The heaviest verse has been dealt with. Now the heaviest rival. And let me remove at once the temptation to belittle him, because that would be dishonest and it would weaken the case. The real competitor of Wednesday 25 April AD 31 is not the traditional Good Friday of AD 33 — that year turned out to be astronomically shaky (ch. 7). The real competitor is AD 30. And AD 30 is strong. So strong that I shall use this whole chapter to build him up before I weigh him, and that at the end I shall have to concede honestly where he remains standing.
Begin with the most painful fact for this book's case: the astronomy does not rule out AD 30. In the year 30, 14 Nisan fell on a Friday: the conjunction fell almost on the equinox and the crescent just after it, so that under every observational rule Nisan stayed in March and 14 Nisan fell on Friday 7 April (at the latest Thursday 6 April) — never a Wednesday. That "at the latest Thursday" deserves an honest note: under the generous crescent threshold AD 30's 14 Nisan falls on Thursday 6 April instead of Friday 7 April (appendix C.2, q = −0.15, a borderline case), so that AD 30 itself yields a Thursday candidate the moment one admits the Thursday into the field. Only under the later, computed molad calendar (Hillel II) would AD 30 become a Wednesday — but that same reckoning also takes the Wednesday away from AD 31 (→ Monday 26 March). The weekday outcome of the whole field thus rests on the observational premise (ch. 8), and it is that premise which gives the Wednesday to AD 31 and not to AD 30. The sky, then, does not rule out AD 30. More than that: of the two years that yield a Friday 14 Nisan at all, AD 30 is the more robust. Whoever says "the astronomy points to a Wednesday" is mistaken; it points to a two-horse race, and AD 30 is the other horse.
Add to this that AD 30 has the majority behind it. When Josef Blinzler surveyed a hundred specialists in 1969, fifty-three chose AD 30 — with names such as Bruce, Jeremias, Flusser, Fitzmyer and Metzger among them (Blinzler, 1969). That is no small thing, and I shall not slide past it. But it is also precisely the kind of fact this book approaches with the question it asks everywhere: is this a measurement or a majority? A count of heads is not evidence; it is a snapshot of where the discipline stood at that moment, and in this field dates are inherited, not measured afresh each time (ch. 1). The majority makes AD 30 respectable. It does not make it true.
What, then, makes AD 30 substantively strong? One thing, and it is a good thing: the length of the ministry. If the public career of Jesus was shorter — roughly two and a half years instead of three and a half — then the crucifixion slides back from AD 31 to AD 30. And the key to that is a small, unobtrusive verse: John 5:1, where it says that Jesus went up to "a feast of the Jews". Which feast, John does not say. And that is exactly the point of the AD 30 advocate. For if that feast was a Passover, John counts four Passovers in his Gospel, and then the ministry lasts at least three years and the crucifixion falls in AD 31 — a reading this book carries as support and not as compulsion, on a test whose criteria were fixed before the sources were read and of which two of the three criteria for it were not met (ch. 9; appendix E.3). If it was not a Passover, three Passovers remain, a shorter ministry, and AD 30. On this one nameless feast hangs the difference between the two horses.
Expertthe nameless feast of John 5:1
John 5:1 reads, in the oldest and best manuscripts, heortêwithout the article — "a feast", unnamed (so P66, P75, A, B, D, W; the article, "the feast", stands in ℵ, C, L and a few later witnesses). The anarthrous reading is the stronger. It argues against the assumption that this would self-evidently be the great feast — Passover — for John always names his Passovers explicitly elsewhere: 2:13 ("the Passover of the Jews was near"), 6:4 ("the Passover, the feast of the Jews, was near"), 11:55 (likewise). Here he does not. That is the strongest internal argument against a fourth Passover — and therefore for AD 30. But it proves nothing: "unnamed" is not "no Passover"; it is "undecided". Whoever infers from the anarthrous reading that it was certainly no Passover asks of the manuscript apparatus more than it gives.
Here, then, AD 30 stands at its strongest: astronomically unshakeable, in the majority, and with a real internal argument — the nameless feast — for a shorter ministry. This is no straw man. If someone asked me to build the best case for AD 30, I would build exactly this one. And yet it falls. Not with a single blow, but on a conjunction — and it is honest to show precisely where.
Countercasethe full case for AD 30, unabridged
Let the advocate of AD 30 lay everything on the table here. One: the astronomy gives him the robust Friday (7 April 30), stronger than that of AD 33. Two: the majority of the discipline stands behind him (Blinzler, 1969). Three: John 5:1 is unnamed, and John names his Passovers by name elsewhere — so probably three Passovers, a shorter ministry, crucifixion in 30. Four: there is an old Jewish tradition that "forty years before the destruction of the temple" — so around the year 30 — signs in the temple began to fail: the lot for the scapegoat no longer came up favourably, the crimson band no longer turned white, the western lamp went out, the temple doors opened of themselves (b. Yoma 39b), and the Sanhedrin lost the right to judge capital cases (b. Sanhedrin 41a). Forty before 70 is 30. Five: beside Pentecost, Tabernacles and Purim, Passover is only one of the candidates for the nameless feast — the burden of proof lies with whoever makes a fourth Passover of it. All this together is a serious, coherent case, and it deserves a serious answer.
