Regnal Year Reckoning
in the Last Years of the Kingdom of Judah
Published in On the Way to the Postmodern: Old Testament Essays 1967-1998, Volume 1 (JSOTSup, 292; Sheffield: Sheffield Academic Press, 1998), pp. 395-425A long-debated question in studies of the Israelite calendar and of pre-exilic chronology is this: Was the calendar year in Israel and Judah reckoned from the spring or the autumn? The majority verdict has come to be that throughout most of the monarchical period an autumnal calendar was employed for civil, religious and royal purposes. Several variations on this view have been advanced. Some have thought that the spring reckoning, which we Þnd attested in the postexilic period, came into operation only during the exile,/1/ but many have maintained that the usual autumnal calendar gave way in pre-exilic Judah to a spring calendar as used by Assyrians and Babylonians./2/ Some have believed that even when a spring calendar was adopted in Judah for civil purposes, regnal years continued to be reckoned from the autumn./3/ Several scholars have claimed that in any case it was only in Judah that the autumnal reckoning was employed, northern Israel having adopted a spring new year at the time of the division of the Solomonic kingdom./4/ The minority view on the whole question has been that calendar and regnal years in both Israel and Judah were reckoned from the spring throughout the monarchy./5/ In a paper published elsewhere/6/ I have examined the biblical data usually advanced in favour of an autumnal new year, and have concluded that positive direct evidence for such a calendaric system in pre-exilic Israel and Judah is entirely lacking.
The purpose of the present article is to approach the problem rather more indirectly by subjecting to examination, not some isolated biblical passages that have been thought to be relevant to the question, but a set of chronological data, both biblical and extrabiblical, relating to a comparatively brief span of pre-exilic history, in the hope that either the spring or the autumn reckoning may prove to satisfy the exigencies of the data. For this purpose the closing decades of the kingdom of Judah are among the most promising for investigation, since we have for the period 609587/6 bce not only several precise biblical year and month dates, but also, for the greater part of that period, the Babylonian Chronicle texts/7/ containing many exact datings. The present study is thus a fragmentary contribution to exploring the relationship between archaeological discovery and the biblical texts, and as such is presented to the recipient of this volume with the respect and appreciation of the author.
1. Arguments for an Autumn New Year
The arguments that may be advanced in favour of an autumn (that is, Tishri)/8/ new year reckoning/9/ are these: 1. The dates given in 2 Kgs 25.8 and Jer. 52.12 for the fall of Jerusalem may be synchronized, it is said, only if Judaean regnal years are reckoned from Tishri 1./10/ These passages date the destruction of the temple, the breaching of the walls, and the captivity to 7/10.V.11/11/ of Zedekiah, which is also said to be the 19th year of Nebuchadrezzar. The 19th year of Nebuchadrezzar began in Nisan 586, when Zedekiah's 11th year, reckoned from Tishri, was still in progress, but when his 11th year if reckoned from Nisan was over (according to the most usual chronologies; see Table, columns II and VIII).
There are, however, three possible methods of avoiding this inference short of supposing that 'nineteenth' in the parallel passages in 2 Kings 25 and Jeremiah 52 is a transmissional or computational error.
(a) First, it is possible that while standard Babylonian reckoning of Nebuchadrezzar's regnal years began his year 1 with Nisan 604 bce (the 7 months from his accession on Elul 1 to Nisan 1 being reckoned, according to the normal practice, as his accession year), in the west Nebuchadrezzar's reign was reckoned as beginning with Nisan 605/12/ since he had by that time already taken his father Nabopolassar's place as commander of the armed forces./13/
According to this view, this was the system that was normally followed in the references to Nebuchadrezzar's regnal years in 2 Kings and Jeremiah; only in Jer. 52.28-30 was the standard Babylonian reckoning employed./14/ Thus the 8th year of Nebuchadrezzar in 2 Kgs 24.12 is the same as the 7th year in Jer. 52.28, and the 19th year in 2 Kgs 25.8 and Jer. 52.12 is the same as the 18th year in 52.29./15/
(b) Secondly, it can be claimed that the captivity in the 19th year of Nebuchadrezzar is not identical with the captivity said elsewhere to have occurred in the 18th year of Nebuchadrezzar (Jer. 52.29),/16/ and that therefore the date of the burning of Jerusalem and the captivity by Nebuzaradan, the captain of the guard, 7/10.V.19 of Nebuchadrezzar (2 Kgs 25.8//Jer. 52.12), need not be synchronized with the 11th year of Zedekiah. On this view the destruction of the city by Nebuzaradan occurred 13 months later than the taking of the city on 9.IV.11 of Zedekiah, that is, in the 18th year of Zedekiah./17/
The difÞculty with this view is that it conþicts with the natural sequence of events in Jeremiah 52: the walls are breached on 9.IV.11 of Zedekiah (vv. 6-7), and on 10.V, presumably of the same year, Nebuzaradan burned the city and took away captives (vv. 12-15).
