This page uses Javascript.

TorahTimes Messianic Ministry

Teaching Truths about Torah, Time, and Messiah

The Inception of the Priestly Divisions

Abstract: Having demonstrated a solution for the division of Abijah, in relation to the birth of John the Baptist, and leading up to Yeshua’s birth on Tishri 1, 2 BC (August 31/Sept 1), I will present additional evidence here which confirms the arrangement of the priestly divisions.

Scholars have proposed several systems of calculating priestly divisions in the hope of correctly calculating the time of the birth of Messiah. I will defend two points. Firstly, that it was well known at the time Rabbi Yose Ben Ḥalaphta composed Seder Olam, sometime between the end of the Bar Kochba Revolt and AD 150, which division of priests would (or should) have been on duty when the Second Temple was destroyed on 9-10 Av, AD 70. Secondly, that the rotation of priests run continuously back to their inception point. I will discuss the other systems, and I will show that a number of them give the same results in 2 BC that continuous rotations do, thought not so elegantly as the solution I have given. I will start with the last point first.

I will lay out four systems, and a nomenclature for this paper. For continuous rotations working backward from the destruction of the Temple, CRT (Continuous Rotations up to the Temple Destruction). For the system promoted by Roger Beckwith, which renews the first division to be on duty during the new moon of Tishri 1 every year: T1A (Tishri 1 Annual). For the system assumed to begin on Nisan 1, N1A. For a variation of N1A which skips a week during each of the three pilgrim feasts N1AS. Below is a comparison of the 4 systems for 4-2 BC.

To compare these on the same basis, I assume the shortest time from the first annunciation to the birth of Messiah, which is 14 months, and the longest time which is 16.5 months. The short time computes both terms from conception at 38 weeks each, and matches the 6th month of Elizabeth with the 1st month of Miryam. The long term, computes 40 weeks from each annunciation, and makes the 1st month of Miyram follow the 6th month of Elizabeth, and allows the delivery of Messiah to be 2 weeks overdue. The difference between the short and long terms is 2.5 months. Since each division serves twice in a year, there are two windows of 2.5 months each year, or 5 months, for which each method will work! Therefore, the chances of a wrong method working by accident for a randomly selected birth date are 5/12 = 42%. On the other hand, the chances of the short term method accidentally working out to the exact date, and both annunciations falling on a new moon day, and both births on a new moon day, etc. for CRT are so small as to be meaningless to compute.

Each method has two calculations, for the two services of Abijah, one for the first service and one for the second service. One method has three calculations here since the lunar year is shorter than the solar year. (I include the Julian day number of each date on the first line of each entry. The proleptic Julian Calendar dates appear on the next line. For the corresponding Roman Calendar dates, add 1 day. The last line gives the lunar month with a Roman numeral, and the day of the month, calculated from the sighted new moon. Column D gives the ideal least time birth date, and Col. E stretches the time line as far as possible, i.e. 2.5 months. This chart only covers dates in 2 BC which meet the requirements of Luke 3:1 and 3:23.)

                       A           B            C            D             E          
                       Yōḥanan     Yĕshūa       Yōḥanan      Yĕshūa        Farthest   
                       Conception1 Conception2  birth1       birth2        Limit      
Method   Service       1st off     A+5mnths+1   A+266        B+266-1       +75        
CRT      Abijah-1st    1720354     1720499.5    1720620      1720766.5     1720841.5  
                       Jan 27,3 BC Jun 22,3 BC  Oct 20,3 BC  Mar 16,2 BC   May 30,2 BC
                       XI.9        IV.9         VIII.9       I.9                      
CRT      Abijah-2nd    1720522     1720670.5    1720788      1720935.5     1721010.5  
                       Jul 14,3 BC Dec 10,3 BC  Apr 06,2 BC  Sep 01,2 BC   Nov 15,2 BC
                       V.I         X.1          II.1         VII.1                    
T1A      Abijah-1st    1720284     1720432.5    1720550      1720697.5     1720772.5  
                       Nov 18,4 BC Apr 16,3 BC  Aug 11,3 BC  Jan 06,2 BC   Mar 22,2 BC
                       VIII.27     I.30         V.29         X.28                     
T1A      Abijah-2nd    1720452     1720600.5    1720718      1720865.5     1720940.5  
                       May 05,3 BC Oct 01,3 BC  Jan 26,2 BC  Jun 23,2 BC   Sep 06,2 BC
                       II.19       VII.20       XI.19        IV.20                    
N1A      Abijah-1st    1720270     1720418.5    1720536      1720683.5     1720758.5  
                       Nov 04,4 BC Apr 02,3 BC  Jul 28,3 BC  Dec 23,3 BC   Mar 08,2 BC
                       VIII.13     I.16         V.15         X.14                     
N1A      Abijah-2nd    1720459     1720607.5    1720725      1720872.5     1720947.5  
                       May 12,3 BC Oct 08,3 BC  Feb 02,2 BC  Jun 30,2 BC   Sep 13,2 BC
                       II.26       VII.27       XI.26        IV.27                    
N1A      Abijah-1st    1720627     1720775.5    1720893      1721040.5     1721115.5  
                       Oct 27,3 BC Mar 25,2 BC  Jul 20,2 BC  Dec 15,2 BC   Feb 28,1 BC
                       VIII.16     I.18         IV.29        X.17                     
N1AS     Abijah-1st    1720291     1720439.5    1720557      1720704.5     1720779.5  
                       Nov 25,4 BC Apr 23,3 BC  Aug 18,3 BC  Jan 13,2 BC   Mar 29,2 BC
                       IX.5        II.7         VI.6         X.6                      
N1AS     Abijah-2nd    1720466     1720614.5    1720732      1720879.5     1720954.5  
                       May 19,3 BC Oct 15,3 BC  Feb 09,2 BC  Jul 07,2 BC   Sep 20,2 BC
                       III.4       VIII.4       XII.3        V.4                      

