One tradition places Jesus' birth on 6 January, another on the twenty-fifth of a month. Their basis is His true birth date, which was 6 January (25 Kislev on the Jewish calendar) in 5 BC. It was the first day of the feast known as Hanukkah, or the Feast of Lights. The coincidence was intended by God to underscore who Jesus was. He was the light coming into the world (John 1:1-5; 9:4-5).
The date we are presenting for Christ's birth is consistent with all relevant historical and circumstantial evidence. The same evidence is mishandled by all attempts to set His birth earlier than 5 BC or later than 4 BC. For a survey of all relevant questions, see a companion article. This article, accepting Josephus's testimony that a lunar eclipse fell shortly before Herod's death, will focus on discovering exactly which eclipse Josephus was remembering.
The Lunar Eclipse before Herod's Death
Josephus places Herod's death between a lunar eclipse and the Feast of Passover (1). Many have assumed that Josephus is referring to the partial eclipse on 13 March 4 BC (2), about a month before the beginning of Passover on 11 April (3). W. E. Filmer and his followers believe, however, that they can better accommodate the available evidence by supposing that Josephus intends the spectacular total eclipse on 10 January 1 BC (4), about three months before the beginning of Passover on 8 April (5).
They rest their case primarily on two arguments: (1) Josephus suggests that the eclipse was generally remembered as an ominous sign, yet it is doubtful that a mere partial eclipse would have left such an impression on the public mind; (2) when treating the interval between the eclipse and the Passover that bracketed Herod's death, Josephus recounts a complex chain of events that seemingly could not have happened within the scope of a single month.
Although these events make a tight fit in 4 BC, they do not quite burst the seams. Douglas Johnson has demonstrated that Ernest L. Martin, for one, has greatly exaggerated the compression (6). Martin imagines that Herod's funeral procession took 25 days to carry his body from Jericho to Herodeion, a distance of 23 miles (8). Yet, as Johnson shows, the body must have been transported to its burial place within a single day (9). Martin admits that apart from Herod's funeral, the remaining events could have occurred within 33 days (10). If 33, why not 30? It is impossible from our perspective to set dogmatic lower limits. We conclude that Josephus's account of these events does not forbid placement of Herod's death after the lunar eclipse in 4 BC.
Yet to keep Herod’s death no later than 4 BC, we need not squeeze Josephus's narrative into a narrow time frame. Timothy Barnes suggested that the lunar eclipse intended by Josephus is not the partial eclipse on 13 March 4 BC but the spectacular total eclipse on 15 September 5 BC (11). The time between the earlier eclipse and the following Passover is more than ample to accommodate all the events that Josephus assigns to it. Moreover, it was an eclipse that observers were more likely to consider noteworthy. Also, instead of constraining us to set Herod's death after 13 March 4 BC, it allows a date anytime after 1 January 4 BC.
Against Barnes’s placement of Josephus’s eclipse in 5 BC, Martin has raised three objections, all depending on subtle conjectural distortion of the account we find in Josephus—distortion similar to what Johnson also found in Martin’s picture of events when we set the eclipse in 4 BC. These are the objections.
A death date in 4 BC conflicts with what Josephus tells us about Varus, one of the Roman governors of Syria (12). Varus assisted as a judge when Herod brought his son Antipater to trial for plotting his father’s assassination. According to Martin, Josephus describes Varus as having recently come to Syria (13). The trial was held immediately upon Antipater's return (14) from a seven-month stay in Rome (15). Again according to Martin, the governor of Syria while Antipater resided in Rome was still Saturninus (16). It is known from surviving coins that Varus was the Syrian governor in years 25, 26, and 27 of the Actian era; that is, from 7/6 BC to 5/4 BC (17). How could Varus be described as a new governor at the time of Antipater’s trial, held only a few weeks before Herod’s death, if he assumed office in 7/6 BC and Herod’s death was in late 5 BC or early 4 BC?
If we look carefully at this reasoning, we find two flaws. Martin says, "Josephus implies strongly that Varus had just recently replaced Saturninus as the new governor" (18). He implies no such thing. When he introduces Varus to the narrative, he says only, "Now Quintillius Varus was at this time in Jerusalem, being sent to succeed Saturninus as president of Syria" (19). Josephus is merely explaining why the governor of Syria was no longer the person he mentioned previously. Far from suggesting that Varus had just come to Syria, Josephus, when speaking of Varus's departure from Herod, says he "departed out of the court, and went away the day following to Antioch, where his usual residence was" (20).