(d) Thirdly, the 11th year of Zedekiah (reckoned from Nisan) and the 19th of Nebuchadrezzar may be synchronized by postulating that Zedekiah acceded to the throne some time after Nisan 1, 597 bce./18/ In that case, his Þrst regnal year would have begun in Nisan 596, and his 11th year in Nisan 586, the beginning of Nebuchadrezzar's 19th year.
The more common view is that within days of the fall of Jerusalem to the Babylonians on Adar 2 (March 16), 597, Nebuchadrezzar personally installed Zedekiah as his own appointee, and that Zedekiah's Þrst regnal year therefore began in a few weeks' time, in Nisan 1, 597. This more usual view has rested on two supports: (i) that 2 Chron. 36.10, 'Nebuchadrezzar sent and brought Jehoiachin to Babylon', shows that Nebuchadrezzar was no longer in Judah at the time of Jehoiachin's deportation on Nisan 10,/19/ and that he must therefore have installed Zedekiah as king before the beginning of Nisan; (ii) that the appointment of Zedekiah is assigned to the 7th year of Nebuchadrezzar in the Babylonian Chronicle,/20/ and thus occurred before Nisan 1, 597.
These arguments are not strong. Against (i) it may be remarked that if inferences may be drawn from 2 Chron. 36.10, it may also be signiÞcant that the appointment of Zedekiah is mentioned after the exile of Jehoiachin to Babylon; and against (ii) it may be observed that if Zedekiah were actually appointed shortly after the beginning of Nebuchadrezzar's 8th year we might still expect the event to be recorded in the context of the king's dealings with Judah in his 7th year, since the appointment of Zedekiah was simply the last event of a series./21/ Accordingly, the way is open for a reconsideration of the possibility that Zedekiah acceded after Nisan 1, 597 bce./22/
In favour of dating the accession of Zedekiah after Nisan 1, 597 bce there are two points: (i) The synchronisms of Ezek. 40.1 are satisÞed. Here the fall of Jerusalem, which we know occurred on 9.IV.11 of Zedekiah (2 Kgs 25.2-3; Jer. 29.2; 52.5-7), is apparently Þxed in 586 bce; and 586 is Zedekiah's 11th Nisan year only if his Þrst regnal year began in Nisan 596, that is, if he acceded to the throne after Nisan 1, 597. (ii) Jeremiah's oracle concerning Elam (49.34-39), dated 'in the beginning of the reign of Zedekiah' (ber'¡ît malkût ßidqiyyâ), Þnds a more appropriate setting on this supposition. The Babylonian Chronicle for Nebuchadrezzar's ninth year (beginning in Nisan 596) records a confrontation between Nebuchadrezzar and the king of Elam in that year, the king of Elam retreating in panic before Nebuchadrezzar had come within a day's march of him./23/ On the usual reckoning of Zedekiah's years, Nebuchadrezzar's 9th year paralleled Zedekiah's 2nd (reckoned from Nisan), so the prophecy may be presumed to have been uttered in Zedekiah's Þrst year, or even early in his 2nd, which the phrase bër'¡ît malkût can refer to only in a very general sense. But if Zedekiah's Þrst regnal year did not begin until Nisan 596, then the prophecy against Elam could naturally fall in the accession year of Zedekiah, that is, the ber'¡ît malkût ßidqiyyâ in a strict sense./24/
But against these two points it may be observed: (i) that another interpretation of Ezek. 40.1, which is expounded below (under I.2), does not lead to the date 586 for the fall of Jerusalem, and that it is therefore of no advantage to have Zedekiah's 11th year begin in 586; and (ii) that it is uncertain whether ber'¡ît malkût is a technical term for 'accession year'; even if it is, it is far from impossible that Jeremiah's oracle against Elam should have been spoken in March, 597 (Zedekiah's accession year, on the usual reckoning) though Nebuchadrezzar did not conduct a campaign against Elam until a year later.
The arguments thus far considered are therefore rather inconclusive.
A more signiÞcant objection to the theory of an accession of Zedekiah after Nisan 1, 597 is the following: Ezek. 24.1 dates the beginning of the siege to 10.X.9 (sc. of the captivity of Jehoiachin), while 2 Kgs 25.1, Jer. 39.1 (the day is lacking) and 52.4 date it to 10.X.9 of Zedekiah./25/ We know that Jehoiachin's captivity began in Nisan 597/26/ (Ezek. 40.1 with 2 Kgs 24.12). So it seems clear that Zedekiah's year 1 must also have begun in that year, and not in 596./27/
This difÞculty may be met, however, by supposing that the years of Jehoiachin's captivity were reckoned on a postdating principle like the accession-year system, that is, that year 1 of the captivity of Jehoiachin on Nisan 10, 597. The 'fourteenth year after the city was conquered' in Ezek. 40.1 must on this view also be calculated in the same way; that is, the 'Þrst year after the fall of the city' begin on Nisan 1 of the year following its fall on 9.IV.11 of Zedekiah (= 19 of Nebuchadrezzar), 586 bce. This method of reckoning, though lacking any established parallels other than the regnal year system, cannot easily be dismissed. If it is natural to call the 'Þrst year' of a king the Þrst calendar year that begins during his reign, it is not unnatural to call the 'Þrst year of the captivity' the Þrst calendar year that begins after that event, and the 'Þrst year after the fall of the city' the Þrst year that begins after that event.
