Daniel 9's Seventy Weeks and the Sabbatical Cycle

Which Years Were the Sabbatical Years?

by Bob Pickle

1. The Pledge of Nehemiah 10:31
2. Alexander's Grant of Tax Exemption
3. Judah Maccabee's Defeat at Beth-Zur
4. Simon's Murder
5. Herod's Conquest of Jerusalem
6. Herod Agrippa's Reading of the Law
7. Note of Indebtedness from Nero's Reign
8. Destruction of Second Temple
9. Land Contracts of Bar Kochba
10. Tombstones from Zoar

"1. The pledge to keep Shemitah" (pp. 157, 158)

Regarding this pledge in Nehemiah 10:31, Wacholder says:

Briefly, we regard Neh. 10:32(31)c as a reasonably datable allusion to the observance of Shemitah during the post-exilic period, but because of the many doubts involved, forego its precise dating. (p. 158)

The doubts he refers to are those suggested by higher criticism. Wacholder is uncertain whether the passage belongs to Ezra or Nehemiah, whether it is connected with Artaxerxes I or Artaxerxes II, and whether it belongs to this monarch's seventh year or twentieth year.

William H. Shea summarized Wacholder's ten sections of evidence in Symposium on Daniel (Biblical Research Institute (1986), pp. 252-255). Unhampered by higher criticism constraints, Shea takes the passage just as it reads and arrives at some definite conclusions:

The biblical text describes the occasion upon which the people of Judah took a pledge to observe the provisions of the Sabbatical year. This occurred when they met to celebrate the Festival of Tabernacles after Nehemiah had repaired the walls and gates of Jerusalem. Nehemiah returned to Palestine to accomplish this task in the twentieth year of Artaxerxes I (444/443 B.C., fall-to-fall). The first Sabbatical year which they took the pledge to observe began in the same fall of 443 they met to celebrate this festival.

Well put, but there is one major problem due to a minor mathematical error: Artaxerxes's twentieth year was in 445/444 BC, not 444/443. And Shea would wholeheartedly agree given the fact that he is certain that Artaxerxes' seventh year is 458/457 BC (cf. p. 225). 

If the twentieth year was in 445/444, then Tishri in the twenty-first year would be in 444 BC, not 443. This supports Zuckermann's sabbatical dates. [table]

Admittedly, this line of evidence is not real strong, since it is based on the assumption that the pledge of Neh. 10:31 was made in a sabbatical year. Shea correct summarizes:

This text does not prove that that year was a Sabbatical year, but its chronological fit with the years listed below adds significance to this part of the service on that occasion. [return]

"2. Alexander Exempts the Jews from Taxation During Shemitah" (pp. 158-160)

Wacholder acknowledges that either Zuckermann's sabbatical dates or his own could fit the historical record of this incident, but he considers his dating "preferable" (p. 160).

Though not explicitly stated in his paper, Wacholder assumes that Josephus's dating of Alexander's grant is incorrect. He postulates that the date of the grant was really in the spring or summer of 331 BC after Alexander had appointed a governor over Cole-Syria, while Josephus plainly put the date of the grant in the previous fall around 6 months earlier.

If we thus revise the source, at some point we can no longer use that source to support our position. If an unemended source is untrustworthy as far as the date goes, how can we use it to prove the date when a sabbatical year occurred?

Suppose for the sake of argument that Wacholder's hypothesis is correct, and that the grant occurred in the spring of 331 BC, not the fall of 332. This would have still been within a sabbatical year according to Zuckermann. Whether that year or the next was a sabbatical year, either way the Jews would have been concerned about paying taxes during sabbaticals, and would have requested an exemption from Alexander. Either they had no harvest that year with which to pay the new taxes, or they were anticipating not having anything to pay the taxes with the following year. 

Thus even if Wacholder's hypothesis is correct, Zuckermann's dates can still fit. [table] [return]

"3. Judah Maccabee's Defeat at Beth-Zur Ascribed to Shemitah" (pp. 160-163)

If the siege cannot be dated beyond reasonable doubt, then the associated sabbatical year likewise cannot be dated, and Wacholder therefore concludes:

. . . the evidence from here alone is not conclusive. (p. 163)

Yet the lack of conclusiveness is apparent rather than real because of certain discrepancies in his argument. To illustrate:

But the Maccabean books, like all other biblical sources, without exception, take it for granted that Nisan was the first month. (pp. 161, 162)

This is incorrect, since Nehemiah 1:1 and 2:1 is an explicit exception. Simply saying that these texts are "probably corrupt" (n. 8, p. 155) does not change this fact.

Other discrepancies appear when one analyzes his Table One, reproduced below:

A, B, and C are supposed to support Zuckermann, while D and E are supposed to support Wacholder. [table] Regarding B and C Wacholder says:

C accepts B's reasoning, but, because of other alleged divergences between the chronological schemes of the two Maccabean books, assumes that I Macc. posits a Seleucid era which started in Nisan 311; II Macc., in October 312 B.C.E. (pp. 163) 

Yet Table One plainly shows that B and C are identical, not different. Furthermore, "Nisan 311" in the quoted sentence should have read "Nisan 312" if the 150th year is going to be 163/162.

D differs from E in that it presumes an error in First Maccabees; . . . . (Ibid.)

The error presumed is really in 2 Maccabees, according to the table. Only an assumed error in 2 Maccabees, not in 1 Maccabees, will harmonize the data with Wacholder's dates.

. . . while E grants that I Maccabees began the Seleucid era in October 312 and II Maccabees, in Nisan 311; both sources agree that Antiochus V's campaign occurred in the spring or summer of 162. (Ibid.)