The answer begins with an observation that splits the AD 30 case from within: the two things it leans on — the majority dating of Tiberius and the year 30 — cannot abide each other. John's Gospel gives three explicitly named, mutually separated Passovers (2:13; 6:4; 11:55). Now take the standard reading of Luke 3:1, the "fifteenth year of the emperor Tiberius": on the ordinary, sole-reign count that year falls around AD 28/29, and then the ministry begins at the earliest in the autumn of 28 (Ogg, 1940; Finegan, 1998). Autumn 28 plus three separated Passovers brings the crucifixion at the earliest to the spring of 31 — not 30. Whoever holds the standard Tiberius and chooses AD 30 is left with a chronology that cannot house John's own three Passovers; only two will fit. The AD 30 majority must therefore either fall back on a co-regency count that pulls the baptism to AD 26 — and that count the discipline itself has demolished numismatically, for in no coin, papyrus or inscription is Tiberius's reign counted from an earlier point (worked out in ch. 9) — or it must let go of John's chain of Passovers. The consensus here is not a position but a fault line.
Dossierthe arithmetical pincer
The crucifixion falls at: (baptism year) + (number of Passovers). Two variables, and they are not free to choose. (The baptism falls in the autumn, so that the first Passover of the ministry comes only in the spring following: autumn 27 plus three Passovers gives the third in the spring of 30, not of 29.)
baptism reading
15th year of Tiberius ≈
baptism
+ 3 Passovers
+ 4 Passovers
sole reign (standard)
AD 28/29
autumn 28
AD 31
AD 32
autumn count / non-accession
autumn 27
autumn 27
AD 30
AD 31
co-regency (12/13)
AD 26/27
autumn 26
AD 29
AD 30
Read the table honestly, for it cuts both ways. The co-regency count yields AD 29 with three Passovers and AD 30 only with four — and it is precisely that count which is numismatically untenable (ch. 9). On the autumn count (Syro-Macedonian) — the same count this book uses to arrive at the AD 27 baptism — AD 30 is possible, but only with exactly three Passovers (John 5:1 not a Passover, and no hidden year in the chain); with four Passovers that same count gives AD 31. And on the standard count the chain gives AD 31 and beyond. In other words: every Tiberius reading that lets AD 30 breathe also gives the AD 27 baptism on which this book builds — and on that baptism John's chain pushes past 30 the moment John 5:1 is a Passover. *(The full treatment of the ministry and the forty-six years of John 2:20 stands in ch. 9; here all that counts is that AD 30 and AD 31 rest on the same baptism and part company over a single feast. The four-Passover column is carried as support and not as compulsion — appendix E.3.)* [CALCULATED]
What, then, of AD 30 remains standing? I say it without varnish, for honesty is here the strength of the book. One configuration survives this entire front: baptism in the autumn of 27 (via the autumn count — this book's own route), exactly three Passovers (John 5:1 not a Passover, no hidden year), a ministry of about two and a half years, and the crucifixion on Friday 7 April 30. That configuration is not incoherent. It does not fall on the astronomy, and it does not fall on the arithmetic. That may be said, and it stands.
One measurement makes it weaker than this chapter formerly allowed, and the same measurement weakens this book's own route in the same breath. Steinmann's survey of the first- and second-century evidence — literary, numismatic and epigraphic — finds no surviving witness that numbers Tiberius' regnal years inclusively (Steinmann, 2022a, p. 115). And the Syrian provincial coins, which run on an autumn era and are therefore the nearest parallel a Syro-Macedonian autumn count could ask for, set Tiberius' year 1 at 14/15 but cannot separate inclusive from non-inclusive reckoning, because the news of Augustus' death reached Antioch only after the new year (Roman Provincial Coinage I 4270, Antioch; I 4330, Seleucia; ch. 9). The autumn-inclusive count that the surviving AD 30 configuration needs is attested nowhere: not probable, but not impossible.
And it is the count this book needs too. The baptism in the autumn of 27 rests on that same reckoning, and chapter 9 registers it there in the only form it will bear: unattested, not refuted, not probable but not impossible — no document attests it for Tiberius, and neither coin corpus can test it, because at both new years the accession stub had already elapsed before the news of Augustus' death could arrive. This therefore does not knock AD 30 out. It puts the surviving AD 30 configuration on exactly the footing this book's own baptism stands on: whoever lets the count fall loses AD 30 and the AD 27 baptism together. The finding is entered in the register as favourable, with the second argument of the same article — which excludes AD 30 and AD 31 by name and supports AD 33 (2022a, p. 116) — entered beside it (appendix E.5; ch. 9).