Finally, in order to test thoroughly this proposal to date Zedekiah's years from Nisan 596 rather than Nisan 597, it is necessary to examine the other dates in Ezekiel that are reckoned from the captivity of Jehoiachin; for it has just been argued that to date Zedekiah's years from 596 requires also the dating of the years of Jehoiachin's captivity from 596.
On the remaining dates in Ezekiel, 24.1 and 40.1 having already been considered, it is clear that some are not integrated with known historical events and are therefore irrelevant to the question of the base point for the years of the captivity. This is the case with the dates in 1.1-2; 8.1; 20.1 and 32.1, 17. But with regard to those dates that may be associated with known historical events, the following observations may be made:
(i) Ezek. 26.1 dates an oracle against Tyre, in which the fall of Jerusalem is presupposed (26.2), to the 11th year of the captivity. This date has seemed too early to many, since if the city fell in 586 bce the 12th year of the captivity had already begun. Accordingly some have claimed that the date of 26.1 is applicable not to the oracle of 26.2-6, but to the subsequent oracle about the siege of Tyre (26.7-14)./28/ Others have attempted to solve the problem by reading with lxxa 'in the twelfth year'./29/ But if the present proposal, that year 1 of the captivity began in Nisan 596, is followed, the 11th year of the captivity began in 586, and the date of Ezekiel's oracle may be in order. The month is not given, but if it were later than the 10th month, in which Ezekiel received news of the fall of the city (33.21), it would make good sense./30/ It would also Þt well with the received opinion, following Josephus,/31/ that Nebuchadrezzar's siege of Tyre began in 586 bce.
(ii) Ezek. 29.1, from 12.X.10 (26 January, 586, on the present proposal),/32/ and 30.20, from 1.I.11 (20 April, 586), both threaten the defeat of Egypt, that is, of Egyptian troops who had come to the relief of the siege of Jerusalem (Jer. 37.5). The arrival of this Egyptian assistance has been recently dated to the early months of 587,/33/ but there is no reason why it should not have come only a few months before the fall of the city in mid-July 586 (cf. also Jer. 32.1-2).
(iii) Ezek. 31.1, from 1.III.11 (11 June, 586), depicts the defeat of the Egyptian relief as already accomplished. This again makes good sense a month before the Þnal fall of the city.
(iv) Ezek. 29.17, from 1.I.27 (26 March, 570), presupposes the end of the siege of Tyre, which we know from Josephus concluded in 573 or perhaps 572 bce./34/ The usual reckoning of the years of the captivity has year 27 beginning in 571; but if the oracle is dated a year later, that is neither here nor there. If in fact the recent suggestion is accepted, that 29.17 contains a reference to the preparations for a Babylonian offensive against Egypt in 568,/35/ the slightly later date of 570 would provide an even more likely setting for the oracle than the usual dating.
Thus, since the usual arguments against the supposition that Zedekiah's reign began after Nisan 1, 597 can be shown to have little weight, and the most substantial objection to it can be met by a plausible suggestion, it may be concluded that the present proposal is worthy of serious consideration as a method of synchronizing the 11th year of Zedekiah with the 19th of Nebuchadrezzar on a Nisan basis.
The evidence on this point may be summarized thus: three explanations of the discrepancies between the regnal years of Nebuchadrezzar and Zedekiah may be offered, making recourse to a Tishri system of reckoning unnecessary. The explanations are mutually exclusive, and it is hard to choose even between the two most plausible of them, the Þrst and the third. But either of these provides a satisfactory account of the chronology in terms of a Nisan new year.
2. Some dates in Ezekiel appear to support the 586 bce dating of the fall of Jerusalem. The signiÞcance of that date for the fall of the city is that in the opinion of most scholars it would imply that Zedekiah's regnal years were reckoned on a Tishri basis, since the siege is said to have ended in the 4th month of his 11th year (2 Kgs 25.2; Jer. 52.5), which would have fallen in 587 if a Nisan reckoning were employed. We have already considered an argument that would exclude the necessity for linking a 586 date with a Tishri new year,/36/ but it is as well to consider the present evidence separately.
Ezek. 40.1 dates a prophecy of Ezekiel's to the 25th anniversary of Jehoiachin's exile on the 10th day of the Þrst month./37/ The date of that prophecy is also said to be in the 14th year after the destruction of the city. 'In accord with these data the fall of the city would have taken place in the twelfth year of Jehoiachin's exile, and that is the year when an escapee brought word to Babylon of the city's fall (Eze. 33:21).'/38/
Ezekiel, it is generally agreed, used the Babylonian reckoning of the year from Nisan. Whether he reckoned 1 Nisan or 10 Nisan (the actual date of Jehoiachin's exile) as the Þrst day of the year is of no consequence for the present argument.