Table One shows instead 1 Maccabees commencing the Seleucid Era in Nisan 311, and 2 Maccabees in October 311. Shall we correct his table or his text? Let's put the two possibilities side by side:

Neither appears acceptable. If we correct the table, 2 Maccabees ends up supporting Zuckermann, 1 Maccabees ends up supporting neither, and neither account puts the campaign in the spring or summer of 162. If we correct the text, we end up with a Seleucid Era for 2 Maccabees that commences in October 311, over a year after Seleucus's entry into Babylon.

The only way to make this line of evidence support Wacholder's sabbatical dates is to either assume an error in 2 Maccabees or assume that the Seleucid Era in 2 Maccabees commenced in the fall of 311 BC. Thus we must use Zuckermann's sabbatical dates if we are going to avoid rewriting the source and commencing the Seleucid Era unusually late. [table] [return]

"4. The Murder of Simon the Hasmonean in 177 A.S." (pp. 163-165)

Again without stating such, Wacholder has assumed that there is an error in the source material. How then can it be relied upon to give any sabbatical date? Wacholder states:

Josephus' dating of Simon's death during the Shemitah of 177 A.S. [Seleucid Era], which is equivalent to 135/34 B.C.E., offers unambiguous testimony for the calendar of Sabbatical cycles appended to this study. (p. 165)

Notice carefully what the above statement said: "Josephus' dating of Simon's death during the Shemitah." Yet as Wacholder himself quotes Josephus, Simon's murder by his son-in-law occurred as much as eight months before the commencement of the sabbatical year, not four months after its commencement:

Furthermore, Josephus adds in both the Bellum and Antiquities that John Hyrcanus' efforts to avenge the heinous crime were futile, for while John besieged Ptolemy's fortress, which Josephus calls Dagon, "there came round the year in which the Jews are wont to be inactive, for they observe the custom every seventh year, just as on the seventh day." (p. 163)

Apparently, Wacholder felt that the murder had to be in February 134 BC instead of February 135 BC. The only way to harmonize this with his sabbatical dates is to have the murder take place during the sabbatical instead of before its commencement, and assume Josephus to be in error. Thus Wacholder is synchronizing February of the 177th year with a sabbatical year rather than with the sixth year of the seven-year cycle. [table]

The chart below gives the various possible dates for the murder and the sabbatical under the different dating schemes of the Seleucid Era:

Now let's compare this with the data from the previous section, assuming that 1 Maccabees is using the same dating scheme for both the 150th and the 177th year. Since the two sabbatical years in question must be 28 years apart (177 minus 149 or 178 minus 150), only dates that are 28 years apart are shown. According to Zuckermann, 136/135 was a sabbatical year. According to Wacholder, the sabbatical year was 135/134.

The only way to have Wacholder's sabbatical dates work is to use a dating scheme for the Seleucid Era that commences in the spring of 312 or the fall of 311 BC for 1 Maccabees. Yet as we saw in the previous section, 1 Maccabees does not use either of these dating schemes for the siege of Beth-Zur. Thus the only way to harmonize the data without assuming an error in either 1 Maccabees, 2 Maccabees, or Josephus is to use Zuckermann's sabbatical dates. [table]

It is true that all the varying chronological statements by Josephus are impossible to reconcile if left unemended. Yet if his testimony is tampered with too much, then how can it be used to prove Wacholder's position? There must be a certain level of trust in Josephus's testimony, or it cannot be used to prove his case. That being said, we now examine several questions that have a bearing on this section.

1. Should the year begin in the spring or fall? Wacholder maintains that the years of the Seleucid Era must commence with Nisan. Notice that this can still be done if one assumes that Josephus erred in placing the murder before the sabbatical. Yet the sabbatical date arrived at is still Zuckermann's regardless, since only Zuckermann's dates allow us to harmonize a sabbatical associated with the 177th year with one in the 150th year.

2. Which scheme of Seleucid dating does 1 Maccabees use? That depends on if we let Josephus answer that question. 1 Maccabees tells us that the temple was desecrated by Antiochus Epiphanes in Kislev in the 145th year of the Seleucid Era, and Josephus says that this was in the 153rd Olympiad (Antiq., bk. 12, ch. 5, sect. 4).

Olympiads are four-year periods that run from July 1 through June 30.

Since the 153rd Olympiad ran from July of 168 to June of 164 BC, commencing the Seleucid Era in the fall of 313 BC doesn't work at all, for the temple's desecration would occur in the 152nd Olympiad, not the 153rd. 

For that matter, since Josephus synchronizes 1 Maccabees's date of the rededication of the temple in Kislev of the 148th year with the 154th Olympiad (164 to 160 BC; Antiq., bk. 12, ch. 7, sect. 6), his synchronisms rule out entirely the first three schemes. Only a rededication in the fall of 164 BC would synchronize with the 154th Olympiad. But this causes some definite problems.

As can be seen from the table of the four schemes of the Seleucid Era dating under the first section, if we must use a Spring 311 beginning for the Seleucid Era, we end up with a sabbatical date for the 150th year that agrees with Wacholder's dates. Yet we are then left with the conclusion that 2 Maccabees erred in identifying the siege of Beth-Zur with the 149th year. 2 Maccabees should then have agreed with 1 Maccabees that this siege occurred in the 150th year. That is the only way to synchronize the two accounts if 1 Maccabees's Seleucid Era begins in the fall of 311.

But we also have 2 Maccabees 14:1-4 saying that three years after the 149th year was the 151st year. While there are three years from 149 to 151 inclusive, there is no way that there can be three years inclusive from 150 to 151. So we must keep the reading of 149 in 2 Maccabees 13:1.