But it carries three burdens, and the third is decisive. The first: it must set aside the earliest tradition. The oldest known reader of John 5:1, Irenaeus around the year 180, reads the nameless feast without hesitation as a Passover — "the second time" that Jesus went up for the Passover — and he does so in passing, in an entirely different argument, not to rescue a chronology (Irenaeus, Adversus Haereses 2.22.3). And the earliest explicit statement on the duration of the ministry, that of Eusebius around 315, calls it "three and a half years, and this the evangelist John makes clear" (Eusebius, Demonstratio Evangelica 8.2). The AD 30 configuration must go against both. (In fairness, and it is said every time: Irenaeus drew from his own count a conclusion no one has followed — that Jesus was approaching fifty and ministered for more than ten years, 2.22.5–6. His count is adopted, not his conclusion.) The second burden: it must make Luke's "about thirty years" (Luke 3:23) and the forty-six years of the temple's building (John 2:20; the temple begun around 20/19 BC, Josephus, Antiquitates 15.380) sit together with a ministry shorter than tradition gives it.
And then the third burden, which makes all the others superfluous: AD 30 is a Friday. Even if the whole configuration holds — baptism 27, three Passovers, two and a half years — it still delivers a Friday crucifixion. And a Friday cannot carry the three nights of the sign of Jonah (ch. 3), and it cannot house the two sabbaths of the passion week (ch. 4). The day-axis, which was measured in Part II, pointed to the middle of the week; and AD 30 cannot be a Wednesday. AD 30 does not fall on the year, then. It falls on the weekday. The year-axis is truly a dead heat between two horses; it is the day-axis that cuts the knot — and it cuts it against AD 30, because AD 30 is a Friday and the text asks for a Wednesday.
And here this book's own measure applies, in both directions. To let AD 30 win, one must — after everything — return to the Friday, and the Friday needs that same shortened "three days and three nights" idiom that Chapter 3 found attested nowhere. AD 30 loses at precisely the point where the whole Friday reading loses: not on a measurement that excludes it, but on the conjunction of a day-axis it cannot reach with an idiom that no one has ever demonstrated.
Excursusthe small school, and the honest bookkeeping
Two things belong here, and both cost something. The first costs the opposing side its size: the so-called "short chronology" is smaller than its reputation. George Ogg, often filed with the AD 30 school, himself chose AD 33 in the end (Ogg, 1940); Jack Finegan, in his revised handbook of 1998, swung round to AD 33 (Finegan, 1998); Harold Hoehner defended AD 33 (Hoehner, 1974). The list of names behind AD 30 is thus in part a list of people who, pressed further, ended up somewhere else. The second costs this book something, and precisely for that reason it stands here: the book's own AD 27 baptism formerly rested in part on a co-regency count, and that count is — as said (ch. 9) — numismatically untenable. It has been dropped; this book's AD 27 baptism now rests solely on the autumn count. A measure laid against AD 30 (no untenable auxiliary route) applies in the same breath to this book itself.
Excursusthe map of the schools, and where AD 30 still has a defender
The literal reading of the sign is no discovery of this book, and a map of the positions shows it. Whoever takes "three days and three nights" as a measure comes out in the middle of the week; others did so before me, with other calendars and other years. Fred Coulter reads the sign literally and lands on a Wednesday in AD 30, but he reckons with the back-projected calculated calendar ("all the feasts [...] were calculated by the same method that is in use today", Harmony, p. 50), the molad reckoning that ch. 8 rejects as the late model and that does give AD 30 a Wednesday (14 Nisan = 5 April). Floyd Jones reads the sign just as literally, rejects with this book the 360-day scheme and the Friday, and lands on a Thursday in AD 30, by way of the decree of Nehemiah, which he places in 454 BC (Chronology of the Old Testament, p. 235). Beside them stand two Friday schools that fix the year by way of Daniel: traditional Adventism (DARCOM, Shea) on Friday 31, with the weekday via a shifted month-beginning, and Hoehner with the majority on Friday 33, by way of 444 BC and the years of 360 days (ch. 17).