These four methods leave two gaps, 1. May 31, 2 BC to June 23, and 2. Sep 20, 2 BC to Dec 15, 2 BC. The missed days are 107, or 30% of the year. So for any random date picked, and then intelligently choosing the right method, the chances of confirming it are 70%. Columns D-E gives the span of dates for each instance of each method. Four of 9 instances cover the Sep. 1, 2 BC date (Tishri 1). And all of the methods cover that date: CRT, T1A, N1A, and N1AS. It is therefore a useless exercise to try to confirm a date without first eliminating the methods which are not actually historical or Biblical.

I was a long time believer in T1A, and then my eyes were opened to some major problems with that system, and to the merits of CRT when I sought to justify on the basis of Scripture what Roger Beckwith and Jack Finegan claimed for T1A. In the process of doing so, I was reading a Rabbinical paper on the subject which cited Deuteronomy 18:6-8:

And when the Levite comes from one of your gates from all of Yisra’ēl, where he sojourns, and he has come with all the desire of his soul unto the place where Yăhwēh will choose, then he will have served in the name of Yăhwēh his Almĭghty like all his brothers, the Leυi̱yim that are standing there before the face of Yăhwēh. Portion like portion he shall eat, apart from his sales concerning fathers.

Deuteronomy 18:6-8

The point is that every Levitical Priest must be given equal access to the benefits of serving at the Sanctuary, and all those benefits must be divided up equally. If someone says to a priest that he can serve two weeks a year, and to other priests that they may serve three weeks a year, then that directive is not legal. For it causes some priests to eat unlike their brothers. And the importance of a job at the Sanctuary is underscored by the details of the curse on Eli’s family (1 Samuel 2:36).

I am very much more suspicious of arguments based on improbable coincidence than I used to be. Beckwith observed that when the first division is worked backwards from 9 AV, AD 70 to the previous Tishri, that the first division just happens to align with the week for the new moon of Tishri 1. Well, it does, and I have no argument that it matches there. But Beckwith’s conclusion (based on the unstated assumption of only a 1/24 chance of accidental alignment, and 23/24 probability that the alignment was due to an annual reset on Tishri 1) was that the divisions reset annually every Tishri 1, hence T1A! But I have worked with calendars enough and astronomy enough to know that it is cyclical by nature, and that what appear to be coincidences too amazing to be called coincidences happen all the time. Exploiting the coincidences and making them seem like intelligent determination is the craft of astrologers. Additionally, it turns out that the first week of Tishri was not the starting point in either the first or Second Temple, so the supposition that a Tishri 1 alignment should mean anything in the first place is incorrect.

The problem with accepting Beckwith’s conclusion is that now I have a number of Scriptures that when put together contradict it, and as a bonus, I have discovered enough new coincidences that agree with those Scriptures to truly think that Beckwith really was taken in by a coincidence. And this he has passed on. But it is really the Scripture that has caused me to switch to CRT. Thankfully, I did not have to reinvent it since it was already correctly calculated by Thomas Lewin in his Fasti Sacri. Lewin’s needs some correction, updating, and expansion.

