The 2,000th anniversary of Jesus’s death on the cross is approaching fast. It’s the kind of attention-getting milestone that will give believers an outstanding opportunity to share the gospel. But when should we observe the anniversary?
Historians agree that Jesus must have been crucified sometime between AD 26 and 36, when Pontius Pilate was stationed in Judea. The exact year, however, has long been a matter of dispute.
Although I lean heavily toward one particular year, my main aim is to lay out the relevant factors so that anyone wishing to tackle this chronological conundrum will be better equipped for the task. The factors don’t all carry the same weight, but each one requires at least some consideration.
First Factor: Tiberius’s Fifteenth Year
We know from Luke 3:1–2 that John the Baptist began his ministry in the 15th year of the emperor Tiberius, who came to power in September AD 14. Luke was writing for the benefit of Theophilus (Luke 1:3), likely a Roman official.
According to Jack Finegan’s Handbook of Biblical Chronology, Roman historians typically started the clock on an emperor’s rule at the beginning of the first full calendar year of the reign. If we apply that principle to Tiberius, his first regnal year was AD 15, and the all-important 15th year was AD 29. This date has the effect of ruling out AD 30 as the crucifixion year because Jesus’s ministry lasted for at least two full years (see the second factor, below).
In his book Jesus in Jerusalem, Eckhard J. Schnabel argues that Luke was counting from Tiberius’s co-regency with the emperor Augustus, beginning in AD 13. If so, the 15th year of Tiberius’s rule would have been AD 27 or 28. The co-regency dating theory is attractive to those, such as Schnabel, who think Jesus was crucified in AD 30.
Arguably, however, the key question isn’t “Which start date for Tiberius’s reign is theoretically possible?”; instead, it’s “Which start date did ancient chroniclers use in practice?” The historian J. K. Fotheringham gives a clear answer: “All our evidence points to one conclusion, that the regnal years of Tiberius throughout the whole empire were reckoned from his succession to full imperial authority, not from his co-regency.”
Similarly, Finegan states that first- and second-century sources “do not count Tiberius’s own reign from what was only his joint rule with Augustus.” He asserts that the co-regency dating theory “may be left out of further consideration”—a verdict also adopted in this article.
Second Factor: Three Passovers in Jesus’s Ministry
We’ve just concluded that John the Baptist’s ministry began in AD 29. Jesus entered public life after John’s work was already underway, and his ministry included at least three Passovers (John 2:13; 6:4; 12:1).
It doesn’t seem plausible for the first of those to have occurred in AD 29 itself, because Passover comes early in the year, and John the Baptist needed sufficient time to get his own ministry up and running. If the first Passover of Jesus’s ministry was in AD 30, it would mean the crucifixion couldn’t have happened until AD 32, which shrinks the range of potential dates to AD 32–36.
Third Factor: Paul’s Personal Timeline
Paul says in Galatians 2:1, “After fourteen years I went up again to Jerusalem.” This most likely means the famine-relief visit recorded in Acts 11:27–30 happened 14 years after his conversion (see the discussion in R. Alan Cole’s commentary on Galatians).
There was a famine in Jerusalem when Tiberius Alexander was procurator of Judea, which was from AD 46–48. If we allow one year for the events before his conversion, Paul’s timeline requires 15 years between the crucifixion and the end of the famine (circa AD 48).
It’s possible to fit 15 years, counted inclusively, between the crucifixion and AD 48 if Jesus died before AD 35. But this chronological factor seems to rule out AD 35 and 36 as potential crucifixion dates. The range of dates thus tightens to AD 32–34.
Fourth Factor: Phlegon’s Olympiads and the Noontime Darkness on Good Friday
Phlegon of Tralles was a non-Christian historian active around AD 150. His book Olympiads is only known through excerpts quoted by other writers. He gives the following account of a darkness that could have been the noontime darkness of the crucifixion (Mark 15:33): “In the fourth year, however, of Olympiad 202, an eclipse of the sun happened, greater and more excellent than any that had happened before it; at the sixth hour [noon], day turned into dark night, so that the stars were seen in the sky.”
Historians agree that Jesus must have been crucified sometime between AD 26 and 36, when Pontius Pilate was stationed in Judea.
The fourth year of Olympiad 202 ran from the summer of AD 32 to the summer of AD 33. Jesus was crucified during the springtime Passover festival, so if Phlegon is referring to the darkness at the time of the crucifixion, his dating would indicate that Jesus died in AD 33.
