The Jewish calendar is lunisolar, New Moons start new months and periodically, seven times every nineteen years, the rabbis insert a thirty-day month, Adar I, keeping the lunar months and the seasons synchronized. The Jewish calendar started as an observational calendar, witnesses appearing before the priests, and later the Sanhedrin, testifying to seeing the New Moon. If the High Priest/Sanhedrin president identified two reliable witnesses, he sanctified the month, then sending out messengers announcing the event.

Things slowly changed after the conquest of Judea by Alexander the Great and the diffusion of Hellenic philosophy, science and mathematics throughout the Middle East. Some rabbis learned Greek astronomy and mathematics, resulting in a calculated Jewish calendar based on the average speed of the moon, calling the time of the New Moon’s arrival the molad, from the Hebrew root for the word birth, as though the moon were being “born anew.” By the late Second Temple period until the mid-fourth century CE, rabbis from the Sanhedrin’s “calendar council” secretly computed the molad and the witness system becoming a perfunctory exercise.

Seeing the growing violence against Jews, particularly after the Council of Nicaea, in 325 CE, took steps to establish Christianity as a state religion, and the severe edicts enacted by the Roman emperors, including banning the distribution of the Jewish calendar, President of the Sanhedrin Hillel II, with the Sanhedrin’s approval, in 359, publicly disseminated the heretofore secret rules governing the construction of the calculated calendar, thereby allowing anyone with sufficient knowledge to construct one. Sanhedrin President Gamaliel VI died in 425 without a male heir and Emperor Theodosius II abolished the office, yet historical documents show that Babylonian Jewry received its calendrical information from the Land of Israel as late as 835.

According to the Book of the Calendar Controversy, found in the Cairo Genizah in the early twentieth century, Aaron Ben Meir, a highly esteemed scholar and Head of the Jewish community living in Muslim-occupied Israel, challenged, in A.M. 4682 (921 CE), Babylonian Jewry’s power to construct the Jewish calendar by declaring on the Mount of Olives that the months of Heshvan and Kislev would be defective (both having 29 days), and as a result, Passover 4682 will fall on Sunday, contradicting the pronouncement of the Babylonian Sanhedrin whose calendar said the months of Heshvan and Kislev will be full (both having 30 days) and Passover falling on Tuesday, two days later.

I digress to make two points:

First, for us, there are 60 minutes in each hour; for Jewish-calendar makers, there are 1,080 parts per hour. Each part equaling 31/3 seconds. Since we are using minutes in this article and not parts, you see fractions, but Jewish-calendar makers see whole numbers.

Second, a molad has two parts, the calculated day and time for a New Moon. The molad for the month in which Rosh Hashanah falls is the lynchpin upon which the entire Jewish calendar hangs, because this molad also fixes the length of the year. Together, the day of the week on which Rosh Hashanah occurs and the length of the year determine the days of the week on which all Jewish holidays fall and the order in which synagogues read Torah portions throughout the year.

Time is the tricky part of the molad, as complex rules delineate cut-off points, which might cause postponement of Rosh Hashanah by one or two days and possibly changing the length of the year. The rules are so precise that a change of as little as 31/3 seconds can alter Rosh Hashanah’s day of the week, length of the year, or both. So, Ben Meir’s calendar affects more than just Passover, it alters the entire structure of the calendar for that year.

Two different molad calculations create the discrepancy between Ben Meir’s and the Babylonian rabbis’ calendars. Ben Meir determined Rosh Hashanah should take place on Thursday, September 6, 921 (Julian-calendar year) at 5:16:62/3 AM, making the year a deficient leap year with 383 days and Passover falling on Sunday, April 14, 922. Babylonian Jewry calculated Rosh Hashanah occurring on the same day at 5:51:462/3 AM in an abundant leap year having 385 days, and Passover falling on Tuesday, April 16, 922. two days later. In essence, Ben Meir declaring the time between New Moons is actually 35 minutes, 40 seconds, or 352/3 minutes sooner than Babylonian Jewry’s calculated molad.

Ben Meir’s proclamation led to a schism between the two Jewish communities. The Babylonian rabbis pleaded with him to recant, but he refused. According to Broydé, Ben Meir “denied them any authority in astronomical matters; and, owing to his own reputation and that of his family, won the confidence of Jews in many countries.” Even the intervention by the illustrious Saadiah Gaon could not dissuade Ben Meir, and in the end, most Jews living in the Land of Israel and some Babylonian communities followed Ben Meir in celebrating Passover two days earlier than the Babylonian rabbis’ determination.