Now, according to the dates in Ezekiel, year 1 of captivity begins Nisan 597/39/ so year 12 of captivity begins Nisan 586, that is, in the year in which the fugitive arrived in the 10th month, bringing news of the fall of Jerusalem (Ezek. 33.21). It would be perverse not to agree that the city must have fallen in the same year (in the 4th month) as news reached Ezekiel in the 10th. Also, year 25 of captivity begins Nisan 573 which is the 14th year (Ezek. 40.1) after the fall of Jerusalem only when that event is dated in 586.
There thus seems to be a sound case for claiming that the dates in Ezekiel demand the date 586 for the fall of Jerusalem, which date in turn is said to demand a Tishri reckoning of Judaean regnal years.
An alternative interpretation, yielding a 587 date for the fall of Jerusalem, must be considered, however. First, there is sufÞcient textual evidence for treating '11th year of our exile' in Ezek. 33.21 as a serious alternative to the reading '12th year of our exile'./40/ A corruption from '11th' to '12th' could be accounted for as due to the appearance of '12th year' already in 32.1, 17, so that it may have seemed to a scribe implausible that the next dated oracle should belong to the 11th year. The reason for the break in chronological sequence is not far to seek, however; oracles about Egypt, dating from 12.X.10 (29.1-16), 7.I.11 (30.20-26), 1.III.11 (31.1-18), 1.XII.12 (32.1-16), 15.I(?).12 (32.17-32), have been collected together; but the oracle of 33.21-22 has nothing to do with Egypt, and its date is consequently not in series with the dates of the Egyptian oracles. Secondly, it may be questioned whether 'in the 14th year after the city was conquered' (40.1) must refer to a calendar year beginning with Nisan (whether Nisan 1 or 10). It may well be that Ezekiel reckons the exact anniversaries of the fall of the city, just as he does the exact anniversaries of the captivity of Jehoiachin (and cf. 24.2 'this very day'). In fact, the 10.I.25 year of exile, Nisan 10 of 573 bce, falls three months before the Þfteenth anniversary of a fall of Jerusalem in 587 bce, and so in the 14th year after that event. Reckoning of years from the fall of Jerusalem was not a normal method of expressing the date and in fact the only other occurrence of such a reckoning seems to be a literary reminiscence of this verse./41/ This further suggests that the reference concerns the anniversary years after that famous event, and is not an alternative nomenclature for the calendar year.
In summarizing the evidence on this point it may be said that although some dates in Ezekiel apparently point to the date 586 bce for the fall of Jerusalem and are thus at Þrst sight evidence for Tishri reckoning, closer examination of the evidence makes it equally possible that the dates in Ezekiel are consonant with a dating of the fall of Jerusalem in 587 bce.
3. Dan. 1.1 refers to a siege of Jerusalem by 'Nebuchadnezzar king of Babylon' and the captivity of some of the people in the 3rd year of Jehoiakim. Only on a Tishri reckoning of Jehoiakim's regnal years can a siege by Nebuchadrezzar have occurred in Jehoiakim's 3rd year (see Table).
Some kind of an attack upon Jerusalem during the reign of Jehoiakim is demanded by the evidence of 2 Chron. 36.6-7, which recounts that Nebuchadrezzar took Jehoiakim in fetters to Babylon, and also carried off some of the temple vessels. This cannot have happened at the time of the siege of Jerusalem in early 597 bce, since Jehoiakim died on December 7, 598, just before the Babylonian troops set out for Jerusalem./42/ So if the biblical Chronicles tradition is sound, it is reasonable to suppose that at some time during his reign 'Jehoiakim took part of the temple treasure [to Babylon] as a qatr-offering or as biltu ('tribute') to buy off the Babylonians'./43/ 2 Kgs 24.1, 'Nebuchadnezzar came up, and Jehoiakim became his servant three years', may in turn be related to the same circumstances.
Now the Babylonian Chronicle makes no mention of a siege of Jerusalem before 598597 bce, but it may reasonable be asked whether such a siege could be Þtted into the record of Nebuchadrezzar's campaigns in about the 3rd year of Jehoiakim.
Several dates have been suggested for such a siege of Jerusalem:
(a) In Nebuchadrezzar's Þrst year, in Dec. 604Jan. 603, between the sack of Ashkelon in Kislev (Dec.) 604 and Nebuchadrezzar's return to Babylon in Sebat (Feb.) 603./44/ If this was Jehoiakim's initial act of submission to the Babylonians, the three years of his vassaldom (2 Kgs 24.1) will have run until Kislev 601 (Nebuchadrezzar's 4th year). That is exactly the date of the near-defeat inþicted by the Egyptians upon the Babylonians/45/-a setback severe enough to keep Nebuchadrezzar at home the next year (his 5th), 'gathering together his horses and chariots in great numbers'./46/ The apparent turning of the tide in favour of the Egyptians will then explain why at this moment Jehoiakim 'rebelled against' Nebuchadrezzar (2 Kgs 24.1)./47/
The objection to this view is that on no system of reckoning does Dec. 604Jan. 603 fall in Jehoiakim's 3rd year.