T1A has the property of running 24 priestly divisions twice every year, and divisions 1-2 get to repeat a third time in every regular year, and often divisions 1-3. In a leap year with an Adar II, divisions 1-7 can repeat three times. The case with T1A then is often that divisions 1-2 or 3 in the 6th month are immediately followed AGAIN by divisions 1-3 in the seventh month. That is not equitable treatment of the priests! What has just been said about T1A also applies to N1A and N1AS. N1AS has the additional problem that the three skipped weeks put with the 48 weeks of two cycles of 24 add up to 51 weeks, and this may be longer than a lunar year. The shortest lunar year is 353 days, or 50 weeks and 3 days. In some cases the 24th division will only serve once in a year under N1AS. It appears, then, that the only method of rotation that can stand up to Deuteronomy 18:6-8 is CRT.

Finegan stated,

It is difficult to see, however, how such a system could have been carried out without confusion about the calendar, or how it could have been reestablished accurately after such a time as when the temple was desecrated by Antiochus Epiphanes on Dec 15, 167 B.C. (1 Macc 1:54), and lay desolate for three years.

§242, Handbook of Biblical Chronology, rev. ed., pg. 133

I don’t think Finegan thought this criticism through too well, which is taken straight from Beckwith, whose arguments I deal with later on, because the seventh day Sabbath is a continuously rotating system of seven days, without any connection to astronomical timekeepers of the calendar. The Sabbath can only be determined by counting back to a previous Sabbath. It was the job of the priests to promote and maintain Sabbath observance in Israel. And the priestly rotations are inextricably synchronized with the Sabbath. The divisions begin and end on every Sabbath at noon. There is no confusion at all, except for one who fails to count. Every priest is assigned to a division with a name and a number 1-24. And all one needs to be able to do is to count from 1 to 24. The same division will be on duty every 24 weeks. Keeping Sabbath, where there are no external reminders of the weekdays, requires one to count days from one to seven. Every division knew which division it should follow in the service. The many priests in every division were better than a peg board or tally marks. They kept their place in time. Since there were twenty four divisions, then there were twenty four groups of priests keeping 24 separate counts of 24 weeks to their next service. And apt comparison would be that there are more backup systems to prevent a failure in the counting of the divisions than there are backup computers in the Space Shuttle!

The three years or so that Finegan speaks about the temple being deprived of sacrifice and offering is not logically an argument against CRS. It is only an argument against the long term continuity of CRS, and that by itself is incredibly weak. For a space of three years away from the temple, it is not hard to continue counting the weeks, especially if each division is in the habit of counting weeks in the first place. Keeping the right times was the foundation of sanctifying holy time, an essential part of godliness. The temple service was disrupted also in 63 BC when Pompey invaded it, and may have caused a division to miss its week, or a good part of one. What do priests do in such a case? They keep counting their place in the order, just as Egypt managed to keep counting its continuous cycle of months through all their wars, invasions, and desolations.

We should not be concerned with the length of 24 weeks. It is no harder to count a continuous rotation of 24 weeks than a continuous rotation of seven days. I will show later how a forgotten count could be recovered for the 24 weeks.

According to Ezra 6:18, the divisions were put back in order when the new Temple was finished:

And they had made stand the priests in their divisions and the Levites in their divisions according to the service of the Almĭghty, which is in Yerūshalem according to what is written in the book of Mōsheh.

Ezra 6:18

The Second Temple was finished up to the 3rd of Adar in the 6th year of King Darius (Ezra 6:15). Adar in the 6th year of Darius was in 515 BC. The 3 day of Adar was March. 12, 515 BC, a Sabbath. Now like with Solomon’s Temple, they did not start the services immediately. For this year had an Adar II. This time was spent putting the equipment into the temple, and sanctifying the priests, and then the divisions picked up with the week of Nisan 1. When the first Temple had been built, it was finished in month VIII of the 11th year of Solomon, but the priests were not sanctified for it until the first week of the following Tishri, the start of the 12th year of Solomon.

It appears to me that the divisions, even when they were disrupted, were counted in strict sequence, so long as there was a community of priests in the land, and when the disruption was over, the rotation simply continued with the division whose week it was. On the other hand, if the whole people were exiled, then the counting might be suspended, and the divisions would pick up where they left off in the old Temple when the new Temple was built. In other words, the priests could keep to their counting if there were a community of them, and the disruption was relatively short. But if it was a generational disruption, then it would simply be remembered which division was cut off from the old temple, and then that division would head up the rotations in the new Temple.