If we do not tamper with the dates given in 2 Maccabees, we have but one other option: Did somehow Josephus err in the Olympiads that he gave for the 148th and/or 145th year? In light of the following quote, perhaps:

But Antiochus, being very uneasy at the miseries that Simon had brought upon him, he invaded Judea in the fourth year of his reign, and the first year of the principality of Hyrcanus, in the hundred and sixty-second olympiad. (Antiq. 13.8.2)

This takes us back to the subject of this section: Simon's murder and John Hyrcanus's accession in the 177th year. The problem is that the 162nd Olympiad did not begin until July 1, 132 BC. Since Simon's murder and Hyrcanus's accession took place in 134 BC at the latest, it occurred in the 161st Olympiad, not the 162nd. Since the existence of this discrepancy is indisputable, we have here an example of Josephus being one Olympiad off when pairing up Seleucid dates with Olympiads. Since we are talking about the same general time period, he could easily have been off one year in his reference to the 154th and/or the 153rd Olympiads.

The indisputable error of the synchronism of the 162nd Olympiad, though, does have a bearing on whether or not Zuckermann's sabbatical dates are correct, as we shall now see.

3. Has anyone attempted to synchronize a sabbatical with the 162nd Olympiad? Josephus gives us three different figures for the reign of John Hyrcanus: thirty years, thirty-one years, and thirty-three years (Antiq. 20.10.1; 13.10.7; Wars 1.2.8). In contrast, Sulpitus Severus (d. 420 AD) gives us a figure of only twenty-six years (The Sacred History of Sulpitius Severus, bk. 2, ch. 26). In view of the Hasmonean chronology that Severus gives, it appears that he or whomever he was following was attempting to place Simon's murder in the 162nd Olympiad:

[Hyrcanus] died, after having held the supreme power for twenty-six years. After him, Aristobulus being appointed high-priest, ... At the close of a year, he died. Then Alexander, his son, who was both king and high-priest, reigned twenty-seven years .... He having left two young sons named Aristobulus and Hyrcanus, Salina or Alexandra, his wife, held the sovereignty for three years. Hyrcanus [II] held the chief power for thirty-four years; but while he carried on war against the Parthians, he was taken prisoner.

Then Herod, a foreigner, the son of Antipater of Askelon, asked and received the sovereignty of Judaea from the senate and people of Rome. ... Under this Herod, in the thirty-third year of his reign, Christ was born on the twenty-fifth of December in the consulship of Sabinus and Rufinus. Herod reigned four years after the birth of the Lord; for the whole period of his reign comprised thirty-seven years. (bk. 2, ch. 26, 27)

While Severus cannot be correct regarding the year of Herod's death, we can extract some helpful information. First of all, the consulship of Sabinus and Rubinus occurred in 4 BC. Thus Severus has Herod beginning his reign thirty-three years earlier in 37 BC. This is a fixed date from other sources, the date Herod took Jerusalem by force, as shown under the next section. Secondly, adding up the lengths of reigns from the first Hyrcanus to the last Hyrcanus, we have a total of 91 years as given by Severus. Adding these to 37 BC, we then have the first Hyrcanus beginning his "twenty-six" years of reign in the spring of 128 BC, in the latter end of the 162nd Olympiad.

If Severus allowed some time between the second Hyrcanus' capture by the Parthians and Herod's beginning to reign, which could be expected, then we have more than 91 years from the first Hyrcanus to Herod.

How much more? In actuality, there were three years between Hyrcanus's capture and Herod's taking of Jerusalem, according to Josephus's account (Antiq., bk. 14, ch. 13-15). Soon after the capture Herod was given the kingship by Rome, and it took him three years to make it good. Josephus tells us that Herod reigned 37 years after being given the kingship and 34 years after taking Jerusalem (Antiq., bk. 17, ch. 8, sect. 1; Wars, bk. 1, ch. 33, sect. 8). Since Severus has the 37 years ending four years after the birth of Christ in 4 BC, then he cannot be extending his reign back beyond 37 BC. Moreover, Severus makes mention only of Herod's receiving the kingdom from the Romans, thus indicating that he associated that event with the commencement of Herod's 37-year reign, not his taking of Jerusalem. The Chronicle of Eusebius gives the same dates for the beginning and end of Herod's reign.

That being so, if we add a minimal amount of time between Hyrcanus II's capture and Herod's receiving the kingdom, we push back Hyrcanus I's accession one year to the spring of 129 BC. This date allows a sabbatical year to begin in the following fall while still in the 162nd Olympiad, and while Hyrcanus was still in his first year of reign.

It therefore appears that Severus's chronology is an attempt, by him or whomever he is following, to synchronize the beginning of Hyrcanus' reign with the 162nd Olympiad, and perhaps even with a sabbatical year in that Olympiad. Why is this significant? Because only Zuckermann's sabbatical dates have a sabbatical commencing in the 162nd Olympiad. That Olympiad ended on July 1, 128 BC, and Wacholder's sabbatical does not commence until the following fall. [table]

The following table gives the sabbatical dates according to Zuckermann and Wacholder in relationship to the 161st, 162nd, and 163rd Olympiads.

All things considered, it appears that this section is solidly in support of Zuckermann, and that there is no good way to harmonize the data with Wacholder's sabbatical dates. [return]

"5. Herod's Conquest of Jerusalem" (pp.165-167)

Some scholars other than Wacholder would like to have Herod conquer Jerusalem in 36 BC instead, yet this is not possible. Twice Josephus informs us that the Battle of Actium (summer of 31 BC) occurred in the seventh year of Herod's reign (Antiq., bk. 15, ch. 5, sect. 2; Wars, bk. 1, ch. 19, sect. 3). If he took Jerusalem in 36 BC, then 31 BC would have been his sixth year by non-accession-year reckoning, not his seventh year. So the data Josephus gives us regarding the Battle of Actium mandates that Herod's taking of Jerusalem be in 37 BC, not in 36 BC.