Phlegon himself doesn’t connect the darkness to Jesus’s crucifixion, but the early Christian writers Julius Africanus, Origen, and Jerome make that connection in their comments on Phlegon’s work. Moreover, Africanus identifies Phlegon’s darkness with the darkness discussed by Thallus, a first-century historian known to us through other writers. Thallus apparently acknowledged there was daytime darkness at the time of Jesus’s death but dismissed it as a coincidental eclipse.
Christian scholar Edwin Yamauchi, interviewed by Lee Strobel, describes Phlegon’s and Thallus’s darkness references as “nonbiblical attestation of the darkness that occurred at the time of Jesus’ crucifixion.” Gary Habermas similarly views the evidence from Phlegon and Thallus as “two extra-biblical mentions of the darkness in ancient writings.”
We’d expect secular historians to ignore the supernatural three-hour darkness beginning at noon on the day of Jesus’s death. But evangelicals believe in the historical reality of that darkness, so we shouldn’t be ashamed to foreground Phlegon’s chronology when discussing the crucifixion’s date.
Fifth Factor: Full Moon at the Last Supper
The four factors above have led us to the year AD 33. The fifth factor offers a way to test that date, but it requires a tough mental workout.
The Last Supper was a Passover meal on the evening before Friday’s crucifixion. That means Nisan 14, the Jewish date when Passover began, must have been a Thursday in the year of Jesus’s death. (It wouldn’t have been called “Thursday,” but I’ll use that name for the fifth day of the week.)
There would have been a full moon that Thursday night, because Jewish months began when the new crescent moon was seen, and the full moon followed 14 days later. So the Last Supper took place at the end of Thursday, Nisan 14, with a full moon rising. But how does that help us test whether AD 33 was the year of Jesus’s death?
Astronomers know precisely when full moons appeared in the first century. In theory, that allows us to check whether the full moon of Nisan 14, AD 33, fell on a Thursday. If it didn’t, we’d arguably need to conclude that Jesus must have been crucified in a different year.
The relevant full moon information is tied to the Western monthly calendar, which wasn’t fixedly aligned with the Jewish monthly calendar at that time. Nisan sometimes matched up mainly with March, and sometimes mainly with April; this floating was caused by the shorter length of the Jewish year and also by the periodic addition of a leap month. However, since Nisan was always a springtime month, a looser version of the “full moon test” can still be applied: Did any of the full moons in the spring of AD 33 fall on a Thursday? Again, if none of them did, we should probably conclude Jesus was crucified in another year.
According to Calendar and Chronology, by the church historian Roger Beckwith, one of the springtime full moons in AD 33 fell on March 5, which was a Thursday. (Importantly, while the Jewish and Western months weren’t aligned, there’s some evidence that the seven-day week was already becoming standardized throughout the Roman Empire.)
Beckwith comments, “March 5th . . . would be rather early for the Passover to fall, but since the climate of Palestine is not mechanically regular . . . an early spring and therefore an early Passover is entirely possible.” If that was the date of the Last Supper, Jesus’s death would have occurred on Friday, March 6, AD 33.
Two Final Factors: Jesus’s Age and the Date of the Temple
Two further chronological markers need to be taken into account.
The sixth factor is Luke’s statement that Jesus “was about thirty years of age” (Luke 3:23) when he began his ministry. The New Testament’s infancy narratives tell us that Jesus was born before King Herod’s death. While some historians think Herod died in 1 BC, which easily fits an AD 30 start to Jesus’s ministry, most think he died in 4 BC. If so, Jesus would have been at least 33 years old in AD 30 (there’s only one year between 1 BC and AD 1).
Why, then, would Luke say he was “about thirty”? The word “about” obviously indicates Luke wasn’t giving a precise age. He may have chosen to give a rounded-down number to show how Jesus’s ministry echoes David’s kingship (see 2 Sam. 5:4).
The seventh factor is the age of Herod’s temple, which was completed in 18/17 BC. Although John 2:20 can be translated in different ways, its most likely meaning is “This temple was built forty-six years ago.” Finegan summarizes the implications: “After 18/17 B.C., forty-six years brings us to A.D. 29/30. The conversation in question between Jesus and the Jews took place at the Passover described in John 2:13–21, and this Passover is therefore to be dated in the spring of the year A.D. 30.” (Schnabel, who holds that Jesus died in AD 30, says that John 2:20 refers to the start of the temple’s construction, rather than the temple’s completion.)