(b) In Nebuchadrezzar's 2nd year when, it is said, he set out for the Hatti-land with a 'powerful army' for a campaign in which he employed 'great siege-towers'./48/ The lacuna in the Chronicle of this year's events 'most likely contained a report of speciÞc conquests, parallel to that for the campaign to Ashkelon in his Þrst year. This may well have been the conquest of Jerusalem, and its date would then fall during the autumn or winter of 603 bce.'/49/ Such a date corresponds with the 6th year of Jehoiakim, which may, it is suggested, be what is intended in Dan. 1.1; a slight emendation, of ¡ålô¡ 'three' to ¡¡ 'six',/50/ would yield the meaning: 'In the sixth year of the reign of Jehoiakim king of Judah, Nebuchadnezzar king of Babylon came to Jerusalem and besieged it'.
The objections to this suggestion are twofold: (i) If the events of Dan. 1.1-2 occur in Nebuchadrezzar's 2nd year, the events of Daniel 2, three years later, cannot also occur in his '2nd year' (2.1). Thus if the date in Dan. 1.1 (emended) is correct, the date in 2.1 is incorrect. (ii) There is no direct evidence that Nebuchadrezzar campaigned in the Hatti-land in his 2nd year. All indications of the locations of that year's campaign are missing from the Chronicle, and it is only by inference from the campaigns of other years that the editor has supplied a reference to the Hatti-land./51/ It is true that in his accession year, 1st, 3rd, 4th, 6th, 7th, 8th, 10th and 11th years, Nebuchadrezzar campaigned in the Hatti-land, but it is perfectly possible that in his 2nd year his attention was directed elsewhere, as was the case in his 9th year (expedition against Elam)./52/
(c) In Nebuchadrezzar's accession year (7 Sept. 6051 Mar. 604), during which, we are told, he 'went back again to the Hatti-land and until the month of Sebat [Feb.] marched unopposed/53/ through the Hatti-land'./54/ A siege of Jerusalem could have occurred during this period,/55/ but it is impossible that it could have taken place before the end of Jehoiakim's 3rd year; for that ended (on a Tishri reckoning) merely a month after Nebuchadrezzar acceded to the throne in Babylon, which would not allow enough time for Nebuchadrezzar to reach Jerusalem, let alone reduce it, even if he had left Babylon on the day after his accession.
(d) Before Nebuchadrezzar's accession to the throne, between the battle of Carchemish (605 bce, probably May-June)/56/ and Nebuchadrezzar's return to Babylon shortly after news reached him of his father's death on Ab 8 (August 15)./57/ The Babylonian Chronicle in fact records that after the battle of Carchemish 'Nebuchadrezzar conquered the whole area of the Hatti-country',/58/ and 2 Kgs 24.7 notes that 'the king of Babylon took all that belonged to the king of Egypt from the Brook of Egypt to the river Euphrates'./59/ But can this be in the 3rd year of Jehoiakim? Only if it can be assumed that (i) the '3rd year' in Dan. 1.1 was reckoned on a Tishri new year, accession-year system, while (ii) the 'fourth year' of Jehoiakim, which is said to be the year of the battle of Carchemish in Jer. 46.2, was reckoned on a non-accession-year system./60/
Against this suggestion, the following objections can be made:
(i) It may be seriously doubted that a non-accession-year system (in which the months between a king's accession and his Þrst new year are counted as his year 1) was in operation at this time in Judah. The best attested evidence for non-accession-year reckoning in Judah is associated with only a few decades in the ninth century bce,/61/ and it is almost universally agreed that accession-year reckoning must be presumed throughout the period under consideration here. A particular advantage of postulating accession-year reckoning in Jeremiah is that it enables a plausible synchronism to be established between Jeremiah 36 and the Babylonian Chronicle. For Jer. 36.9 records a national fast in Judah in the 9th month of Jehoiakim's 5th year; on accession-year reckoning that was the month (Kislev, 604 bce) in which the Babylonian army was proceeding to the sack of Ashkelon./62/ A convincing historical setting is thus provided for the reading of the scroll of Jeremiah's prophecies, written (or at least ordered to be written) sometime between 9 and 21 months previously, and stored up for such an occasion as this./63/
(ii) It involves the interposition of a Babylonian march to Jerusalem from Hamath (no place further south is mentioned in the Babylonian Chronicle) and a siege of the city in the short interval between May-June (Carchemish) and c. August 25 (news of Nabopolassar's death). Very little time, if any, would be available for the siege itself, when the days spent in marching are reckoned up (see below, under II.2). We do not know where Nebuchadrezzar was when he received news of his father's death, but it must be recalled that only three weeks elapsed between Nabopolassar's death and Nebuchadrezzar's arrival in Babylon.