Concerning the earlier conquest of Jerusalem by Pompey, Josephus writes:

. . . for although the city was taken on the third month, on the day of the fast, upon the hundred and seventy-ninth olympiad, when Caius Antonius and Marcus Tullius Cicero were consuls . . . . (Ibid., bk. 14, ch. 4, sect. 3)

The 179th Olympiad began in the summer of 64 BC, and these men became consuls in 63 BC. Thus the date of the city being taken must be 63 BC.

Concerning Herod's conquest of Jerusalem:

. . . they persisted in this war to the very last; and this they did while a mighty army lay round about them, and while they were distressed by famine and the want of necessaries, for this happened to be a Sabbatic year. (Ibid., bk. 14, ch. 16, sect. 2)

This destruction befell the city of Jerusalem when Marcus Agrippa and Caninius Gallus were consuls of Rome on the hundred eighty and fifth olympiad, on the third month, on the solemnity of the fast, as if a periodical revolution of calamities had returned since that which befell the Jews under Pompey; for the Jews were taken by him on the same day, and this was after twenty-seven years' time. (Ibid., bk. 14, ch. 16, sect. 4)

Some think that this conquest must be in 36 BC since 36 BC is 27 years later than Pompey's conquest in 63 BC. However, if we use inclusive reckoning, similar to what we must use for Herod's reign, 63 to 37 BC is indeed 27 years.

Notice that both conquests are said to have occurred on the same day, the day of the "fast." More on this in a moment.

Regarding events immediately following Herod's conquest of Jerusalem, Josephus writes:

. . . nor was there any end of the miseries he brought upon them; and this distress was in part occasioned by the covetousness of the prince regent, who was still in want of more, and in part by the Sabbatic year, which was still going on, and forced the country to lie still uncultivated, since we are forbidden to sow our land in that year. (Ibid., bk. 15, ch. 1, sect. 2)

Wacholder and others identify the "fast" as being the Day of Atonement, though this opinion is by no means universal. If it were true, then we would have Josephus making an obvious blunder in adjacent parts of his Antiquities. He would be saying that 38/37 BC, when Herod besieged Jerusalem, and 37/36 BC, the year after, were both sabbatical years. Of course, two calendar years in a row can't both be sabbatical years.

In actuality, Josephus' testimony excludes the possibility that the "fast" was the Day of Atonement:

When the rigor of winter was over, Herod removed his army, and came near to Jerusalem, and pitched his camp hard by the city. Now this was the third year since he had been made king at Rome; and as he removed his camp, and came near that part of the wall where it could be most easily assaulted, he pitched that camp before the temple, intending to make his attacks in the same manner as did Pompey. (Ibid., bk. 14, ch. 15, sect. 14)

Now as winter was going off, Herod marched to Jerusalem, and brought his army to the wall of it; this was the third year since he had been made king at Rome; so he pitched his camp before the temple, for on that side it might be besieged, and there it was that Pompey took the city. (Wars, bk. 1, ch. 17, sect. 8)

"When Herod, the son of Antipater, brought upon us Sosius, and Sosius brought upon us the Roman army, they were then encompassed and besieged for six months, till, as a punishment for their sins, they were taken, and the city was plundered by the enemy." (Ibid., bk. 5, ch. 9, sect. 4)

Indeed, though they had so great an army lying round about them, they bore a siege of five months, till some of Herod's chosen men ventured to get upon the wall, and fell into the city . . . . (Ibid.,  bk. 1, ch. 18, sect. 2)

To have the city taken on the Day of the Atonement after a siege of five months would require that Herod arrive at Jerusalem in the second month. Yet it is inconceivable that the rigor of winter would not be over until the second Jewish month, roughly our month of May.

Adar 13, roughly early March, is not too early for war in Palestine, as Nicanor's defeat indicates (1 Mac. 7:49; Antiq., bk. 12, ch. 10, sect. 5). Thus if Herod approached Jerusalem by mid-Adar, five or six months would definitely be over before Tishri 1. The "fast" should therefore refer to another day than the Day of Atonement.

Also, using a spring-to-spring calendar, if Herod approached Jerusalem before the new year commenced on Nisan 1, 37 BC, it would be the third year inclusive since he had been appointed king by the Romans in early 40 BC, just as Josephus twice specified (Ibid., bk. 14, ch. 15, sect. 14; Wars, bk. 1, ch. 17, sect. 8). If he instead approached Jerusalem six months before the Day of Atonement, the new year would have already commenced, and it would have been during his fourth year inclusive that he began to besiege Jerusalem.

