Bible Calendar Part 3

Table of Contents

The barley harvest and the calendar

Image of barley harvest - Israel - public domain
Barley harvesting in Israel

What makes a solution to the wave sheaf problem possible is the varied climate in Palestine. There are three areas where barley was grown. In the Jordan valley (Jericho) it ripens first. Along the coast it usually ripens about 8 to 10 days later. In the hill country it ripens even later, usually 3 or 4 weeks later than Jericho. This is a range of almost 30 days.

While the Second Temple stood (until AD 70) the complex rules used to determine the start of a year were based mainly on the barley harvest. If the barley was not expected to be ripe in 2 of the 3 areas by the time of the wave sheaf offering, then a month was “intercalated” and Abib (Nisan) was delayed. The effect of this rule was that the barley ripening in the coastal area (the second area) would usually be the basis for the decision. If we know when barley ripens in the coastal area, we would also know the average time when years would have started then, compared to the equinox. Most of the barley in an area ripens about the same time, even if not planted at the same time. Late in the 19th century, when the mean equinox was March 20, it was observed that the barley harvest starts about the middle of April around Jericho. This was about 26 days after the equinox.

In the coastal area, since barley ripens 9 days later than Jericho, it would be ripe about 35 days after the equinox. This is the time, then, when the average date for the wave sheaf offering would have come. Since the average date of the wave sheaf offering would actually have been Abib 18, which is 17 days after the first day of the month, then the median first day of Abib would be 17 days earlier, or about 18 days after the equinox.

So if, in the first century, the start of a year had been based entirely on the barley harvest, then the years would have started, on the average, about 18 days after the equinox. If each year were instead started as soon as possible after the equinox, from 1 to 30 days after, then the median first day of the year would be 15 days after the equinox. Only a three day difference. From this we see that starting every year after the equinox is virtually equivalent to starting each year when the barley would be ripe in two of the three areas, the standard used in the first century.

Other seasonal indications were taken into account also. If the lambs were young and small, it could be reason to delay. Too much cold rainy weather in itself would delay the harvest. Barley needs a month of sun and afternoon warmth to fully ripen. Winter in Palestine is not totally cold. Spring arrives in short periods of warmth and sun alternating with cold periods, during late winter. We are dealing here with averages, but it is clear that the usual time of harvest is later than has been assumed, being in April, not in March. The harvest in the hill country starts about 47 to 54 days after the equinox. This would be available close to Jerusalem in a late year, starting as late as 30 days after the equinox.

The requirement for the barley was not that every field in an area be ready for harvest. Only enough fully developed barley must be available for the offering, and this would happen a few days before all the fields in the area are ready for harvest. Remember that the average harvest times spoken of earlier are for the coastal area, and we were considering only the average month. Now we can consider the earliest years.

The harvests start about 9 days earlier in the Jericho area. The earliest barley is usually ready to harvest about 26 days after the equinox. If the month always started as soon as possible after the equinox, then the earliest month, starting one day after the equinox, might need barley for the offering as soon as 15 days after the equinox. This is still about 11 days before the earliest harvest usually starts. But the harvest does not start until all the barley in the area is fully ripe. Some would be fully formed and at least close to ripe many days earlier. So this may not be too early even in the earliest possible year. Does it seem too early? Then consider the calculated calendar now in use.

Under the particular 19-year system used today, the earliest year starts much earlier than this. When that system was begun, sometime prior to AD 358, the design was that the 16th day of the first month must always come after the equinox has occurred. That put the average start of the year on the day of the equinox, by observation. (Observation was still being used then, and the start of the religious year was based on the first month, not the seventh.) The earliest year could start as early as 15 days before the vernal equinox. Such a system would never have been used while the Temple stood. If you think there might have been a problem finding enough fully formed barley in a year that starts just after the equinox, starting the earliest year 16 days earlier would often have made it impossible.

In the first century, religious leaders in Palestine actually used the ripening of barley to decide the starting month for each year. How did they know what the weather would be like during the next two weeks? And how did they know how ripe the barley should be, if the Bible does not explain how ripe it should be? They did not need to claim the Bible as authority for this practice. Their authority came from tradition. At some time after the return from the captivity, this had become the method they relied on. But this method is totally subjective. That is, someone in authority has to make a difficult decision every year based on various observations of conditions, and assumptions about probable weather conditions for the next two weeks.

This method served the purpose of establishing the authority of the priests and Levites in Jerusalem, and keeping the returned exiles tied closely to the religious leaders in that city. They feared that the people would become divided, as they had after the death of Solomon. So they instituted a calendar system that would guarantee the continuation of their own authority. If they let the people know how easy it really is, people might start thinking for themselves, instead of letting the “duly authorized religious authorities” do the thinking for them.