(iii) It requires the dating of Jeremiah's speech in Jeremiah 25 to a rather unsuitable time. For if the '4th year of Jehoiakim' (Jer. 25.1) is reckoned on a non-accession-year system,/64/ it overlaps Nebuchadrezzar's accession year by only one month (Elul)/65/ (see Table); therefore Jeremiah's speech must be dated only two months or less after a successful siege by Nebuchadrezzar. Yet in it Jeremiah makes no allusion to that catastrophic event of a few weeks earlier, nor to the king who must have been at this time still absent in Babylon whither he 'may have been personally required to go . . . to take part in the victory celebrations as a conquered and vassal king'/66/ (in fetters, if 2 Chron. 36.6 is relevant here; and cf. Manasseh, 2 Chron. 33.11). Rather, Jeremiah speaks as if Jerusalem's encounter with Nebuchadrezzar lies still in the future (cf. 25.8-9 especially). We may conclude that although the speech Þts well into the months after Carchemish when Judah was growing complacent again, it would be a rather anomalous speech had Carchemish been swiftly followed by such a siege as that depicted in Dan. 1.1./67/
(iv) Likewise, if the '5th year of Jehoiakim' (Jer. 36.9) is reckoned on a non-accession-year system (with Tishri new year), the reading of Baruch's scroll at the fast of the ninth month would have taken place in Kislev (Dec.) 605, less than 6 months after the siege of Jerusalem referred to in Dan. 1.1. Again, the narrative (Jer. 36) does not read as if such events had just taken place; the 'word of the Lord' to Jeremiah after Jehoiakim's burning of the scroll (36.28-31) makes no allusion to any recent humiliation of the king at the hands of the Babylonians, but, like ch. 25, speaks as if Jehoiakim has yet to encounter Nebuchadrezzar.
(v) Further, it may be urged that the dates in Daniel are not self-consistent, and that it is not worth attempting to justify the date in Dan. 1.1. If Daniel was taken to Babylon c. June-July 605, and immediately began his three-year education (1.5), he cannot have completed his education before the end of the '2nd year of Nebuchadnezzar' (ended 9 April, 602) in which the king had his dream (2.1) and Daniel, already graduated (1.18-20), gave its interpretation. This conclusion could be avoided only if it were maintained that the 'three years' of 1.15 are counted inclusively; but against that must be set the evidence for a Þxed period of 3 full years' education, at least in the Persian court./68/
The conclusion of our discussion of this point is this: Dan. 1.1 is the only piece of biblical data that demands a Tishri reckoning of regnal years. But there are several difÞculties and implausibilities involved in the acceptance of Dan. 1.1 at face value. Hence, while the event mentioned in Dan. 1.1 may be authenticated by other biblical references, the date given must be viewed with reserve.
2. Arguments for a Spring New Year
The arguments in favour of a spring (that is, Nisan) new year in Judah at this time/69/ are these:
1. The battle of Carchemish is said in Jer. 46.2 to have taken place in the 4th year of Jehoiakim. The Babylonian Chronicle shows that the battle occurred before the death of Nabopolassar on Ab 8, which can be synchronized with Jehoiakim's 4th year only on a Nisan reckoning/70/ (see Table). On a Tishri new year basis the date in Jer. 46.2 must be corrected to 'third year of Jehoiakim'./71/
The only means of evading this conclusion are: (a) To suppose that a non-accession reckoning is employed in Jer. 46.2. But the improbability of this supposition has been pointed out above (under I.3). (b) To suppose that Jehoiakim ascended the throne before Tishri 1, 609 bce. But this view, when all the data are considered, rather supports a Nisan new year (see below, under II.2). Or (c) to suppose that the date 'in the 4th year' belongs to the prophecy and not to the battle./72/ But this entails the postulation of a very large parenthesis in the text, contrary to the plain sense./73/
Jer. 46.2 provides in fact the clearest piece of chronological data relevant to this period, and its witness to a Nisan system of reckoning can hardly be challenged.
2. Examination of the date of Jehoiakim's accession to the throne makes it more likely than not that his regnal years were reckoned from Nisan, not Tishri. Most scholars agree that Jehoiakim came to the throne after Tishri 1, 609; if so, the eleven years of his reign (2 Kgs 23.36; 2 Chron. 36.5), which ended on Mare¡wan 22 (December 7), 598,/74/ may have been reckoned either on a Tishri or a Nisan basis./75/ But if it can be shown to be likely that he ascended the throne before Tishri 1, 609,/76/ he reigned 11 years only on a Nisan reckoning (it would have been 12 on a Tishri reckoning).
In determining the date of his accession, the events that must be brought into relationship are:1. Battle of Megiddo, death of Josiah, accession of Jehoahaz.
2. March of Egyptians from Megiddo to Euphrates.
3. Egyptian crossing of Euphrates for attack on Harran./77/
4. Capture of Jehoahaz and accession of Jehoiakim.The chronological data are twofold:
a. No. 3 occurred in Tammuz (25 June-23 July), 609 (BM 21901, Rev. 66-67)./78/
b. 'Three months' intervened between No. 1 and No. 4 (2 Kgs 23.31; 2 Chron. 36.2).The question is this: Is No. 4 likely to have occurred before Tishri 1 (21 September), 609?