It would be helpful to know then, which divisions would be on duty in the spring of 515 BC counting back from 9 AV, AD 70, (the synchronized date of the 1st division at the destruction of the Second Temple: Aug 5). And then if we can find the beginning point of the rotations in Solomon’s Temple, and then count forward to its destruction date, then we could find the division whose week it was when the Babylonians burned the Temple. What if we were to find, then, this same division serving first in the spring of 515 BC after the New Temple were opened, or the one following it?

In the Scroll of Bibical Chronology, I show that the First Temple was finished in the 11th year of Solomon and in month VIII. The Temple was opened with a feast of dedication and a week of sanctifying all the priests on Tishri 1 in the year following. This was the 12th year of Solomon. The date for this is 1012 BC. The Law required priests to be consecrated for seven days (Exodus 29:30, 35, 37; Leviticus 8:33), so they were at the dedication of Solomon’s Temple (2 Chronicles 5:3, 11; 1 Kings 8:2, 8:65-66). They assembled in the seventh month to bring the ark to the new Temple for the Yōm Teruah feast. Solomon put on a feast all that week while the altar and the priests were sanctified. 2 Chronicles 5:11 says:

Then it was in the going out of the priests from the holy place that all the priests being found had made themselves holy without regard to divisions.

This appears to mean that the rotation started the following week with the first division. For the text first records that Solomon sent the people away on the 8th day, during the second week of Tishri. Then they returned for the feast of Tabernacles, and then on the 8th day after that, the 22nd of Tishri, the people were not sent away, but they held a solemn assembly on the last great day. (1 Kings 8:65-66 and 2 Chronicles 7:8-10.) I believe all this to mean that the rotations began with the Sabbath following Tishri 1. This month in question appears thus:

Month: VII ETHANIM, 1012 BC 3128 A.M. Sab. Cyc: 7. Jub. Cyc: 42 Cycle No: 63
Q1: 0.677 A Q2: -0.320 F LG:  61m W: 1.125' AL: 22.6 AV: 12.3               
New Moon calculated for longitude: 35.17 and latitude 31.77                 
Location of calculations: Jerusalem Author: Daniel Gregg                    
        I        II        III       IV         V        VI        VII      
 ↑   │   1     │   2     │   3     │   4     │   5     │   6     │   7     │
NM   │ Oct 7   │         │         │         │         │         │         │
     │1352070  │     Seven Days Sanctification of All the Priests│         │
     │         │         │  Without Regard to their Divisions    │         │
     │   8     │   9     │  10     │  11     │  12     │  13     │  14     │
     │ Oct 14  │         │Y. Kippur│         │         │         │         │
     │1352077  │         │         │         │         │         │         │
     │ First Division: Jehoiarib, Yehoiari̱v̱  │         │         │         │
     │  15     │  16     │  17     │  18     │  19     │  20     │  21     │
     │Sukkot   │         │         │         │         │         │         │
     │         │         │         │         │         │         │         │
     │         │         │         │         │         │         │         │
     │  22     │  23     │  24     │  25     │  26     │  27     │  28     │
     │8th Day  │         │         │         │         │         │         │
     │         │         │         │         │         │         │         │
     │         │         │         │         │         │         │         │
     │  29     │  30 ↑   │                                                  
     │         │         │                                                  
     │         │         │                                                  
     │         │         │                                                  

The Julian day number stands in the position of the 8th day of the month. This is when Solomon sent all the officials away who took part in this special feast for the first week of Tishri. Only the first division stayed on for this week. Now as I elsewhere developed a formula for calculating the division number from the Julian date for the Second Temple, I have also done the same for the First Temple:

Notes: Valid 1012 BC to 587 BC.                                                                         
Given:  Julian_day_number                    (i.e. JD)                                                  
        ADD 1                                (Eliminates decimal results from the next step)            
        Divide by 7                          (Reduces JD, to consecutive week number)                   
        SUBTRACT 2                           (offsets the week so that MOD gives 0 for the 1st division)
        Divide by 24 and take the remainder  (MOD 24, returns 0-23, one less than the division number)  
        ADD 1                                (eliminate 0 based counting of the modulus)                
        Drop any decimal                     (hint: avoid decimals by choosing a JD for a Sunday.)      
Result: Priestly_division_number                                                                        
How the function looks in a MS Exel spreadsheet: =INT(MOD((JD+1)/7-2,24)+1)                             
JD = references a cell with a JD.                                                                       

The Julian date for the 8th day of the month is 1352077.