If we identify the "fast" with that of Tammuz 17 instead of the Day of Atonement, as the above evidence suggests that we should do, then Josephus is not forced to contradict himself in adjoining portions of a single account. Therefore, Josephus is referring to the same sabbatical year in the same calendar year, before and after Herod's conquest of Jerusalem that occurred some time before Tishri 1, 37 BC, probably on Tammuz 17. That makes this sabbatical year coincide with 38/37 BC in harmony with Zuckermann's sabbatical dates. [table] [return]

"6. King Agrippa I Recites Deut. 7:15 in a Post-Sabbatical Year" (pp. 167-169)

Wacholder's objection to Agrippa's reading the law in a post-sabbatical year according to Zuckermann's dates (41 AD) is that "An incidental remark in Josephus shows, however, that 40/41 could not have been a Shemitah." (p. 168). [table] This remark has to do with Caius Caligula ordering that his statue be set up in the Jewish temple. Many Jews subsequently went to the city of Tiberias to plead with Petronius, the governor of Syria:

So they threw themselves down upon their faces, and stretched out their throats, and said they were ready to be slain; and this they did for forty days together, and in the mean time left off the tilling of their ground, and that while the season of the year required them to sow it. (Antiq., bk. 18, ch. 8, sect. 3)

But as they could be no way prevailed upon, and he saw that the country was in danger of lying without tillage; (for it was about seed time that the multitude continued for fifty days together idle;) . . . (Wars, bk. 2, ch. 10, sect. 5)

Wacholder writes:

We know that the turmoil in Judaea described by both Josephus and Philo took place during the final months of the reign of Caligula, who was assassinated on the twenty-fourth of January of 41 C.E. (p. 168)

Wacholder maintains that 40/41 AD could not have been a sabbatical year, for why would then Josephus make a point of their voluntarily not sowing their crops if they couldn't sow them anyway? The point would only make sense if 40/41 was not a sabbatical.

Yet this conclusion is based on the assumption that we are talking about the sowing of crops sometime between Tishri 22 and Caligula's death. Indeed, that is when certain grain crops are sown which are later reaped in the spring, but that cannot be what Josephus is talking about here. As Wacholder himself says:

Philo Alexandrinus, referring to the same incident records: "For the wheat crop was just ripe, and so were the other cereals." (Ibid.)

While Josephus was not an eyewitness of these events, Philo was personally involved. If these grain crops were just ripe, as Philo indicates, then we are talking about the spring of 40 AD, not the fall. The neglected sowing Josephus referred to must have been summer vegetables, not grain crops planted in the fall.

In Palestine today, such summer vegetables are "planted after mid spring when the danger of frost has passed and temperatures have begun to warm." These include "tomato, squash, okra, snakecucumber, watermelon, cantaloupe" (http://www.arij.org/pub/dryland/sec6.htm).

Mid spring would be around the beginning of May. Wheat in Palestine back then was harvested in April, May, or June, depending on the area. Since the only sowing that might coincide with ripe wheat would be of summer vegetables, this must be what Josephus was referring to. That being so, there is no conflict with such a sowing and Zuckermann's sabbatical dates, for the sabbatical would not yet have begun.

Essentially, we are left in this section without a case for either Zuckermann's or Wacholder's sabbatical dates. [table] The only case that really could be made is the excluding of one date or the other, and that cannot be done with this line of evidence. [return]

"7. Note of Indebtedness on a Papyrus of Wadi Murabba`at" (pp. 169-171)

Josephus appears to have used a spring-to-spring calendar in his account of the Jewish War:

Now this war began in the second year of the government of Florus, and the twelfth year of the reign of Nero. (Antiq., bk. 20, ch. 11, sect. 1)

. . . at the same time began the war, in the twelfth year of the reign of Nero, and the seventeenth of the reign of Agrippa, in the month of Artemisius [Iyar.] (Wars, bk. 2, ch.14, sect. 4)

This defeat [of Cestius] happened on the eighth day of the month Dius, [Marchesvan,] in the twelfth year of the reign of Nero. (Wars, bk. 2, ch.19, sect. 9)

Since the year of reign did not change between the beginning of the revolt in the second Jewish month and the defeat of Cestius in the eighth Jewish month, both in 66 AD, it appears that he was indeed using a spring-to-spring calendar. For certain, he was not dating Nero's reign from either Tishri or the anniversary date of Nero's accession. 

Possibly Josephus was using accession-year reckoning, for non-accession-year reckoning would make the summer of 66 AD Nero's thirteenth year instead of his twelfth. Then again, he may have been using an incorrect, later date for Nero's accession, since he has Nero reigning but thirteen years and eight days (Wars, bk. 4, ch. 9, sect. 2).

Regarding the destruction of the temple in the fifth Jewish month in 70 AD, Josephus writes:

. . . its destruction, which happened in the second year of the reign of Vespasian . . . . (Wars, bk. 6, ch. 4, sect. 8)

And regarding the conquest of Jerusalem in the sixth Jewish month of 70 AD:

And thus was Jerusalem taken, in the second year of the reign of Vespasian, on the eighth day of the month Gorpeius [Elul]. (Wars, bk. 6, ch. 10, sect. 1)

If the summer of 70 AD was Vespasian's second year, Josephus must have been using non-accession-year reckoning. Vespasian was proclaimed emperor by his troops in July of 69, and did not begin to reign officially until December of 69. Additionally, if Josephus was dating Vespasian's reign from December of 69, then we have further evidence of his use of a spring-to-spring calendar for these reigns.

Wacholder's paper refers to the dating used in the Bar Kochba revolt, a revolt lasting three and a half years from early or mid 132 to mid or late 135 AD (n. 92, p. 179). He cites a document published by Yadin, which was dated in Iyyar (second month) of the first year of the revolt. He also cites Mur 22 and Mur 30, documents dated in Heshvan (eighth month) of the first year and Tishri of the fourth year. 

In order for Mur 30 to be dated in the fourth year of a revolt that lasted three and a half years, non-accession-year reckoning is required. A spring-to-spring calendar must have been used, or else Mur 22 would have been dated in the second year of the revolt, not the first.