Two major uncertainties are: (i) How long precisely were the '3 months' of the reign of Jehoahaz? And (ii) how long did the Egyptian march (No. 2) take?
On (i), we have solid evidence that a 'three months' reign could actually be signiÞcantly longer: Jehoiachim is said in 2 Kgs 24.8 to have reigned 3 months, but in 2 Chron. 36.9 to have reigned 3 months and 10 days. We can presume that a period up to almost 3F(1,2) months could be rounded down to 3 months. On the other hand, it is equally possible that '3 months' could be a rounding up from 2F(1,2) months, or even, if we may postulate an inclusive system or reckoning, from 2 months 1 day. The last suggestion is unlikely; for inclusive reckoning was not usually employed in reckoning lengths of reign, since it complicated the computation of accurate totals. So it may be concluded that the '3 months' of Jehoahaz's reign could have been between 2F(1,2) and 3F(1,2) months.
On (ii), it may be estimated that the Egyptians' march from Megiddo to Carchemish on the Euphrates took about 4 weeks. According to A. Malamat,/79/ the Egyptians would have travelled 35 km a day, 'or even more (remembering that this took place during the long summer days)' and so could have covered the 500+ km, as he reckons it, between Megiddo and Carchemish in about two weeks. But in fact the distance must be more like 650 km (400 miles)./80/ And an average rate of 35 km (22 miles) a day seems rather high, even assuming, as Malamat does, that the statement attributed to Neco in 2 Chron. 35.21, 'God has commanded me to make haste', can count as historical evidence for the speed of the Egyptian advance. References to rates of march in Assyrian and Egyptian texts suggest that rather lower averages, between 10 and 15 miles per day, were more usual./81/ Even allowing the rate of 35 km a day to be reasonable, the distance would have required some 19 days, without taking account of rest days. Whether the 'long summer days' would improve or retard an army's performance is open to question, and it ought also to be recalled that with a force of any considerable size some hours per day are spent in getting the column moving,/82/ and in setting up the camp./83/ Forced marches were not of course unknown in antiquity, the most famous example in classical times being a march of 7 days and 7 nights by Antigonus I in 310 bce, in which he moved an army of 50,000 men almost 300 miles at a speed of 40 miles a day./84/ But surprise was of the essence of Antigonus's march, and this does not seem to have been a factor in the advance of Neco's army. The fastest realistic speed we can assume for the Egyptian army over such a distance is about 15 miles a day; so we should allow at least 27 days for the 400 miles./85/
If we adopt mean Þgures for (i) the days occupied by the march from Megiddo to the Euphrates, (ii) the date in Tammuz on which the Egyptian army crossed the Euphrates, (iii) the length of reign of Jehoahaz, we can propose the following dates for the four key events:1. Death of Josiah 8 June
2. Egyptian march 4 wks 3
3. Crossing Euphrates 8 July " mo
4. Accession of Jehoiakim 8 September Ø ØNow September 8, 609 bce, is a fortnight earlier than Tishri 1 (September 21). The signiÞcance of using a mean Þgure is that it shows that an accession of Jehoiakim before Tishri 1 is more likely than not on the basis of our available evidence./86/ The reign of Jehoahaz need only have lasted 3 months and 1 week and the crossing of the Euphrates have taken place a week later in Tammuz, for the tables to be turned; but the likelihood of these being the case is no greater than that of the opposite possibilities: Jehoahaz may have reigned less than 3 months, the Euphrates may have been crossed earlier in Tammuz.
The point may be put another way: if Jehoiakim ascended the throne after Tishri 1, as is generally thought to have been the case, Jehoahaz more probably than not ascended the throne after Tammuz 1, 3 months earlier. Therefore the Egyptian army must have arrived at the Euphrates in the same month (Tammuz) in which they fought Josiah at Megiddo. That seems, in the light of the foregoing discussion, to be rather unlikely; equally unlikely therefore is the original hypothesis, that Jehoiakim ascended the throne after Tishri 1.
If Jehoiakim's accession occurred before Tishri 1, 609, how is a Tishri new year hypothesis affected? On an accession-year system, the weeks between his accession and Tishri 1 would count as his accession year, and his year 1 would run from Tishri 1, 609. Thus, by the time of his death in Mare¡wan 598, he would be in his 12th regnal year, a conþict with the '11 years' of 2 Kgs 23.36 and 2 Chron. 36.5./87/ On a non-accession-year system (improbable in Judah at this time in any case),/88/ he would die in his 13th regnal year!
On the other hand, a Nisan new year reckoning would create no conþict with the biblical data:/89/ Jehoiakim's year 1 (accession-year reckoning) would begin on Nisan 1, 608, Carchemish would occur in his 4th year (Jer. 46.2) = Nebuchadrezzar's 4th (605 bce), and Jehoiakim would die in his 11th regnal year (late 598 bce).