Why Do Some Scholars Point to April 3, AD 33?
In their article “The Date of the Crucifixion,” Oxford University scientists Colin J. Humphreys and W. Graeme Waddington argue for April 3, AD 33, as the date of Jesus’s death. Their conclusion has been influential. Ask ChatGPT, “When exactly did Jesus die?” and it’ll suggest that date; ask it for links, and it’ll bring up articles such as Andreas Köstenberger’s “April 3, AD 33,” which itself links back to Humphreys and Waddington’s research.
Humphreys and Waddington lay out two alternative timetables for the week of Jesus’s death. In the first, Thursday was Nisan 14 (which would mean the Last Supper was a Passover meal held at the regular time). In the second, Friday was Nisan 14 (which would mean Thursday’s Last Supper was a premature Passover meal, and Jesus died before Passover had begun). Humphreys and Waddington describe the latter option as “John’s date,” and they admit that it “disagrees with” a straightforward reading of the Synoptic Gospels.
Next, they use the previously mentioned full moon research to identify dates that could have been Nisan 14 during the years AD 26–36. It should be noted at this point that any specific Western calendar date proposed for the crucifixion depends on this full moon research. There’s no other way to settle on a precise Western date.
In view of their two timetables, Humphreys and Waddington are interested in springtime full moons (i.e., possible Nisan 14s) on Thursdays or Fridays. By their reckoning, the only relevant Thursdays occurred in AD 27, which they rule out as too early, and AD 34, which they rule out for other reasons. (They ignore March 5, AD 33—the Thursday discussed above—because they mistakenly think “Passover must fall after the vernal equinox”; Beckwith shows in Calendar and Chronology that Passover could predate the equinox.)
Turning to their alternative timetable, which they present as “John’s date,” they highlight Friday, April 3, AD 33. This was the next full moon after March 5 in that year. In Humphreys’s and Waddington’s eyes, it fits all the criteria for the crucifixion.
From an evangelical perspective, however, that conclusion is problematic. It only works if Passover followed the crucifixion, which Humphreys and Waddington see as John’s version of events. But according to Matthew 26:17, Mark 14:12, and Luke 22:7, Passover preceded the crucifixion.
Most evangelical scholars uphold the plain meaning of those verses and harmonize John’s account accordingly. The sequence required if Nisan 14 was a Friday—Last Supper, then crucifixion, then Passover—is a stumbling block for those who accept the Synoptic Gospels’ order of events. This has never been adequately addressed by those who favor April 3, AD 33, as the crucifixion date.
In his book Chronological Aspects of the Life of Christ, Harold Hoehner speculates that, due to time-keeping differences, a Galilean such as Jesus might have celebrated Passover one day before the Judeans. At first glance, this results in a sequence that would support April 3, AD 33: Last Supper (a Galilean Passover), then crucifixion, then the Judean Passover.
But Hoehner himself admits “there is no explicit statement” to validate his hypothesis. What’s more, any time-keeping difference between Galilee and Judea would, as Beckwith explains in Calendar and Chronology, “only have led to a difference in the dating of the night and not of the day.” So if Thursday daytime was Nisan 14 for Galileans, it would also have been Nisan 14 for Judeans, and both groups would have celebrated Passover at the end of that day (see Num. 9:2–3).
2033
The Jewish calendar and the Western church calendar run on different tracks, and so Passover and Good Friday are sometimes weeks apart. But in 2033, Passover and Good Friday will fall in the same week, as they did in the year of Jesus’s death.
For that reason, without setting aside the earlier argument in favor of March 6 as the anniversary date, it makes sense for churches to commemorate the 2,000th anniversary of Jesus’s death on the Good Friday of 2033, which falls on April 15.
It Is Finished
I’ve made the case that Jesus died in AD 33. But far more important than the date of Jesus’s death is the fact of his death. It’s been said that the Maker came to be unmade so that we could be remade. By taking the curse for covenant-breaking on himself, God’s Son freed all who trust in him from the punishment we deserve.
The year 2033 will be a special occasion for remembering Jesus’s death, but every day of every year is a perfect day for remembering the sacrifice of the Son of God who loved us and gave himself for us (Gal. 2:20).
The Gospel Coalition