The first revolt that began during the reign of Nero was reckoned similarly, according to numismatic evidence. Shekels exist bearing dates of year 1 through year 5 (B. Kanael, "Notes on the Dates Used During the Bar Kokhba Revolt," Israel Exploration Journal 21 (1971), n. 11, p. 40). The revolt began in the spring and summer of 66 AD, and was squelched in the summer of 70 AD. Only by using non-accession-year reckoning and a spring-to-spring calendar can we have a year 5 of the revolt by the summer of 70 AD. [return]

"8. Was the Second Temple Destroyed During a Sabbatical or a Post-Sabbatical Year?" (pp. 171-176)

Of course, there are some things in the Talmudic statement that cannot possibly be true, like the year of the destruction of Solomon's temple. By placing the destruction of Solomon's temple in 421 BC instead of the 586 BC, the rabbis vaporized 165 years of history. This was in order to support a non-messianic interpretation of Daniel 9's 70 weeks and/or to avoid finding a fulfillment of those 70 weeks in the ministry of Christ. But as Wacholder admits:

To be sure, the fact that other parts of Rabbi Jose's statement reflect midrashic chronography does not exclude the possibility that his dating of the Shemitah nearest to the destruction of the Herodian Temple in 68/68 [sic] is necessarily inaccurate. It may just be the historical grain upon which Rabbi Jose built the rest of his hermeneutics of Daniel 9:24. (p. 174)

So Rabbi Jose may have been building his faulty chronology on the "historical grain" of an actual synchronism between the sabbatical cycle and the destruction of the second temple. And this is all the more plausible given the fact that Ab 9, when reckoned based on the visibility of the crescent, could have fallen on a Sunday in 421 BC (Aug. 22) and 70 AD (Aug. 5), in agreement with Rabbi Jose.

Is August 22 too late for Ab 9? At some point the Jewish calendar ceased to be reckoned by the visibility of the crescent and started commencing the year earlier in the seasons. Based on the Jewish papyri found at Elephantine, Egypt, Ab commenced on August 20 in 465 BC and August 19 in 416 BC (Horn and Wood, The Chronology of Ezra 7, pp. 157, 159). Thus a date of August 22 for Ab 9 in 421 BC is quite realistic.

Wacholder wants to see the destruction of the temple fall in a sabbatical year rather than in a post-sabbatical year. [table] Four lines of evidence he uses to support such a conclusion are: 1) a statement by Maimonides, 2) Abodah Zarah 9b, 3) Arakhin 12a-13b, and 4) a passage from Josephus.

1. The Statement by Maimonides. As given by Wacholder, Maimonides wrote:

You must say that the year when the second temple was last destroyed, whose beginning commenced in Tishri, about two months after the destruction (since the computation of Shemitot and Jubilees begins in Tishri), that year was a Shemitah. (n. 79, p. 174)

In actuality, Maimonides did not say "that year was a Shemitah." He said that year "was the year following a Shemitah." Wacholder must have realized this, for his argument depends on the correct reading.

So this statement appears to say that Tishri 70 AD began a post-sabbatical year, the first year of the Destruction Era. This makes the actual destruction on the preceding Ab 9 fall in a sabbatical year, an idea that apparently harmonizes with Wacholder.

Apparently? Yes, apparently. It must be pointed out that in the same passage, Maimonides synchronizes a post-sabbatical year with 4936 AM (Era of Creation), 1487 SE (Seleucid Era), and 1107 DE (Era of Destruction of the second temple) (Sabbatical and Jubilee Years, ch. 10, par. 4, 6). Thus 1 DE would coincide with 381 SE and 3830 AM. In order to make these dates fit, we must either assume an error in Maimonides's text or in his chronology.

If Tishri 70/Elul 71 was 1 DE and 381 SE, then 1 SE would begin in the fall of 311 BC, more than a year after Seleucus entered Babylon. This seems a bit late. It makes more sense to use a Seleucid Era for Maimonides that commences in the fall of 312 than one that commences in the fall of 311. This requires us to commence the era of Destruction with the year of the actual destruction rather than with the following Tishri 1.

Since Maimonides identified 1107 DE as a post-sabbatical year, we know that 1 DE was also a post-sabbatical year. This is because the difference between the two, 1106, is evenly divisible by 7. Since 69/70 is a post-sabbatical year according to Zuckermann, this is additional support of his position. [table]

Perhaps we should correct Maimonides's chronology instead of his text. Many rabbinical writers have mistakenly placed the destruction of the temple in the summer of 68 or 69 AD. Suppose Maimonides placed the destruction in 69 AD, as has been suggested by some (Edgar Frank, Talmudic and Rabbinical Chronology (1956), pp. 57, 71). The dates in the above table would then look precisely the same. If we commence the Destruction Era with the Tishri after the actual destruction, and if that destruction is assumed to have occurred in August of 69, then 1 DE begins on Tishri 1, 69 AD. If on the other hand we commence the Destruction Era with the Tishri 1 preceding the actual destruction, and if that destruction was in August of 70 AD, then we still are starting 1 DE with Tishri 1, 69 AD.

Therefore, whether we assume an error in Maimonides's chronology or in his text, we still arrive at a sabbatical date that supports Zuckermann. If the temple was destroyed in a sabbatical year, and if the temple is assumed to have been destroyed in 69 AD, then that still makes 68/69 AD a sabbatical year.

And actually, we have to end up supporting Zuckermann's dates in one of these two ways because of the AM date that Maimonides gives.

AM dating starts with the first conjunction of the new moon of the first year of creation on Tishri 1. When is the first year of creation? Maimonides used a particular form of AM dating that considered the first year to commence with a hypothetical Tishri 1 almost a year preceding the actual Creation week (Sanctification of the New Moon, ch. 6, sect. 8; Frank, pp. 14-19). The sixth day of creation week, the day of Adam's "birth," fell on a second Tishri 1, the first day of a new year. This is why Maimonides says that the "birth" of Adam took place during "the second year after creation," for the second year had just begun (Sabbatical and Jubilee Years; ch. 10, sect. 2).