Martin says that Saturninus was still governor during Antipater's recent stay in Rome. Again, a fair reading of Josephus does not support this inference.
The account in Wars is too vague for sound conclusions (21). There, after informing us that Antipater sailed to Rome, Josephus adds that another who went was Sylleus, a bitter enemy of Herod. He continues by telling us how Sylleus engineered a plot against Herod's life. He engaged Corinthus, one of Herod's guards, to carry out the attack, and afterward, with the promise of money, recruited the support of Fabatus, Herod's steward. Fabatus, in return for the promise of more money, informed Herod of the threat, who then arrested Corinthus and two Arabians also allied with Sylleus. All three confessed under torture, and when Saturninus was informed by Herod, he sent the three to Rome for trial. This appearance of Saturninus in the narrative is the sole basis for Martin's claim that Saturninus was still governor when Antipater was in Rome.
In Antiquities, Josephus clarifies the order of events (22). He places Saturninus's verdict last in sequence, after Antipater's voyage, only because he is filling in the background of events in Rome. He furnishes two details leaving no doubt that the verdict actually came before the voyage. He says that Sylleus traveled with Antipater; also, that after coming before Caesar, Antipater and other enemies of Sylleus charged him with the murder of Fabatus. It follows that Fabatus's death must have preceded the voyage. Since he was assassinated in revenge for betraying the conspirators who were caught by Herod and sent to Rome by Saturninus, Saturninus's role in the drama must have preceded Fabatus's death. It would not be unreasonable to set as much as several months between His verdict and Antipater's departure for Rome.
Clearly then, nothing we know about Saturninus prohibits our reconstruction of events. As we have seen from coinage, Saturninus's tenure as governor of Syria probably lapsed in the Syrian year running from the fall of 7 BC to the fall of 6 BC. In a timetable we will present later, we place Antipater's departure for Rome in the fall of 6 BC. It is evident that he sailed not long after Saturninus left office.
On the night of the eclipse, Herod took the high priesthood from Matthias (23), who had received it shortly after Antipater’s crimes against his father were first discovered (24). It was seven months from this discovery until Antipater’s return to Judea (25). On the next day after his return Antipater was tried and convicted (26). Soon afterward Herod’s ambassadors “made haste to Rome” seeking Caesar’s permission for Antipater’s execution (27). It was while Herod awaited an answer that the eclipse occurred (28). Caesar’s answer came back only a few days before Herod’s death (29). Martin has estimated that a round trip by official couriers on an urgent mission could not have taken more than six weeks to two months (30).
These presumed facts create two impossibilities.
- Matthias could not have held the priesthood more than eight or nine months. Yet Josephus (31) and the Babylonian Talmud (32) testify that his tenure spanned the Day of Atonement. The Day of Atonement preceding 15 September 5 BC fell on 24 October 6 BC (33), about eleven months earlier.
- The events that Josephus places between the eclipse and Herod’s death could not have stretched out to at least four months (from 15 September 5 BC to early 5 BC).
Although Martin’s reconstruction of events is compelling on the surface, inside we again find a framework of unwarranted compression. In three respects he has collapsed a larger story to smaller dimensions.
- Josephus by no means tells us that it was seven months from discovery of Antipater's crime until his return to Judea. Rather, he says that Antipater had been in Rome for seven months when he wrote a letter to his father informing him of his plan to come home (34). How soon afterward he actually embarked is uncertain, and the time required for the journey is uncertain as well, but Herod evidently expected it to be a leisurely affair, stopping at ports of call along the way, for in his letter of reply, he urged Antipater to "not delay" (35). Therefore, the time from installation of Matthias until Antipater's trial was more than seven months, at least eight months.
- At the conclusion of the trial, "when Herod had bound his son, he sent letters to Rome to Caesar about him, and such messengers withal as should, by word of mouth, inform Caesar of Antipater's wickedness" (36). Martin treats this wave of accusations as reason enough to expect Caesar's immediate consent to Antipater's execution. But Herod was not content with the case already forged against his son, as we see in his response when further evidence came to light. Josephus says in Antiquities only that it appeared "at this very time" (37); in Wars, the earlier of these two histories, the time marker is "after this" (38). Yet we may assume that it did not surface on the day of the first trial, because in consequence, Herod "sent more ambassadors and letters [to Rome] to accuse his son" (39). So besides the first trial, there was a second, which Martin overlooks in his time estimates.