JD = 1352077                              
Formula: ((((JD + 1) / 7) - 2) mod 24) + 1
Add 1:          1352078                   
Divide by 7:    193154                    
Subtract 2:     193152                    
Mod 24:         0                         
Add 1:          1                         

The priestly division number is 1. So on the Sabbath following the new moon of Tishri 1, the first division was on regular duty.

Now I would like to work a second example. What if we count the divisions backwards proleptically to the week for the new moon of Nisan that year? Here is the calendar for Nisan 1012 BC:

Month: I AVIV, 1012 BC 3128 A.M. Sab. Cyc: 6. Jub. Cyc: 41 Cycle No: 63     
Q1: 1.096 A Q2: -0.726 G LG:  90m W: 0.817' AL: 18.5 AV: 18.1               
New Moon calculated for longitude: 35.17 and latitude 31.77                 
Location of calculations: Jerusalem Author: Daniel Gregg                    
        I        II        III       IV         V        VI        VII      
     Division 22 - Gamul (proleptic)     ↑   │   1     │   2     │   3     │
     AVIV/NISAN                          NM  │New Moon │         │         │
      1351888   1351889   1351890   1351891  │ APR 12  │         │         │
     │   4     │   5     │   6     │   7     │   8     │   9     │  10     │
     │         │         │         │         │         │         │         │
     │         │         │         │         │         │         │         │
     │  11     │  12     │  13     │  14     │  15     │  16     │   17    │
     │         │         │         │Passover │Passover │ Sheaf   │         │
     │         │         │         │         │         │         │         │
     │  18     │  19     │  20     │  21     │  22     │  23     │  24     │
     │         │         │         │7thULB   │         │         │         │
     │         │         │         │         │         │         │         │
     │  25     │  26     │  27     │ 28      │  29     │                    
     │         │         │         │         │ MAY 10  │                    
     │         │         │         │         │         │                    

Again I have put up the Julian day number for the week containing the new moon of Nisan. I will use the first day of that week: 1351888. Let us work the math, and find the proleptic division number. Proleptic means before it was instituted. I have a reason for doing that that I will disclose in a while.

JD = 1351888                              
Formula: ((((JD + 1) / 7) - 2) mod 24) + 1
Add 1:          1351889                   
Divide by 7:    193127                    
Subtract 2:     193125                    
Mod 24:         21                        
Add 1:          22                        

The division number is 22. That is the division number of Gamul (1 Chronicles 24:17). The relevance of this point is that the Qumran community considered Gamul (division 22) to be the starting point of their cycles in the spring. And they also assumed that Gamul was on duty (proleptically) the week of creation. The point is that these assumptions must be based on some general knowledge of the divisions. It means that the first division did not start out on the first week of Tishri as claimed by Lutheran chronologist Roger Beckwith.

The Second Temple was burned in 587 BC on the 10th of Av (Jeremiah 52:12). Jewish tradition puts the fast on 9 Av. The difference will not concern us. The calendar for the month is here:

Month: V AV, 587 BC  3553 A.M. Sab. Cyc: 4. Jub. Cyc: 25 Cycle No: 72       
Q1: 1.056 A Q2: -0.239 E LG:  88m W: 1.040' AL: 21.6 AV: 16.5               
New Moon calculated for longitude: 35.17 and latitude 31.77                 
Location of calculations: Jerusalem Author: Daniel Gregg                    
        I        II        III       IV         V        VI        VII      
                                                   ↑   │   1     │   2     │
         AV                                       NM   │New Moon │         │
                                              1507222  │ JUL 21  │         │
     │   3     │   4     │   5     │   6     │   7     │   8     │  9   ♦  │
     │         │         │         │         │         │         │         │
     │1507225  │         │         │         │         │         │         │
     │  10     │  11     │  12     │  13     │  14     │  15     │  16     │
     │14th Division      │         │         │         │         │         │
     │1507232  │         │         │         │         │         │         │
     │Jeshebeab│         │         │         │         │         │         │
     │  17     │  18     │  19     │  20     │  21     │  22     │  23     │
     │         │         │         │         │         │         │         │
     │         │         │         │         │         │         │         │
     │  24     │  25     │  26     │  27     │  28     │  29 ↑   │          
     │         │         │         │         │         │         │          
     │         │         │         │         │         │         │          

The week of the destruction was 1507232.