defender
weekday
year
ground
this book
Wednesday
31
sign as measure; observational calendar; 457 as convergence (ch. 17)
Coulter, Harmony of the Gospels
Wednesday
30
sign as measure; calculated (molad) calendar
Jones, Chronology of the Old Testament
Thursday
30
sign as measure; Nehemiah decree in 454 BC
DARCOM / Shea; Maxwell, de Oliveira (ch. 15)
Friday
31
457/Daniel; weekday via a shifted month-beginning
Hoehner et al. (majority)
Friday
33
444 BC and years of 360 days; sign as figure of speech
Two things follow. The difference within the Wednesday camp, 30 against 31, is exactly the difference between computing and looking: Coulter's molad gives the Wednesday in 30, the visible crescent gives it in 31 (ch. 8). And Jones' Thursday does not lie outside the observational calendar: under the generous crescent threshold 14 Nisan of AD 30 falls on Thursday 6 April (appendix C.2, q = −0.15), so that AD 30 has a defender within the literal school too. That defender falls not on the weekday but on the year and on the threshold: his 454 BC requires an accession of Artaxerxes in 474, against Ptolemy's canon and the cuneiform (ch. 17), and his Thursday hangs on the borderline evening of 23 March that Schaefer himself calls "not certain" (appendix C.2). The weekday is thus carried more broadly than by one book; the year hangs on the calendar choice, and that is the choice ch. 8 accounts for. [DOCUMENTED — Coulter, p. 50, and Jones, pp. 232 and 235, by page reference; CALCULATED — the molad Wednesday of AD 30 and the threshold evening of 23 March AD 30]
Where does the investigation stand, then, at the end of this chapter? With a rival who has not been written off but honestly located. AD 30 is astronomically as strong as AD 31; it is the majority choice; and there is one internal configuration — baptism 27, three Passovers — that survives this entire front, on a count attested for Tiberius nowhere and shared with the baptism year this book assumes. But that configuration founders on the weekday: AD 30 is a Friday, and the day the text asks for is a Wednesday. The choice between the two horses is not decided by the sky, but by the axis that was measured first.
There remains the reigning date itself — AD 33, with its famous Good Friday. If AD 30 and this book's own case have been laid against the strictest measure, then the consensus deserves the same treatment. That is the next question.
HeartHEART
There is something liberating about a case that need not fear its strongest opponent. I have built AD 30 up here as well as I could — the robust Friday, the majority, the nameless feast — and the case is not weakened by it; it is made more honest by it. Whoever seeks the truth and not the win can afford to draw his opponent at his best. That is not merely good method; it is a form of reverence — for the text, for the reader, and for the many scholars who in good conscience chose another year. The date that remains does not remain because the others were silenced, but because they were heard at their strongest.
Note on the sources
The astronomical robustness of AD 30 (Friday 7 April, under every observational rule — at the latest Thursday 6 April under the generous crescent threshold; under the computed molad a Wednesday instead, and then AD 31 loses its Wednesday too) is recomputable and is worked out in ch. 7 and appendix A. The arithmetical pincer (baptism year × number of Passovers) is a counting-logical inference and marked as such. The early witnesses (Irenaeus, Eusebius) have been checked in translation in the source itself, with location; the Greek remains a control point still to be closed. The four-Passover column of the pincer is carried as support and not as compulsion, on the test fixed in advance and accounted for in appendix E.3; earlier versions of this chapter carried it more strongly than that test allows. That no surviving witness counts Tiberius' regnal years inclusively rests on Steinmann (2022a, p. 115), seen at first hand; the Syrian provincial coins RPC I 4270 and I 4330 fix the era but cannot separate the counts (ch. 9). The Tiberius numismatics and the majority survey were consulted via the standard works named. Where this chapter says of AD 30 that it "survives this front", that is an honest concession, not a concession of the final conclusion: AD 30 falls on the day-axis, not on the year-axis, and there is a proof for neither side.
Literature (APA)
Blinzler, J. (1969). Der Prozess Jesu (4th ed.). Friedrich Pustet.
Coulter, F. R. (1974). A harmony of the Gospels in modern English (p. 50). York Publishing. [Page reference taken from the older dossier; edition still to be verified.]
Eusebius of Caesarea. (1920). The proof of the Gospel (Demonstratio Evangelica) (W. J. Ferrar, Trans.; Book 8, ch. 2). SPCK. (Original work ca. 315)
Finegan, J. (1998). Handbook of biblical chronology (rev. ed.). Hendrickson.
Hoehner, H. W. (1974). Chronological aspects of the life of Christ, Part IV: The day of Christ's crucifixion. Bibliotheca Sacra, 131(523), 241–264.
Irenaeus of Lyons. (1885). Against heresies (A. Roberts & J. Donaldson, Eds.; Ante-Nicene Fathers, Vol. 1; book 2, ch. 22, §3–6). Christian Literature Company. (Original work ca. 180)
Jones, F. N. (2005). The chronology of the Old Testament (pp. 232, 235). Master Books. [Co-regency and baptism 26 on p. 232; the Thursday of AD 30 on p. 235.]
Ogg, G. (1940). The chronology of the public ministry of Jesus. Cambridge University Press.
Steinmann, A. E. (2022a). Reckoning Tiberius's reign and Jesus's baptism: First- and second-century evidence concerning Tiberius's fifteenth year (Luke 3:1). Tyndale Bulletin, 73, 91–118. https://doi.org/10.53751/001c.37789 ("No evidence for inclusive reckoning of Tiberius's reign" at p. 115; the conclusion excluding AD 30 and AD 31 by name, and supporting AD 33, at p. 116. Seen at first hand.)