Each year contains twelve lunar months, plus a little extra. This extra is a fixed amount and is used in Maimonides's calculations for the beginning of each new year. Every year after the hypothetical first conjunction, that fixed amount is added to determine on what day and time the next Tishri 1 will be, a day and time that coincides with observable astronomical phenomena.

All this being so, 4936 AM becomes a fixed date of Tishri 1175/ Elul 1176 AD. Since Maimonides said that this was a post-sabbatical year, we have solid evidence that a scholarly twelfth-century rabbi believed that Zuckermann's sabbatical dates were correct.

2. Abodah Zarah 9b. In Abodah Zarah 9b, we have a formula given for calculating the year of any sabbatical cycle based on the Destruction Era. The formula is to add one year, divide by seven, and the remainder is the year of the sabbatical cycle. If there is no remainder, then that year happens to be a sabbatical year.

Add one year to what? This has been interpreted in a number of ways. Suppose we add one year to the year number of the Destruction Era, using a Destruction Era that commences with Tishri 70 AD. We use this example since Wacholder's citation of Maimonides (uncorrected textually or chronologically) indicates that this is when the Destruction Era began. So if 70/71 AD = 1 DE, then we add 1 to 1 DE to get 2. Since dividing 2 by 7 leaves a remainder of 2, this makes 70/71 the second year of a sabbatical cycle. This date agrees with Zuckermann's sabbatical dates. [table]

The formula does not work correctly when trying to verify that 1107 DE was a post-sabbatical year. 1107 + 1 = 1108. 1108 divided by 7 leaves a remainder of 2. Thus the formula in Abodah Zarah cannot work with Destruction Era dates if Maimonides was correct in labeling 1107 DE as a post-sabbatical year. We are then left with the possibility that Edgar Frank gives, that of adding one year to the difference between the year of the actual destruction, and the year to be calculated. Thus,

This interpretation makes 1107 DE a post-sabbatical year, which is what Maimonides said it was. And it also makes 69/70 a post-sabbatical year in agreement with Zuckermann when beginning the Destruction Era with Tishri 1, 69 AD:

3. Arakhin 12a-13b. It is true that this portion of the Talmud argues that the destruction of Solomon's temple took place during a sabbatical rather than a post-sabbatical year, and that this is then used to show that the destruction of the second temple likewise took place during a sabbatical year. Yet several crucial points should be noted.

First of all, it would appear from the Arakhin's discussion that the view that the destructions took place during post-sabbatical years was a common one.

Second, Arakhin's main argument is based on a misinterpretation of Ezekiel 40:1. It was felt that this verse synchronized the fourteenth year after the destruction of Solomon's temple and the twenty-fifth year of the Babylonian captivity with a jubilee year. Yet this was all based on the rabbinical revision of ancient chronology that vaporizes 165 years of history in an attempt to avoid arriving at a messianic or Christian interpretation of Daniel 9's 70 weeks. 

The fact of the matter is that, using a legitimate chronology, there is no way to have the fourteenth year after the destruction in 586 BC be a jubilee year using either Zuckermann's or Wacholder's sabbatical dates. [table] According to Wacholder, the summer of 586 was the third year of a sabbatical cycle. According to Zuckermann it was the fourth year. Since the relationship to the sabbatical cycle of the date of the second destruction is derived from this faulty reasoning, it is impossible to use it to prove Zuckermann's sabbatical dates to be in error.

4. The passage from Josephus. Wacholder writes:

Finally, a passage in Josephus implies that the year 68/69 was not Sabbatical. According to B.J. 4:529-37, "Simon the son of Gioras, the leader of the Zealots, invaded Idumaea in the winter of 68/69 and gained abundant booty and laid hands on vast supplies of corn." This clearly indicates that it was not a part of a Sabbatical season, for surely the Idumeans by now appear to have been following the traditions of Jewish law. (p. 176)

What Josephus wrote was this:

Thus did Simon unexpectedly march into Idumea, without bloodshed, and made a sudden attack upon the city Hebron, and took it; wherein he got possession of a great deal of prey, and plundered it of a vast quantity of fruit. (Wars, bk. 4, ch. 9, sect. 7)

According to Zuckermann, the sabbatical year would have begun just prior to the winter of 68/69. [table] Since we aren't talking about a lengthy time period after the non-existent harvest of a sabbatical year, then surely there would be large stores of grain in existence. Indeed, we might expect the stores of grain to be larger during the winter of a sabbatical year than during the winter of a non-sabbatical year, for the supply of grain must last until the harvest of the post-sabbatical year.

Thus the fact that Simon found large stores of grain in Idumea, if it suggests anything at all, really suggests that the winter of 68/69 was during a sabbatical year. And that supports Zuckermann's sabbatical dates.

Some Calendar History

At this point we suggest what appears to be part of the history of the Jewish calendar. Ta'anith 29a put the destructions of both temples on Sundays corresponding with Ab 9, and other Talmudic statements put these two dates 490 years apart. If we begin the months with the observation of the new crescent moon, then astronomically speaking, Sunday, August 5, 70 AD, and Sunday, August 22, 421 BC were likely Ab 9.

At some point in the first millennium of the Christian Era, rabbinical Judaism switched to using calculations instead of the sighting of the new crescent. They also switched to dating the destruction of Jerusalem in 69 AD instead of 70, and many rabbis eventually dated it in 68 AD.