JD = 1507232                              
Formula: ((((JD + 1) / 7) - 2) mod 24) + 1
Add 1:          1507233                   
Divide by 7:    215319                    
Subtract 2:     215317                    
Mod 24:         13                        
Add 1:          14                        

Duty was passed to Jeshebeab, the 14th Division on the 9th of Av at noon, and the Temple was burned on their duty shift. The formula is: [(JD + 1)/7 — 2] MOD 24 + 1.

Using the formula for the Second Temple, let’s see who was on duty in the spring of 515. The Julian date for the week of the new moon of Nisan is 1533412. The Second Temple formula is: = MOD((JD+1)/7+3,24)+1 (MS Excel Notation).

JD = 1533412                              
Formula: ((((JD + 1) / 7) + 3) mod 24) + 1
Add 1:          1533413                   
Divide by 7:    219059                    
Add 3:          219062                    
Mod 24:         14                        
Add 1:          15                        

Counting back from the destruction of the First Temple to its begining leads us to division 15: Bilgah. As a standard math expression: [(JD + 1)/7 + 3] MOD 24 + 1.

Month: I AVIV, 515 BC  3625 A.M. Sab. Cyc: 6. Jub. Cyc: 48 Cycle No: 73     
Q1: 1.131 A Q2: -0.761 G LG:  87m W: 0.936' AL: 19.5 AV: 17.8               
New Moon calculated for longitude: 35.17 and latitude 31.77                 
Location of calculations: Jerusalem Author: Daniel Gregg                    
        I        II        III       IV         V        VI        VII      
      Division Number 15                           ↑   │   1     │   2     │
     AVIV/NISAN                                   NM   │New Moon │         │
     │ Bilgah  │         │         │         │APR 7    │ APR 8   │         │
     │1533412  │1533413  │1533414  │1533415  │1533416  │         │         │
     │         │         │         │         │         │         │         │
     │   3     │   4     │   5     │   6     │   7     │   8     │   9     │
     │         │         │         │         │         │         │         │
     │         │         │         │         │         │         │         │
     │  10     │  11     │  12     │  13     │ 14♦     │  15     │16-1-1   │
     │         │         │         │         │Passover │Passover │ Sheaf   │
     │         │         │         │         │         │         │         │
     │         │         │         │         │         │         │         │
     │17-1-2   │18-1-3   │19-1-4   │20-1-5   │21-1-6   │22-1-7   │23-2-8   │
     │         │         │         │         │7thULB   │         │         │
     │         │         │         │         │         │         │         │
     │         │         │         │         │         │         │         │
     │24-2-9   │25-2-10  │26-2-11  │27-2-12  │28-2-13  │29-2-14  │          
     │         │         │         │         │         │         │          
     │         │         │         │         │         │         │          
     │         │         │         │         │         │         │          

So going back to the beginning of services in the Second Temple leads us to Bilgah, the 15th division, and this follows the 14th Division during which the First Temple was destroyed.

So, after the special service during the first week of Tishri in 1012 BC, regular services could start in Solomon’s Temple, with the first division. Counting up to the destruction of the Temple on 10 Av, 587 BC, we find the 14th Division on duty, or scheduled to be on duty. They were probably brave enough to sacrifice their lives if they were not already dead. Then when the Second Temple is completed we find that counting back from the First Temple, the 15th Division would be on duty for the week of Nisan 1. And we find a statement in Ezra 6:18 that says they put the divisions in order at that time. Was this the way it really was? For the present, it appears so. It may be that the 14th division was given a rotation in the week before Nisan 1 in 515 also. The inferred idea is that they would pick up where they left off in the First Temple.

The above coincidence appears to me to negate Beckwith’s observation that the first division lands on the week of Tishri 1, AD 69, and then to deduce T1A from it. One can just as well conclude CRT from the Biblical data I have given as Beckwith can conclude T1A from a chance alignment. But I should point out one thing. Beckwith’s chance alignment does not really align! The first division should be on duty on the Sabbath after Tishri 1 according to the inception point in the First Temple! (See 1 Chronicles 5:11). The first week was used up sanctifying all the divisions. Finegan mentions this as a feature of the Qumran Calendar. Finegan highlights the difference (cf. §243, §246), stating that Qumran synchronized the first division to the Sabbath after Tishri 1 in the first year of their six year cycle. Then like most early Dead Sea Scholars, Finegan overrates the opposition of Qumran to Jerusalem (cf. §249, Handbook of Biblical Chronology, rev. ed.), and implies that the Qumran rotation was in opposition to that at Jerusalem. This is highly unlikely. Qumran and Jerusalem shared common traditions, just as Catholics and Protestants agree on a great number of items. There is no evidence that the lunar cycle was tracked in 4Q320 as a polemic against the new moon. Rather it was tracked as a schematic means of double dating. For using it an Essene could receive news of the new moon from Jerusalem, or look it up in the current cycle, a kind of Almanac to find the corresponding dates in the Solar Calendar. And since it was tied to the weekly cycle, only a Metonic type cycle accuracy was required. The scattered Essenes could use the double dating and the announcement of the new moon to find their place in their calendar. The recording of details like the last visibility of the old moon, and the first day that the sun sets before the full moon rises, are measures to ensure accuracy, and not polemical devices. More on this below. I will return now briefly for the reason we are doing all of this:

Luke seems to have carefully constructed his chronology based on the rotation of the priestly divisions. He mentions Abijah once (Luke 1:5), but he refers to the division duty twice more in his narrative (Luke 1:8; 1:23). Luke is writing more than 30 years after the events, but he can interview the primary source: the mother of Yeshua. Would Luke have placed a threefold emphasis on the division of Abijah, and then connected it with the annunciations and months if he had any knowledge that the rotation was disrupted? It is highly unlikely. Everything Luke does is calculated to give us a chronology in literary style. And such style is frequently used in Scripture to indicate chronological details. The Scripture in Deuteronomy 18:6-8 supports the CRT system, and leaves us wondering how T1A, NIA, or N1AS can meet the requirement to treat all the priests equally. When this system is worked back from 9 AV, AD 70, it fits perfectly with the Luke 1-2 narrative, and connects exactly with a Tishri 1 birth date. It also suggests that both conceptions, and the birth of Yohanan were also on new moon days. This arrangement of coincidences is much more agreeable to the Scripture than the single coincidence that Roger Beckwith offers in support of T1A, which turns out according to Chronicles to be an incorrect inception point, proving again the banal proverb: assume what you prove, and prove what you assume. The difference here is that the Scriptures match what I am saying, whereas they do not match what Beckwith claimed.


In the first version of this article I had picked Feb. 10 as the Adar 3 date on which the Temple was finished, and did not state that there was an Adar I and and Adar II in 515 BC. The reason I rejected Adar II was that the 3rd day is a Sabbath. Finishing on that day implies work on that day. In the Biblical Calendar, both Adar I and II are in the sixth year of Darius. Is it possible that Adar I and II are not distinguished? The answer is that they never are. The Hebrew Bible does not include a I or II to indicate a first or second Adar. According to the normal Persian/Bablyonian method of inter calculating the year, there would be no Adar II that year. There would only be an Adar (I) starting with a new moon sighting on March 9th. So the question is whether the Scripture means Adar I or Adar II? I ruled out Adar II since the English text said “on the third day.” I should have checked the Hebrew text.

Now normally the Hebrew in back of the word “on” is ב. This what one would assume. However the Aramaic in Ezra 6:15 does not say this. It has the word עַד! This discovery throws a whole new light on the matter. This word may mean “until” or “up to.” So it does not matter that the Sabbath is Adar 3. It is just the endpoint. I am chagrined at this oversight. Not only does this Hebrew word permit sunset on the sixth day of the week to be the terminal point, it implies that it was the terminal point given the two choices. Normally one expects the ב. But the author by putting the עַד is implying that he had a reason for doing so, and that is that he knew there was an Adar I and Adar II and therefore wanted to leave a clue as to which one. The author had a choice of writing “on the second day of Adar” or “up to the third day of Adar.” The first choice gives no clue. The second does because it is not the normal choice and there needs to be a reason for it. The reason is the final work was the day before a Sabbath.

So taking the II Adar we have the following:

Month: XIII ADAR_II, 515 BC  3624 A.M. Sab. Cyc: 6. Jub. Cyc: 48 Cycle No: 73
Q1: 1.791 A Q2: -0.199 D LG: 109m W: 1.336' AL: 23.4 AV: 22.4                
New Moon calculated for longitude: 35.17 and latitude 31.77                  
Location of calculations: Jerusalem Author: Daniel Gregg                     
        I        II        III       IV         V        VI        VII       
                                         ↑   │   1     │   2     │   3     │ 
       ADAR_II                               │New Moon │ FIN     │         │ 
     │         │         │         │ MAR 9   │ MAR 10  │         │         │ 
     │         │         │         │         │         │         │         │ 
     │   4     │   5     │   6     │   7     │   8     │   9     │  10     │ 
     │         │         │         │         │         │         │         │ 
     │  Installation of Equipment  │         │         │         │         │ 
     │         │         │         │         │         │         │         │ 
     │  11     │  12     │  13     │  14     │  15     │  16     │  17     │ 
     │         │         │         │         │         │         │         │ 
     │  ALL DIVISIONS SANCTIFIED SEVEN DAYS ±1 week    │         │         │ 
     │         │         │         │         │         │         │         │ 
     │  18     │  19     │  20     │  21     │  22     │  23     │  24     │ 
     │         │         │         │         │         │         │         │ 
     │  #14?   │         │         │         │         │         │         │ 
     │         │         │         │         │         │         │         │ 
     │  25     │  26     │  27     │  28     │  29 ↑   │                     
     │         │         │         │         │         │                     
     │ DIVISION│ BILGAH  │         │         │         │                     
     │  #15    │         │         │         │         │                     

This means that there are only three possible weeks between the completion date and the service dates of division #15 in which all the divisions could be sanctified for the New Temple. If the 14th division was repeated from 587 BC, then there are only two weeks available.

We can again use the directive that all the priests are to be treated fairly from Deuteronomy 18:8. It does not matter that only four divisions returned from Babylon. The principle of the matter is that the remaining divisions were made up by lot, and that the principle of fairness demands that the divisions be resumed where they left off. The only legal ambiguity is whether the 14th division gets to try again after being cut off from service in 587 BC. My guess is the answer is no, since there were no doubt other shorter disruptions and the division count then proceeded onward regardless. The priests kept their week counting at all times except with a major exile. Evidently here they suspended counting. So my guess is that service passed to the 15th division as the first service in the new Temple.

If we suppose for the sake of argument that the divisions started again with the first, then we can well imagine the complaints of divisions 15, 17-24 re-chosen by lot. Four divisions returned from Babylon: Jedaiah (#2), Harim (#3), Pashur (#5), and Immer (#16). We can suppose that Immer would have complained it was a violation of the Law, and the legal scholars would have to agree. Again if the divisions started with #22, then the skipped divisions #15-21 would have a grievance. We have to consider that division #15, due to the exile, had to wait the longest to see new service in the Temple. It was therefore correctly their turn.

One week has to be used for the sanctification of all the divisions. This means that division #15 can only be moved backward two weeks. Upon considering the possibility of a 3 BC birth of Messiah we can see what effect only being able to move division #15 back two weeks has. Division 15 finished on April 9, 515 BC (1533418). Backing up two weeks brings us to a finish date March 26, 515 BC (1533404). From this I find the proleptic date of the 8th division, which is 7 weeks earlier or 49 days. This comes to 1533355. Now to find recurring dates that the 8th division finishes we go forward whole cycles of 24 weeks. The magic number here is 1112 cycles. So 1533355 + 1112 x 24 x 7 = 1720171 = July 28, 4 BC. In this thought experiment, this date is the date the 8th division finished in 4 BC. Now add one day for the conception of John: July 29, 4 BC (1720172). To this date we add the minimum time to Messiah’s birth. It is 5 months 1 day and then 266 days or 414 days. This is assuming no delays. 1720172 + 414 = 1720586. The Julian day #1720586 is September 16, 3 BC. This is past the Sept. 11, 3 BC date proposed for his birth by about 5 days. Conclusion: divison #15 would have to be moved back 3 weeks to make 3 BC possible. But only two weeks are available.

The actual dates of the 8th service in 4 BC are: Aug 4 to Aug 11. Adding 414 days to the end date brings us to 9/30 3 BC, which is 19 days past the proposed Sept. 11, 3 BC date. It can be easily seen that one must move the divisions back three weeks to make the date possible.

I should review here the basis of the minimum calculation from John’s conception to Messiah’s birth. This is five months, and 1 day to make it the first day of the sixth month that the annunciation was to Miryam. Since the child was born when the days were full, a full term of 266 days is required. Five months is in the rough 29.5 days x 5 = 147.5 days. Adding 266 days to this gives 413.5. Round off to 414. This is exactly 14 months and 1 day. Therefore any proposed birth date for Messiah must follow the 8th division’s ending date by 14 months and 1 day.

The calculation put forth here demonstrate three independent witnesses to the cycle of priestly divisions. The first witness is Luke himself. The second is the citation in Seder Olam that the first division was only duty when the Second Temple was destroyed by the Romans. And the third witness is that the divisions actually do mesh together between Solomon’s Temple and the Second Temple. All three lines of evidence converge on the same answer.