Primary texts: Luke 3:1, 23 · John 2:13, 2:20; 5:1; 6:4; 11:55 · Josephus, Antiquitates Judaicae 15.380 · Babylonian Talmud, Yoma 39b; Sanhedrin 41a · Roman Provincial Coinage I 4270 (Antioch) and I 4330 (Seleucia) · Novum Testamentum Graece (NA28), apparatus ad John 5:1. This chapter quotes no Greek of its own; where it refers to the apparatus of NA28, that is expressly meant as the apparatus of the critical edition and not as the text-form this book maintains — that is the received text (chs. 3, 4, 13). My own working translation.
The heaviest verse has been dealt with. Now the heaviest rival. And let me remove at once the temptation to belittle him, because that would be dishonest and it would weaken the case. The real competitor of Wednesday 25 April AD 31 is not the traditional Good Friday of AD 33 — that year turned out to be astronomically shaky (ch. 7). The real competitor is AD 30. And AD 30 is strong. So strong that I shall use this whole chapter to build him up before I weigh him, and that at the end I shall have to concede honestly where he remains standing.
Begin with the most painful fact for this book's case: the astronomy does not rule out AD 30. In the year 30, 14 Nisan fell on a Friday: the conjunction fell almost on the equinox and the crescent just after it, so that under every observational rule Nisan stayed in March and 14 Nisan fell on Friday 7 April (at the latest Thursday 6 April) — never a Wednesday. That "at the latest Thursday" deserves an honest note: under the generous crescent threshold AD 30's 14 Nisan falls on Thursday 6 April instead of Friday 7 April (appendix C.2, q = −0.15, a borderline case), so that AD 30 itself yields a Thursday candidate the moment one admits the Thursday into the field. Only under the later, computed molad calendar (Hillel II) would AD 30 become a Wednesday — but that same reckoning also takes the Wednesday away from AD 31 (→ Monday 26 March). The weekday outcome of the whole field thus rests on the observational premise (ch. 8), and it is that premise which gives the Wednesday to AD 31 and not to AD 30. The sky, then, does not rule out AD 30. More than that: of the two years that yield a Friday 14 Nisan at all, AD 30 is the more robust. Whoever says "the astronomy points to a Wednesday" is mistaken; it points to a two-horse race, and AD 30 is the other horse.
Add to this that AD 30 has the majority behind it. When Josef Blinzler surveyed a hundred specialists in 1969, fifty-three chose AD 30 — with names such as Bruce, Jeremias, Flusser, Fitzmyer and Metzger among them (Blinzler, 1969). That is no small thing, and I shall not slide past it. But it is also precisely the kind of fact this book approaches with the question it asks everywhere: is this a measurement or a majority? A count of heads is not evidence; it is a snapshot of where the discipline stood at that moment, and in this field dates are inherited, not measured afresh each time (ch. 1). The majority makes AD 30 respectable. It does not make it true.
What, then, makes AD 30 substantively strong? One thing, and it is a good thing: the length of the ministry. If the public career of Jesus was shorter — roughly two and a half years instead of three and a half — then the crucifixion slides back from AD 31 to AD 30. And the key to that is a small, unobtrusive verse: John 5:1, where it says that Jesus went up to "a feast of the Jews". Which feast, John does not say. And that is exactly the point of the AD 30 advocate. For if that feast was a Passover, John counts four Passovers in his Gospel, and then the ministry lasts at least three years and the crucifixion falls in AD 31 — a reading this book carries as support and not as compulsion, on a test whose criteria were fixed before the sources were read and of which two of the three criteria for it were not met (ch. 9; appendix E.3). If it was not a Passover, three Passovers remain, a shorter ministry, and AD 30. On this one nameless feast hangs the difference between the two horses.
Expertthe nameless feast of John 5:1
John 5:1 reads, in the oldest and best manuscripts, heortê without the article — "a feast", unnamed (so P66, P75, A, B, D, W; the article, "the feast", stands in ℵ, C, L and a few later witnesses). The anarthrous reading is the stronger. It argues against the assumption that this would self-evidently be the great feast — Passover — for John always names his Passovers explicitly elsewhere: 2:13 ("the Passover of the Jews was near"), 6:4 ("the Passover, the feast of the Jews, was near"), 11:55 (likewise). Here he does not. That is the strongest internal argument against a fourth Passover — and therefore for AD 30. But it proves nothing: "unnamed" is not "no Passover"; it is "undecided". Whoever infers from the anarthrous reading that it was certainly no Passover asks of the manuscript apparatus more than it gives.
Here, then, AD 30 stands at its strongest: astronomically unshakeable, in the majority, and with a real internal argument — the nameless feast — for a shorter ministry. This is no straw man. If someone asked me to build the best case for AD 30, I would build exactly this one. And yet it falls. Not with a single blow, but on a conjunction — and it is honest to show precisely where.