- Martin's case depends on another assumption, never brought into view, that a casual student might miss. Martin takes for granted that Caesar was inclined to rule in Herod's favor. Yet the trial of Antipater was not the first occasion when Herod sought Caesar’s permission to execute a son. In about 12 BC (40), he had found his sons Alexander and Aristobulus guilty of treason and pronounced them worthy of death, but to carry out the sentence he needed Caesar's approval, which Caesar refused to give, instead insisting on reconciliation (41). Only later, after Herod pressed him further on the basis of new evidence to permit their execution, Caesar reluctantly give Herod jurisdiction over his sons. Yet he advised clemency and stipulated that they be tried before an assembly of other rulers (42).
Even at that time, it was Caesar's judgment that Herod was an old man, with few years remaining before another succeeded him (43). Several years later, when Herod sought to execute Antipater, Caesar must have known that Herod's death was imminent. The ambassadors sent by Herod would surely not have concealed the truth from Caesar. Therefore, the interests of the Roman state required Caesar to consider a larger question than whether Antipater had plotted against his father. The question of overriding importance was who would succeed Herod.
Antipater for a long time been viewed as essentially a coregent with his father. "He governed the nation jointly with his father, being indeed no other than a king already" (44). He was well-known in Rome. Years earlier, in 13 BC (45), Herod had sent Antipater to Rome in the hope that Antipater would be made Caesar’s friend (46). Josephus reports that Antipater succeeded in making himself “a great figure among the influential of the city” (47). In the years following the execution of Alexander and Aristolulus, he continued to curry favor with the Roman elite, sending "great presents to his friends at Rome particularly, to gain their good will" (48). For seven months prior to his trial in Jerusalem, he had been in Rome, "getting interest among the great men; and on that account had bought splendid ornaments to present them withal, which cost him two hundred talents" (49).
Thus, when Herod's request to execute Antipater arrived at Caesar's palace, we can imagine that it found the atmosphere unfriendly. Antipater was a man with known abilities, established connections with powerful people, and, apparently, some experience in governing. The others who might succeed Herod lacked a comparable reputation. We might imagine that at the very least, Caesar made further inquiries before consenting to Herod's request, perhaps communicating with authorities in Syria. He needed information bearing on the likely aftermath of putting Antipater out of the way. He needed information also about all the others in line for the throne.
Evidence that the first round of charges against Antipater did not win Caesar’s consent to his execution appears in the second trial of Antipater. Sometime after the first trial, letters came to light incriminating Antipater in a plot to alienate Herod from his strongest political ally, his sister Salome (50). According to Josephus, “Hereupon Herod was in such great grief, that he was ready to send his son to Rome to Caesar, there to give an account of these his wicked contrivances. But he soon became afraid, lest he might there, by the assistance of his friends, escape the danger he was in; so he kept him bound as before, and sent more ambassadors and letters [to Rome] to accuse his son" (51). Here we find at least a hint that Antipater’s friends had won dismissal of the first accusations sent to Caesar or at least a postponement of action. Yet Herod could not have known what happened in Rome unless some news had returned from there. So, as much as two months may have separated the two trials.
Calendar of events
If we make allowance for delay in Caesar's deliberations, the date we are proposing for Herod’s death fits well into a reasonable timetable. Antipater went to Rome in the fall of 6 BC, either still en route or arriving before the Day of Atonement, on 24 October. He stayed in the capital during the next seven months after charges against him surfaced in Judea. Josephus confirms that it was winter, for he says that messages could not reach Antipater "because the roads were exactly guarded" (52). The implication is that Antipater’s sojourn fell in the season when the only long-distance communication was by land; ships were not making the journey. As Martin correctly observes, the sailing season on the Mediterranean lasted from about 10 March to 11 November (53).