It seems pretty obvious Ben Meir aimed to return Jewish-calendar construction, a highly-prized prerogative, back to the Land of Israel. But why choose 921? The two calculation methods coincide better than 97 percent of the time, but they differed in the tenth century in five years and for the first time in 921.

Babylonian leaders excommunicated Ben Meir because of his arrogance and warned people not to follow his teachings. We do not know Jewish reaction to this warning, but if it had little or no effect, then some Jews might celebrate Rosh Hashanah A.M. 4683 (922) on Tuesday in a year with 354 days and Passover 923 falling on Thursday, while others would celebrate Rosh Hashanah on Thursday in a year with 354 days and Passover falling on Saturday. Ben Meir’s Rosh Hashanah calculations in 923, 926, and 927 also differ from the dates announced by the Babylonian rabbis.

More difficult to explain is how Ben Meir arrived at his conclusion, as he left no records. Jewish calendar experts suggest some possibilities. We present two. The first is the choice of “head of the year.”Reingold and Dershowitz presume Jews living in Israel “did their calculations from Nisan, instead of Tishri, and rounded the time of the epochal new moon differently.” The Torah describes Nisan as the first of the months when referring to Passover and national liberation, but was Nisan ever the first month of the year? Maybe.

There are four pertinent biblical references. 2 Samuel 11:1 reads, “At the turn of the year, the season when kings go out [to battle].” I Kings 20:22 and 26 and II Chronicles 36:10 also refer to kings battling at the “turn of the year,” implying these events occurred in Nisan, as kings “go out to battle” from spring into summer and not fall into winter, and suggesting at some time in the remote past, perhaps as far back as the tenth century BCE, Nisan represented the head of the year.

Another “head of the year” explanation says that during the Babylonian captivity, beginning in the sixth century BCE, Judaism adopted its host county’s “head of the year,” Ni’sannu, thereby creating two New Year celebrations. Even today Jews celebrate two: One on January 1 and another on 1 Tishri. In support of these, the Talmud, Rosh Hashanah 12a, quotes a baraita, an old tradition in Jewish oral law not incorporated into its findings, saying, “The Jewish Sages count the years from Creation and the flood in accordance with the opinion of Rabbi Eliezer, from Tishri, and they calculate the cycles of the sun and the moon in accordance with the opinion of Rabbi Yehoshua, from Nisan.”

A second possibility comes from locations on Earth. Adam lived in the Garden of Eden when he saw his first New Moon. Tradition holds that God placed the Garden of Eden in Babylonia, where the Tigris and Euphrates rivers run into the Persian Gulf. Perhaps in his attempt to bring the calendar back to Jerusalem, Ben Meir changed the location for molad computations, from the Garden of Eden (Babylonia) to Jerusalem. We don’t know the exact coordinates chosen, but the separation between Jerusalem and Pumbedita (near the modern city of Fallujah), the location of an important Jewish academy and seat of the Sanhedrin responsible for constructing the calendar, is a little more than 8.9 degrees of longitude. The earth takes 4 minutes to rotate 1 degree of longitude, or about 352/3 minutes between the two cities.

The two calendars are in complete agreement all other years in the tenth century and all years in the eleventh century. There is evidence, however, also from the Cairo Genizah, that another calendar controversy ensued in 1107 and 1108, the next time the two calendars differed. There are no recorded calendar disagreements after this date. Ben Meir’s and the Babylonian rabbis’ calendars last differed during the fourth decade of the nineteenth century, 1840, 1843, and 1844. They will differ five times in the twenty-first century: 2024, 2025, 2027, 2028, and 2029, then not again until 2246.

An outsider might consider it very strange that a calendar, a seemingly mundane object, and not some grand philosophical disagreements about theology, or even Jewish law, ritual, or prayer provoked a religious rupture. This acrimony is understandable in Judaism, however, because producing an accurate religious calendar is a grave responsibility, proving the veracity of Rabbi Samson Rafael Hirsch’s words, “The catechism of the Jew is his calendar.”

Republished from San Diego Jewish World