Countercasethe full case for AD 30, unabridged
Let the advocate of AD 30 lay everything on the table here. One: the astronomy gives him the robust Friday (7 April 30), stronger than that of AD 33. Two: the majority of the discipline stands behind him (Blinzler, 1969). Three: John 5:1 is unnamed, and John names his Passovers by name elsewhere — so probably three Passovers, a shorter ministry, crucifixion in 30. Four: there is an old Jewish tradition that "forty years before the destruction of the temple" — so around the year 30 — signs in the temple began to fail: the lot for the scapegoat no longer came up favourably, the crimson band no longer turned white, the western lamp went out, the temple doors opened of themselves (b. Yoma 39b), and the Sanhedrin lost the right to judge capital cases (b. Sanhedrin 41a). Forty before 70 is 30. Five: beside Pentecost, Tabernacles and Purim, Passover is only one of the candidates for the nameless feast — the burden of proof lies with whoever makes a fourth Passover of it. All this together is a serious, coherent case, and it deserves a serious answer.
The answer begins with an observation that splits the AD 30 case from within: the two things it leans on — the majority dating of Tiberius and the year 30 — cannot abide each other. John's Gospel gives three explicitly named, mutually separated Passovers (2:13; 6:4; 11:55). Now take the standard reading of Luke 3:1, the "fifteenth year of the emperor Tiberius": on the ordinary, sole-reign count that year falls around AD 28/29, and then the ministry begins at the earliest in the autumn of 28 (Ogg, 1940; Finegan, 1998). Autumn 28 plus three separated Passovers brings the crucifixion at the earliest to the spring of 31 — not 30. Whoever holds the standard Tiberius and chooses AD 30 is left with a chronology that cannot house John's own three Passovers; only two will fit. The AD 30 majority must therefore either fall back on a co-regency count that pulls the baptism to AD 26 — and that count the discipline itself has demolished numismatically, for in no coin, papyrus or inscription is Tiberius's reign counted from an earlier point (worked out in ch. 9) — or it must let go of John's chain of Passovers. The consensus here is not a position but a fault line.
Dossierthe arithmetical pincer
The crucifixion falls at: (baptism year) + (number of Passovers). Two variables, and they are not free to choose. (The baptism falls in the autumn, so that the first Passover of the ministry comes only in the spring following: autumn 27 plus three Passovers gives the third in the spring of 30, not of 29.)
Read the table honestly, for it cuts both ways. The co-regency count yields AD 29 with three Passovers and AD 30 only with four — and it is precisely that count which is numismatically untenable (ch. 9). On the autumn count (Syro-Macedonian) — the same count this book uses to arrive at the AD 27 baptism — AD 30 is possible, but only with exactly three Passovers (John 5:1 not a Passover, and no hidden year in the chain); with four Passovers that same count gives AD 31. And on the standard count the chain gives AD 31 and beyond. In other words: every Tiberius reading that lets AD 30 breathe also gives the AD 27 baptism on which this book builds — and on that baptism John's chain pushes past 30 the moment John 5:1 is a Passover. *(The full treatment of the ministry and the forty-six years of John 2:20 stands in ch. 9; here all that counts is that AD 30 and AD 31 rest on the same baptism and part company over a single feast. The four-Passover column is carried as support and not as compulsion — appendix E.3.)* [CALCULATED]
What, then, of AD 30 remains standing? I say it without varnish, for honesty is here the strength of the book. One configuration survives this entire front: baptism in the autumn of 27 (via the autumn count — this book's own route), exactly three Passovers (John 5:1 not a Passover, no hidden year), a ministry of about two and a half years, and the crucifixion on Friday 7 April 30. That configuration is not incoherent. It does not fall on the astronomy, and it does not fall on the arithmetic. That may be said, and it stands.
One measurement makes it weaker than this chapter formerly allowed, and the same measurement weakens this book's own route in the same breath. Steinmann's survey of the first- and second-century evidence — literary, numismatic and epigraphic — finds no surviving witness that numbers Tiberius' regnal years inclusively (Steinmann, 2022a, p. 115). And the Syrian provincial coins, which run on an autumn era and are therefore the nearest parallel a Syro-Macedonian autumn count could ask for, set Tiberius' year 1 at 14/15 but cannot separate inclusive from non-inclusive reckoning, because the news of Augustus' death reached Antioch only after the new year (Roman Provincial Coinage I 4270, Antioch; I 4330, Seleucia; ch. 9). The autumn-inclusive count that the surviving AD 30 configuration needs is attested nowhere: not probable, but not impossible.
And it is the count this book needs too. The baptism in the autumn of 27 rests on that same reckoning, and chapter 9 registers it there in the only form it will bear: unattested, not refuted, not probable but not impossible — no document attests it for Tiberius, and neither coin corpus can test it, because at both new years the accession stub had already elapsed before the news of Augustus' death could arrive. This therefore does not knock AD 30 out. It puts the surviving AD 30 configuration on exactly the footing this book's own baptism stands on: whoever lets the count fall loses AD 30 and the AD 27 baptism together. The finding is entered in the register as favourable, with the second argument of the same article — which excludes AD 30 and AD 31 by name and supports AD 33 (2022a, p. 116) — entered beside it (appendix E.5; ch. 9).