The next summer he came home and was brought to trial twice. How long did the voyage from Rome to Judea require in an ancient Roman ship? Pliny the Elder, to show how quickly it was now possible to cross the Mediterranean, gives the time of passage for several recent voyages (54). Based on this information, James Smith, using the calculations of a British admiral, concluded that the rate of sailing under good conditions was about seven miles per hour (55). An account in the Book of Acts provides corroboration (Acts 28:13). Luke reports that the ship carrying Paul's party, including himself, took about one day to sail from Rhegium to Puteoli, a distance of 182 miles (56). For comparison, twenty-four hours of travel at seven miles per hour would progress 168 miles.
On Antipater's return voyage to Judea, his ship evidently took the customary route hovering close to Asia Minor, for Herod's reply to the letter Antipater sent before leaving caught up with him in Cilicia, in the southeast corner of that region (57). The distance from Puteoli, Rome's port city, to Sebastus, where Antipater disembarked, is roughly (very roughly, since we do not know the exact route) 1600 miles. The required time of transit was therefore no less than 9½ days even without bad weather or stopovers. How long after he wrote the letter before he set off from Puteoli is uncertain, but no doubt there was some delay. His voyage had at least one interruption, when he went aside into Cilicia. After hearing from his father, he sailed to Celenderis, but then hesitated before moving on, because he was unsure what his reception would be (58). A realistic estimate of the true interval between the dispatch of his letter and his arrival in Jerusalem would be four to six weeks.
The eclipse fell on 15 September. But the messages and ambassadors sent in two waves earlier to Caesar encountered resistance in the capital. Rather than immediately authorize Antipater’s execution, Caesar stalled his decision until December. Therefore, the answer did not reach Judea until after 1 January 4 BC, just days before Herod died.
Here is our proposed timetable.
- Antipater left Judea in the fall of 6 BC.
- Accusations against him surfaced in Judea perhaps in mid-October, before the Day of Atonement on the 24th.
- Antipater was in Rome for the next seven months.
- He sent a letter home about mid-May, 5 BC.
- He departed from Rome about 1 June.
- Antipater’s first trial was held about 1 July.
- His second trial was held about 1 September.
- The eclipse fell on 15 September.
- Caesar’s issued his consent to Antipater’s execution in about mid-December.
- The authorization to proceed did not arrive until early January, 4 BC.
- Herod died a few days later.
- The following Passover fell on 11 April.
The dates we have surmised for the eclipse and Herod’s death provide both a comfortable and a natural framework for the events that Josephus places within it. Throughout this period Herod was battling severe illness. For a while he was in one of his palaces west of Jordan. Then he spent time east of Jordan seeking relief in the warm baths at Callirrhoe. Not long before he died, he returned to Jericho (59).
Josephus reports that Herod died thirty-four years after he conquered Jerusalem and thirty-seven years after he was appointed king of Judea by the Roman Senate (60). He furnishes dates for both starting points. Herod's appointment was in the Roman year (running from January to January) when Caius Domitius Calvinus (for the second time) and Caius Asinius Pollio held the office of consul (61), the same as 40 BC (62). His conquest of Jerusalem was in the consular year of Marcus Agrippa and Caninius Gallus (63), which indisputably was 37 BC (64). The first date marks the beginning of his de jure authority, the second the beginning of his de facto authority.
The ancient kings of Mesopotamia, Syria, and Judea normally counted their regnal years on lunar calendars employing either Nisan or Tishri (these being the Jewish names) as the first month. Some dynasties included the partial year after a king took office (so-called nonaccession year reckoning), whereas others did not (known as accession year reckoning). Martin argues that if Herod died before Nisan of 4 BC, his reign could not have been regarded as spanning 37/34 years on any conventional calendar (65). If his first year started in Nisan of 40 BC, his thirty-seventh started in Nisan of 4 BC by nonaccession year reckoning and in Nisan of 3 BC by accession year reckoning. Also, if his first year started in Tishri of 40 BC, his thirty-seventh started in Tishri of 4 BC by nonaccession year reckoning and in Tishri of 3 BC by accession year reckoning.
But as we will show in our extended discussion of Herodian chronology, Herod measured his reign in Roman years running from January to January. So far as we know, the custom everywhere in the ancient world was to view the number of a king’s last regnal year as the length of his reign (66). Thus, if Herod's first de jure and de facto years were 40 and 37 BC, he reigned a total of 37/34 years if his last year was 4 BC; that is, if he died anytime during 4 BC.