God’s plan, on the other hand, is a system so simple that everyone could know, by just observing the sun and moon, exactly when each year should start, and the barley for the wave sheaf offering would always be ready when needed. It’s guaranteed. It’s the sun that ripens the barley. If we start the year in the right season, when the sun is in the right place, there’s no need to worry about barley.

Should the entire Feast Of Tabernacles be in Autumn?

The traditional calculated calendar drifts forward in the seasons over the centuries. Because 19 lunar years don’t quite match 19 solar years, today the average start of a year is 6 days after the equinox, by molad calculation. That is still 9 days earlier than it would be if every year started after the equinox, as in ancient times. Five of every 19 years still start before the day of the equinox. Before that calendar drifted forward, it was possible for all the fall holy days, even the Eighth Day, to come entirely in summer. This happened many times. It last happened in AD 1348 that all 7 days of the Feast were in summer. Even now most of the Feast of Tabernacles can be in summer.

Some defenders of the traditional calendar have claimed that “Tabernacles must never end before the fall equinox… That Feast must never occur wholly in summer!” (Global CoG.) Or they have said that if the year starts too early, “the entire Feast and the Last Great Day would occur before the autumnal equinox – before autumn actually begins! This is no small error.” (Philadelphia CoG.)

Why do you think they worded it like this? Why would one not expect the whole feast to be in the same season, entirely after the autumnal equinox, as Lev.23:39 (“when ye have gathered in the fruit of the land”) and Ex.34:22 (“at the revolution of the year”) imply? It doesn’t always do that with the current Jewish calendar. In some years, only part of the feast is in autumn now. But if it is “no small error” to have the entire feast in summer, how do they explain the fact that using the same calendar it has happened many times. If you know how to calculate that calendar, check out these years. (Dates are Julian.):

AD 664, when the first of Tishri was Aug. 27, and the “Eighth Day” (the 22nd of Tishri) was September 17. The autumnal equinox that year was not until September 20. This is the worst example, with the Eighth Day ending over 2 days before the end of summer.

AD 911, when the first of Tishri was also Aug. 27, and the Eighth Day Sept. 17. The equinox that year was Sept. 18. This was the last time that even the Eighth Day was in summer, before the “turn of the year” (tekufah). Ex.34:22, KJV margin: “revolution of the year.”

AD 1348, when the first of Tishri was Aug. 25, and the Eighth Day was Sept.15. The equinox that year was on Sept. 14, at about the end of the seventh day of Tabernacles. This is the last time that all but the Eighth Day were in summer. Since then, the calendar has drifted several days forward in the season, because the length of a 19-year cycle (235 lunations) is about 2 hours more than 19 solar years, so today at least some of the feast is always in the fall, as all of it would be if the signs were still used, and every year were started after the spring equinox. It has advanced about 7 and a half days in the season since 358 AD, so that at least the last 3 days of the Feast are in autumn now. But it still has another 4 days to go to get the fall holy days right, which will take another 875 years.

The Hebrew word “tequfah” or “tekufah”, Strong’s #8622, is found in Exodus 34:22, where it refers to the end of a season, the “revolution” of the year, the fall equinox, which marks the end of summer. In 2 Chron. 24:23 it refers to one of the two equinoxes. Which one is not clear. The word seems to mean a turning point or the completion of a circuit. In Psalm 19:6 the word seems to be referring to a single day’s course of the sun across the sky, so the word “tequfah” means a complete period of time. This is what it means in I Sam 1:20, where it refers to the birth of Samuel at the completion of the required nine months. So the “completion” of a year would be when the “tekufah” or equinox has occurred.

Exodus 34:22 says, (KJV) “And thou shalt observe the feast of weeks, of the firstfruits of wheat harvest, and the feast of ingathering at the year’s end.” The margin in the KJV offers “revolution of the year” as an alternate translation for “year’s end.” The word is “tequfah.” So the clear implication is that the whole of the “feast of ingathering” should be after the equinox, and after the harvest has been completed.

Simplicity and chaos: How the traditional calendar changed after the first century

There is a simple system, in which we are assured that if we start each year after the spring equinox, with the first visible new moon, the holy days will always be at the right time. Even the fall holy days, the whole Feast of Tabernacles, will always be in fall, after the harvest is completed, instead of in summer (Lev.23:39). It is the simple way that God intended from the beginning. For all the complicated rules used in the first century, before the Temple was destroyed, rules about the ripening of barley, the size of lambs, the condition of roads, the size of fruit, etc., the result was that each year was started exactly when God intended that it should be – with the first new moon after the spring equinox. In this way Jesus was able to keep each holy day on the right day, in spite of the rules that men had contrived to establish their own authority over other men.