But it carries three burdens, and the third is decisive. The first: it must set aside the earliest tradition. The oldest known reader of John 5:1, Irenaeus around the year 180, reads the nameless feast without hesitation as a Passover — "the second time" that Jesus went up for the Passover — and he does so in passing, in an entirely different argument, not to rescue a chronology (Irenaeus, Adversus Haereses 2.22.3). And the earliest explicit statement on the duration of the ministry, that of Eusebius around 315, calls it "three and a half years, and this the evangelist John makes clear" (Eusebius, Demonstratio Evangelica 8.2). The AD 30 configuration must go against both. (In fairness, and it is said every time: Irenaeus drew from his own count a conclusion no one has followed — that Jesus was approaching fifty and ministered for more than ten years, 2.22.5–6. His count is adopted, not his conclusion.) The second burden: it must make Luke's "about thirty years" (Luke 3:23) and the forty-six years of the temple's building (John 2:20; the temple begun around 20/19 BC, Josephus, Antiquitates 15.380) sit together with a ministry shorter than tradition gives it.
And then the third burden, which makes all the others superfluous: AD 30 is a Friday. Even if the whole configuration holds — baptism 27, three Passovers, two and a half years — it still delivers a Friday crucifixion. And a Friday cannot carry the three nights of the sign of Jonah (ch. 3), and it cannot house the two sabbaths of the passion week (ch. 4). The day-axis, which was measured in Part II, pointed to the middle of the week; and AD 30 cannot be a Wednesday. AD 30 does not fall on the year, then. It falls on the weekday. The year-axis is truly a dead heat between two horses; it is the day-axis that cuts the knot — and it cuts it against AD 30, because AD 30 is a Friday and the text asks for a Wednesday.
And here this book's own measure applies, in both directions. To let AD 30 win, one must — after everything — return to the Friday, and the Friday needs that same shortened "three days and three nights" idiom that Chapter 3 found attested nowhere. AD 30 loses at precisely the point where the whole Friday reading loses: not on a measurement that excludes it, but on the conjunction of a day-axis it cannot reach with an idiom that no one has ever demonstrated.
Excursusthe small school, and the honest bookkeeping
Two things belong here, and both cost something. The first costs the opposing side its size: the so-called "short chronology" is smaller than its reputation. George Ogg, often filed with the AD 30 school, himself chose AD 33 in the end (Ogg, 1940); Jack Finegan, in his revised handbook of 1998, swung round to AD 33 (Finegan, 1998); Harold Hoehner defended AD 33 (Hoehner, 1974). The list of names behind AD 30 is thus in part a list of people who, pressed further, ended up somewhere else. The second costs this book something, and precisely for that reason it stands here: the book's own AD 27 baptism formerly rested in part on a co-regency count, and that count is — as said (ch. 9) — numismatically untenable. It has been dropped; this book's AD 27 baptism now rests solely on the autumn count. A measure laid against AD 30 (no untenable auxiliary route) applies in the same breath to this book itself.
Excursusthe map of the schools, and where AD 30 still has a defender
The literal reading of the sign is no discovery of this book, and a map of the positions shows it. Whoever takes "three days and three nights" as a measure comes out in the middle of the week; others did so before me, with other calendars and other years. Fred Coulter reads the sign literally and lands on a Wednesday in AD 30, but he reckons with the back-projected calculated calendar ("all the feasts [...] were calculated by the same method that is in use today", Harmony, p. 50), the molad reckoning that ch. 8 rejects as the late model and that does give AD 30 a Wednesday (14 Nisan = 5 April). Floyd Jones reads the sign just as literally, rejects with this book the 360-day scheme and the Friday, and lands on a Thursday in AD 30, by way of the decree of Nehemiah, which he places in 454 BC (Chronology of the Old Testament, p. 235). Beside them stand two Friday schools that fix the year by way of Daniel: traditional Adventism (DARCOM, Shea) on Friday 31, with the weekday via a shifted month-beginning, and Hoehner with the majority on Friday 33, by way of 444 BC and the years of 360 days (ch. 17).
Two things follow. The difference within the Wednesday camp, 30 against 31, is exactly the difference between computing and looking: Coulter's molad gives the Wednesday in 30, the visible crescent gives it in 31 (ch. 8). And Jones' Thursday does not lie outside the observational calendar: under the generous crescent threshold 14 Nisan of AD 30 falls on Thursday 6 April (appendix C.2, q = −0.15), so that AD 30 has a defender within the literal school too. That defender falls not on the weekday but on the year and on the threshold: his 454 BC requires an accession of Artaxerxes in 474, against Ptolemy's canon and the cuneiform (ch. 17), and his Thursday hangs on the borderline evening of 23 March that Schaefer himself calls "not certain" (appendix C.2). The weekday is thus carried more broadly than by one book; the year hangs on the calendar choice, and that is the choice ch. 8 accounts for. [DOCUMENTED — Coulter, p. 50, and Jones, pp. 232 and 235, by page reference; CALCULATED — the molad Wednesday of AD 30 and the threshold evening of 23 March AD 30]
Where does the investigation stand, then, at the end of this chapter? With a rival who has not been written off but honestly located. AD 30 is astronomically as strong as AD 31; it is the majority choice; and there is one internal configuration — baptism 27, three Passovers — that survives this entire front, on a count attested for Tiberius nowhere and shared with the baptism year this book assumes. But that configuration founders on the weekday: AD 30 is a Friday, and the day the text asks for is a Wednesday. The choice between the two horses is not decided by the sky, but by the axis that was measured first.