If we take today's Jewish calendar and extend it back in time, we find that Sunday, July 16, 69 AD, and Sunday, August 3, 422 BC, are identified as being Ab 9. It thus appears that the rabbis started off with the correct dating of the second destruction, and then for some reason pushed it back one year, synchronizing their calendar calculations to fit this change.

It does not appear that in switching to a calculated new moon and a 69 AD destruction date that they changed their understanding of the sabbatical cycles. When they pushed the second destruction back a year, they appear to have likewise switched from identifying the second destruction with a post-sabbatical to identifying it with a sabbatical year. [return]

"9. Renting Land from Bar Kosba" (pp. 176-179)

Wacholder states that the rental period is therefore "five years, six months, and ten days" (p. 179). Actually, it is seven months instead of six, since Shevat is the eleventh month and Tishri is the seventh.

The fact that the Bar Kochba revolt began in the spring of 132 AD is further evidence for Zuckermann's sabbatical dates, based on Wacholder's hypothesis in his paper "Chronomessianism." His hypothesis is that the Messiah was expected to redeem Israel during the Passover season, and during the sabbatical year. Therefore he seeks to date the commencement of messianic movements accordingly. In this context his paper discusses a) the beginning of John the Baptist's ministry; b) the birth of Christ; c) the messianic Jewish prophet of Egypt (Acts 21:38); and d) the revolt of the messianic Bar Kochba.

Regarding the Egyptian prophet, he dates the commencement of his messianic movement in Nisan 56 AD, after the murder of the high priest in early 55 AD. 55/56 is a sabbatical year according to his calculations. Yet there presumably is nothing that would prevent a dating of this messianic movement in Nisan 55, a date in harmony with Zuckermann's sabbatical dates.

Regarding Bar Kochba's movement, Wacholder departs from his hypothesis by commencing it in Nisan of 132, a year he identifies as a pre-sabbatical year. Yet to be consistent, both the Egyptian and Bar Kochba should commence their movements in sabbatical years. Yet if we identify 132 as being a sabbatical, we once again find support for Zuckermann's dates. [table] [return]

"10. Three Fourth and Fifth Century Tombstone Inscriptions in Sodom" (pp. 180-182)

Edgar Frank identifies five different ways of calculating DE, the Destruction Era (p. 41). To these we append the two proposed by Wacholder:

Frank's second method, as we have already seen, likely should start 1 DE on Tishri 1 of 69, not Ab 9 of 70, but this will have no effect on the following discussion.

The first tombstone cited is dated by Wacholder in Heshvan 1, 364 DE, a post-sabbatical year. If we use Frank's first method or either of Wacholder's methods, we end up with Heshvan of 364 DE being in 433/434 AD, a post-sabbatical year according to Zuckermann. [table] None of the proposed methods allow Heshvan 364 DE to fall in 434/435 AD, and thus none of the methods allow Wacholder's sabbatical dates to be supported by this tombstone.

The third tombstone cited is dated on Tuesday, Elul 11, 435 DE, a sabbatical year. Ignoring the day of the week for a moment, we find that Frank's first two methods support Wacholder (503/504 AD), his second two support Zuckermann (502/503 AD), and Wacholder's second method supports himself. [table]

Using today's rabbinical calendar, we find that Elul 11 in 503 AD was on a Wednesday (Aug. 20), and in 504 AD on a Monday (Sept. 6).

Frank mentions that some have concluded from this tombstone that the calendar rule Lo ADU Rosh had not yet been adopted in 503 AD. This rule calls for the postponement of the beginning of the year by one day if Tishri 1 were to fall on a Sunday, Wednesday, or Friday. If we remove the effect of this rule upon these dates, then Elul 11 in 503 AD falls on a Tuesday (Aug. 19). In 504 AD it would still likely fall on a Monday (not Sunday) since probably the preceding Kislev would then have had thirty days instead of twenty-nine.

Frank proposes another method to still have Elul 11 fall on a Tuesday in 503 AD, by having Elul have 30 days instead of 29. This essentially requires both Kislev and Heshvan to have only twenty-nine days as well as Lo ADU Rosh to be applied.

At any rate, this third tombstone, unemended, allows an Elul 11 in 502/503 AD to fall on a Tuesday. That being so, it supports Zuckermann's sabbatical dates, since that year was a sabbatical year according to him. Since Elul 11 could not have fallen on a Tuesday in 503/504, this tombstone can't be used to support Wacholder's sabbatical dates.

Wacholder postulates an emendation of this tombstone that would date it in the second year of a sabbatical cycle. This does not help his case. The only emendation that would help is one that dates it in the sixth year of a sabbatical cycle. This would allow it to synchronize with his pre-sabbatical year of 502/503 AD. [table] Unemended in this way, it can only support Zuckermann. And if Frank is correct that there are two tombstones with this date (p. 38), then it is unlikely that the engraver would make the same error twice.

The second tombstone cited by Wacholder is dated in the third year of a sabbatical cycle, in the month of Shevat, 300 DE. Wacholder proposes emending the date to read 303 DE by adding the word for 3. Yet 303 DE is the third year of a sabbatical cycle only when using Zuckermann's sabbatical dates. If one uses Frank's first method or either of Wacholder's methods of calculating DE, then 303 DE becomes 372/373 AD, the third year of a sabbatical cycle according to Zuckermann. [table]

Wacholder says that Umberto Cassuto has proposed the addition of a single letter, a conjunction, in order to make the tombstone read 346 DE, an idea he rejects. Yet this certainly seems plausible, especially since it doesn't require adding words to the text. (It would, though, require using letters to represent the number forty-six rather than words.) Depending on the method used to calculate DE, either Zuckermann or Wacholder could be supported using this emendation. [table]

[return]