But after that, God allowed the Jewish authorities to do whatever they wanted with the calendar. The result was that they invented a new system completely unrelated to God’s system. Months are started totally without regard to visible new moons, based on darkness rather than light. Holy days are delayed for reason of convenience. In the 5th century, years were begun a month too soon more than half the time, and the seventh month, not the first, is now regarded as the first month of the year. Even the names of the months used in that calendar are based on the life story of the Babylonian sun god, Tammuz. Of course, we use the names of Roman gods and goddesses like Mars and Juno when we use the Roman calendar. But we don’t call that calendar “sacred.”

Every adjustment and addition to this calculated calendar, over a period of perhaps 700 years, was done for good reason, and with the best of intentions. Before the institution of the present 19-year cycle by Hillel II in 358, the Roman Emperor Constantius (son of Constantine) persecuted the Jews to the extent that the Sanhedrin in Palestine was prevented from adding the “intercalary” month at the right time, and resorted to placing it after Ab, the 5th month (Jewish Encyclopedia).

Earlier persecutions had caused chaos in the calendar. The Encyclopedia Judaica says that, “Owing to the omission of intercalation over a period of some length, R. Akiva (d.137) once intercalated three successive years as an emergency measure.” The idea of systematically postponing the day now called “Rosh ha-Shana” (the first day of “Tishri”) if it would fall on Wednesday or Friday came about AD 300. The reason for this is to keep the day of Atonement from falling next to the weekly Sabbath. Before then it had been postponed by manipulation of the “sanctification” process when possible. The year itself was still based on the observed new moon of the first month before then, which they knew how to calculate in advance. That calculation of observed new moons is not part of the calculated calendar now used, and never was. It was only used to check reported observations of the new moon each month.

Once the first 6 months had each been given a fixed length, and it became acceptable to count back from the Molad Tishri instead, the first day of Tishri could also be postponed from Sunday to prevent the 15th of the first month, Nisan, from occurring on a Friday. In fact, the biggest advantage in using calculation of molads instead of visibility is that it is so easy to use postponements, since the start of the month has already been pushed back as far as possible to make room for them. At one time the number of 30-day months varied from 4 to 8 in common years and from 4 to 9 in leap years. It all became much “simpler” after everything was neatly standardized. But if it was God’s Sacred Calendar, why did it have to be improved? And if it became God’s Sacred Calendar only after all the improvements, what was it before all the improvements were made? Just when did it become sacred?

Which calendar was used in the first century?

We have been told that the same calculated calendar used today was used in the first century. How can that be if it was not invented until centuries later? We have been told that the only way Nisan 14 in AD 31 could have been on a Wednesday is if postponements were used then [A Wednesday crucifixion is required by the “sign of Jonah” (Matt 12:40)]. Doesn’t that prove that the same calendar used by the Jews of today was being used then? The answers are, “It cannot be.” and “It doesn’t.”

Besides the evidence of every secular history, and of the Talmud, showing that the calendar used by the Jews of first-century Palestine was based on the actual observation of new moons, the year of the crucifixion is an example showing that even the current 19-year cycle was not used then.

The 70 weeks prophecy (Dan. 9:24) shows AD 31 as the year of the crucifixion. From the going forth of the commandment to rebuild Jerusalem, which was from the first to the fifth month in 457 BC (Ezra 7:6-13), “unto the Messiah”, in AD 27, was 69 weeks or 483 years. Then a three-and-a-half-year ministry, as the gospels indicate, starting in the autumn, brings us to the month of Abib (also called “Nisan”), AD 31. The year 457 BC is certain, and we have other dating methods that fix a narrow range of years around AD 31.

The first month began in AD 31 on Thursday, April 12 (Julian calendar). That makes the Annual Sabbath on the 15th day of the month a Thursday, also, with the crucifixion on Wednesday. We have been told that this proves they were using a calendar much like the one used today, including postponements, but with a different order of intercalation. It proves no such thing.

The vernal equinox that year was on March 23. The first visible new moon after the equinox was seen after sunset on Wed., April 11, making Thursday, April 12, the first day of the new year. This is no evidence of postponements or of calculation. The moon was easily seen that evening, a 28-hour moon. This is the way it was done then, and history proves it. Something else is clear from this also. If the previous month had been the first month instead, it would have started 9 days before the equinox – certainly permitted (required, in fact) by today’s calculated calendar rules, today’s leap-year cycle. If the calculated calendar was used in that century, then someone needs to explain why that year started a month later. Just saying so isn’t enough.