There remains the reigning date itself — AD 33, with its famous Good Friday. If AD 30 and this book's own case have been laid against the strictest measure, then the consensus deserves the same treatment. That is the next question.
HeartHEART
There is something liberating about a case that need not fear its strongest opponent. I have built AD 30 up here as well as I could — the robust Friday, the majority, the nameless feast — and the case is not weakened by it; it is made more honest by it. Whoever seeks the truth and not the win can afford to draw his opponent at his best. That is not merely good method; it is a form of reverence — for the text, for the reader, and for the many scholars who in good conscience chose another year. The date that remains does not remain because the others were silenced, but because they were heard at their strongest.
Note on the sources
The astronomical robustness of AD 30 (Friday 7 April, under every observational rule — at the latest Thursday 6 April under the generous crescent threshold; under the computed molad a Wednesday instead, and then AD 31 loses its Wednesday too) is recomputable and is worked out in ch. 7 and appendix A. The arithmetical pincer (baptism year × number of Passovers) is a counting-logical inference and marked as such. The early witnesses (Irenaeus, Eusebius) have been checked in translation in the source itself, with location; the Greek remains a control point still to be closed. The four-Passover column of the pincer is carried as support and not as compulsion, on the test fixed in advance and accounted for in appendix E.3; earlier versions of this chapter carried it more strongly than that test allows. That no surviving witness counts Tiberius' regnal years inclusively rests on Steinmann (2022a, p. 115), seen at first hand; the Syrian provincial coins RPC I 4270 and I 4330 fix the era but cannot separate the counts (ch. 9). The Tiberius numismatics and the majority survey were consulted via the standard works named. Where this chapter says of AD 30 that it "survives this front", that is an honest concession, not a concession of the final conclusion: AD 30 falls on the day-axis, not on the year-axis, and there is a proof for neither side.
Literature (APA)
Blinzler, J. (1969). Der Prozess Jesu (4th ed.). Friedrich Pustet.
Coulter, F. R. (1974). A harmony of the Gospels in modern English (p. 50). York Publishing. [Page reference taken from the older dossier; edition still to be verified.]
Eusebius of Caesarea. (1920). The proof of the Gospel (Demonstratio Evangelica) (W. J. Ferrar, Trans.; Book 8, ch. 2). SPCK. (Original work ca. 315)
Finegan, J. (1998). Handbook of biblical chronology (rev. ed.). Hendrickson.
Hoehner, H. W. (1974). Chronological aspects of the life of Christ, Part IV: The day of Christ's crucifixion. Bibliotheca Sacra, 131(523), 241–264.
Irenaeus of Lyons. (1885). Against heresies (A. Roberts & J. Donaldson, Eds.; Ante-Nicene Fathers, Vol. 1; book 2, ch. 22, §3–6). Christian Literature Company. (Original work ca. 180)
Jones, F. N. (2005). The chronology of the Old Testament (pp. 232, 235). Master Books. [Co-regency and baptism 26 on p. 232; the Thursday of AD 30 on p. 235.]
Ogg, G. (1940). The chronology of the public ministry of Jesus. Cambridge University Press.
Steinmann, A. E. (2022a). Reckoning Tiberius's reign and Jesus's baptism: First- and second-century evidence concerning Tiberius's fifteenth year (Luke 3:1). Tyndale Bulletin, 73, 91–118. https://doi.org/10.53751/001c.37789 ("No evidence for inclusive reckoning of Tiberius's reign" at p. 115; the conclusion excluding AD 30 and AD 31 by name, and supporting AD 33, at p. 116. Seen at first hand.)
Primary texts: Luke 3:1, 23 · John 2:13, 2:20; 5:1; 6:4; 11:55 · Josephus, Antiquitates Judaicae 15.380 · Babylonian Talmud, Yoma 39b; Sanhedrin 41a · Roman Provincial Coinage I 4270 (Antioch) and I 4330 (Seleucia) · Novum Testamentum Graece (NA28), apparatus ad John 5:1. This chapter quotes no Greek of its own; where it refers to the apparatus of NA28, that is expressly meant as the apparatus of the critical edition and not as the text-form this book maintains — that is the received text (chs. 3, 4, 13). My own working translation.