This argument does not provide conclusive evidence for a Nisan new year, since it rests upon an estimation of probabilities, but within its limitations it is a sound reason for claiming that Nisan reckoning was employed during this period in Judah.
The conclusion reached in this study is this: the balance of evidence points towards the use of Nisan reckoning of regnal years in the closing decades of the kingdom of Judah. Of the three arguments in favour of Tishri reckoning, two (I.1, 2) have been shown to be inconclusive, since the date they rest upon is susceptible of reasonable explanation in a Nisan system; the third argument (I.3) uses a date which requires Tishri reckoning but which also carries with it several improbable corollaries. Of the two arguments in favour of Nisan reckoning, the Þrst (II.1) is clear evidence for a Nisan system, and the second (II.2) suggests that a Nisan system is more probable than not. The whole argument therefore depends largely on the respective weight of the data in Dan. 1.1 and Jer. 46.2. Most scholars would give more weight to Jer. 46.2 than to Dan. 1.1 even if Dan. 1.1 were free of problems, and the argument presented in II.2 (the accession of Jehoiakim) offers further support for the claim that Judaean regnal years were reckoned from Nisan.
Appendix
Esarhaddon's army in an Egyptian campaign marched (they were not mounted on camels, as A.T. Olmstead, History of Assyria [New York: C. Scribner's Sons, 1923], p. 382, thought!) 20 bru in 15 days, 4 bru in 2 days (3 times), 15 bru in 8 days (ANET, pp. 292b = ARAB, II, §558). A bru is 'over 10 kms' (CAD, B, 208), or more precisely, according to F. Thureau-Dangin, JA (10th series) 13 (1909), p. 98 (cf. E. Weidner, AfO 16 [195253], p. 20 n. 138), 10.692 km. Thutmosis III's army in Palestine is reckoned to have marched 'at the respectable rate of 150 miles in 9 or 10 days' and about 80 miles in 11 or 12 days (J.A. Wilson, ANET, p. 235 nn. 16, 18). Rameses II marched the 400 miles from Sile on the Egyptian frontier to Kadesh on the Orontes in exactly one month (ANET, pp. 255b-256a), that is, at the rate of 13 1/2 miles a day. See on the subject of the rate of march of Egyptian armies, E. Edel, 'Die Stelen Amenophis' II aus Karnak und Memphis mit dem Bericht über die asiatischen Feldzüge des Königs', ZDPV 69 (1953), pp. 97-176 (152-53).
In Old Babylonian times a daily rate of travel of 25-30 km (as the crow þies) is attested, for example, in the LarsaEmar itinerary; cf. W.W. Hallo, 'The Road to Emar', JCS 18 (1964), pp. 57-88 (63, 66, 72, 75, 77). But this Þgure does not take rest days into account, some of the journey appears to have been made by boat (cf. p. 69), and above all it is most unlikely that this itinerary relates to the large-scale movement of troops, since only one night was spent at the Þnal destination (Emar) before the homeward journey was begun (cf. p. 85). An army would do remarkably well, judging by this standard, to march an average of 20 km a day, rest days included.
Another set of data, relating to the Babylonian advance to Jerusalem in 598/7 bce, offers the possibility of a rough calculation of the rate of march, though the data are not precise enough to be really helpful. Malamat calculates that the army covered the 1600 km (c. 1000 miles) from Babylon in about two months at an average daily rate of 30 km (18.6 miles) 'during the short and rainy winter days' (IEJ 18 [1968], p. 144; similarly Noth, ZDPV 74 [1958], p. 137). We know that the army was mustered and set out in Kislev (18 Dec. 598-15 Jan. 597), and that its siege of Jerusalem ended on Adar 2 (16 March), 597. If we make two quite plausible assumptions, that they set out early in Kislev and that the siege of Jerusalem lasted only a few days (so Malamat; cf. Noth, ZDPV 74 [1958], p. 138), they had nearly three months for the journey and so need not have covered more than 12 miles a day. But it is in fact impossible to determine whether this speed or that given by Malamat is closer to the reality. Certainly it seems unlikely that the march could have been accomplished in a month (so Wiseman, Chronicles, p. 33), at a rate of 33 miles a day, or even in six weeks (Vogt, VTS 4 [1957], p. 93) at a rate of 24 miles a day. Any calculations are further complicated by the possibility that Nebuchadrezzar's troops were marched not from Babylon but from some garrison town near the borders of Hatti-land (cf. Noth, ZDPV 74 [1958], p. 137).
It has been calculated that Roman armies marched between 14.6 and 19.5 miles a day; but since a day of rest followed each two or three days of marching, an average of more than 15 miles a day over some weeks would have been exceptional (see Harry Pratt Judson, Caesar's Army: A Study of the Military Art of the Romans in the Last Days of the Republic [Boston: Ginn, 1902], pp. 52-53). It is true that the Roman army spent a great part of the day in constructing a camp, a practice that was perhaps not taken to such lengths during the route marches of ancient Near Eastern infantry, but against this waste of time must be set the fact that Roman soldiers covered their daily itinera on roads built especially for marching.
/