The Year of Herod’s Death

The primary source of information about King Herod, other than the New Testament, is Josephus, a Jewish historian of the first century. From him we learn the following:

1. Herod died shortly after a lunar eclipse, but before passover.
2. Herod reigned 37 years from his appointment to the kingship (by Marc Antony).
3. Herod reigned 34 years from his actual accession after deposing his predecessor Antigonus.
   (Ant. XVII vi 4 - ix 3, viii 1, Wars I xxxiii 8)

Eclipses which could be the one referred to occurred on 12/13 March 4 B.C. and 9 January 1 B.C., but none in 3 or 2 B.C. The German scholar Emil Schurer favored the former date in his book Geschichte des Judiscen Volkes (1901), though he mentions an earlier work by one F. Riess, Das Geburtsjahr Christi (1880) who favored the latter date, because "the numerous events that took place between the eclipse and the Passover could not be squeezed into the four weeks available in 4 B.C." (Filmer, p. 284, see reference below).

Emil Schurer opted for the earlier eclipse because of points (2) and (3) above, which he dated beginning with the years 40 and 37 B.C. respectively. These dates are inconsistent with the later eclipse. His choice required him to suppose Josephus (a) used the non-accession year reckoning, and (b) counted Herod’s regnal years from Nisan (so that the few days Herod lived past Nisan 1 in his last year counted as a separate regnal year). His conclusions are reflected in many commentaries and Bible dictionaries extant today.

However, a more recent study by William Filmer, "The Chronology of the Reign of Herod the Great," published in Oxford’s Journal of Theological Studies, October, 1966, argues persuasively that the actual dates beginning Herod’s 37 and 34 years were 39 and 36 B.C., and that Josephus used an accession year reckoning (so the initial year was an "accession year," and "year one" was the next year). These conclusions are consistent only with accepting the 9 January 1 B.C. eclipse as the one mentioned as just preceding Herod’s death and are equally valid whether Josephus used Nisan or Tishri regnal years.

Filmer augments his thesis with evidence respecting Herod’s age, deducing that Herod would have been (as Josephus stipulated) about 70 years old in 2 or 1 B.C., and a suggestion that 2 Shebat, listed in the first century Jewish text Megillat Ta’anit as a celebrated day, marked the day of Herod’s death. Filmer’s conclusions have been widely followed since.