Supposedly, the traditional calculated calendar used today was used in the first century, postponements and all, except that the order of intercalation (adding a 13th month) was different until AD 148. But history and astronomical evidence proves that this calendar was not invented until centuries later, and no calendar even remotely like it was in use in the first century. The Bible proves that this calendar is based on human tradition and what seems right to men, not on biblical principles. So where did the “experts” get the idea that a calendar kept secret until 210 years later had its rules changed (presumably in secret) in AD 148? Where are their historical proofs? Let’s see the evidence.

Egyptian papyrus: Records from the fifth century BC

During the fifth century BC, there was a large community of Aramaic-speaking Jews living on an island in the Nile called Elephantine, in southern Egypt near Aswan. This community was founded by the Persian king in the sixth century as a military colony to protect his empire. A temple was built there in 525 BC. Egypt was then ruled by a local ruler appointed by the Persian king.

About a hundred years ago, a large number of original papyrus documents were discovered there. Many are dated, and at least 22 of them have double dates, in both their own calendar and the 365-day Egyptian calendar. These dates are very useful for understanding the calendar used by the Jews then. Covering the years between 471 BC and 402 BC, these documents can show that

  1. The Jews living there sometimes counted years in their own civil calendar using a fall-to-fall regnal year, counting from the first of Tishri of the king’s reign, just as their ancestors had done before the Captivity, and as Nehemiah was doing at about the same time.
  2. Though their own calendar was very similar to the Persian (Babylonian) calendar, it can be shown that they were not using the same calendar.
  3. They did use the Babylonian month names (in Aramaic form).
  4. Dates were determined by local observation of the new moons.
  5. The first month was started on or after the vernal equinox, not before.
  6. They did not use the 19-year cycle that the Persians used. They relied on actual observation to decide when to start each year.

We can say this about the Jews living at Aswan in the 5th century BC: They had brought with them knowledge of the calendar that had been used by their ancestors in Judah before the captivity. That calendar began each year with the first month starting after, or perhaps “on or after” the vernal equinox. If they were observing the equinox by the method described in Part Two under the heading “Determining Equinoxes by Observing the Sun” it may be that they were indeed starting each year after the observed equinox every time, since the equinox would have been observed one day before the “actual” equinox. While each month was counted from the start of the first (spring) month, years were numbered starting with the 7th (fall) month. An advantage of this practice is always knowing in advance which month will be the last month of the civil year. Each month was begun with the observed new moon. No months had a predetermined fixed length. No 19-year cycles were used.

19-Year cycles

Since the average length of 235 lunar months (conjunction to conjunction) is equal to 19 solar years plus two hours, in every cycle each mean conjunction will be two hours later than in the previous cycle. After 12 cycles, or 228 years, the conjunctions are each an average of one whole day later compared to the equinoxes. About every two centuries or so, the order of intercalation would have to be adjusted to compensate for this drift.

Since there is a range of 30 days for the start of each year, and in each cycle there are 19 different starting dates, one of the years in the cycle is always going to start with a new moon very close to the vernal equinox. Actual conjunction times can vary as much as 10 hours from the seasonal mean because of the “anomalistic month.” That is a 20-hour range. Also, the “nodical month” can affect visibility from year to year. These two variations are out of sync with the 19-year period. So for at least one year in each cycle, that particular new moon may be seen sooner or later than the expectation, with respect to the equinox. This is how we can know that the Jews at Elephantine in the 5th century BC were not just relying on a fixed 19-year cycle, but were using real observations.


  1. We cannot start a new year until the old year is finished. If it’s still winter, the old year is not finished.
  2. If we start each year in spring there has to be barley for the wave sheaf offering – guaranteed.
  3. If we start each year in spring the Feast of Tabernacles will be entirely in fall – guaranteed.
  4. The relationship between the start of the religious year on the first of Abib, and the start of the agricultural and Jubilee years on the tenth of Ethanim (186 days later), is the same as the relationship between the two equinoxes. This relationship is from the first day of Abib, not from the passover or the wave sheaf offering.
  5. The Bible shows that Daniel himself was regulating the Babylonian calendar in the 6th century BC. From that time on years were begun after the spring equinox.
  6. The Jews living at Elephantine in the 5th century BC, at the time of Ezra, were using a calendar based entirely on years starting at or after the spring equinox. But they did not use the 19-year cycle that was then being used in Babylon. It was a different calendar.

Genesis 1:14 says plainly that the light of the sun, moon, and stars alone is all that is needed for the calendar. No barley observations, no 19-year cycles, no fixed-length months, no postponements, no “duly authorized religious authorities”, just our own eyes.

Written by Derek G. Davies. Dated Sept. 22, 2002.
Edited June 2005 and March 2013.