Lectures on the weekly Torah reading by the faculty of Bar-Ilan University in Ramat Gan, Israel. A project of the Faculty of Jewish Studies, Paul and Helene Shulman Basic Jewish Studies Center, and the Office of the Campus Rabbi. Published on the Internet under the sponsorship of Bar-Ilan University's International Center for Jewish Identity.
Inquiries and comments to: Dr. Isaac Gottlieb, Department of Bible, gottlii@mail.biu.ac.il
Parashat Vaethanan 5759/1999
Can Astronomy Help Determine When the Hebrew Calendar was Founded?
Yaakov Lewinger, Eng.
Tel Aviv
One verse in this week's portion has been singled out by the Talmud to refer to the Jewish calendar: "'For that will be proof of your wisdom and discernment' (Deut. 4:6)--This [wisdom and discernment] refers to calculation of the seasons and zodiac signs" (Shabbat 75a). Further, Rambam tells us that "computing the seasons and months means the science of astronomy" (Maimonides, Resp. [Blau], 150).
1. Traditions dating the establishment of the calendar
In a responsum dating to 992, Rav Hai Gaon wrote that a fixed Jewish calendar had been followed since the time of Hillel son of Judah ha-Nasi in 359.[1] As he put it: "Since the time of Hillel bar Judah in the year 670 for dating bills [le-minyan shetarot] (= 4119 by the Jewish calendar), since that year they neither moved up nor postponed, but followed this order [of regular years and leap years--Y. L.]."
Maimonides has the following to say about when the calendar was set:
When did all the Jews begin to go by these calculations [of the calendar--Y. L.]? From the end of the time of the Sages of the gemara, when the land of Israel was destroyed and there was no longer a regular court sitting there ... but until the days of Abaye and Rava they relied on what was determined in the land of Israel [according to when the new moon was seen--Y. L.]. (Maimonides, Mishneh Torah, Hilkhot Kiddush ha-Hodesh 5.2)
Abaye and Rava died near the middle of the fourth century,[2] thus these two views of Rav Hai and Maimonides are not far apart.
When the calendar was established is an important question in principle because of the rules about the body authorized to determine the new moon: "This is not to be done by any other than a Sanhedrin in the land of Israel or a court in Israel vested with the power to do so by the Sanhedrin" (Maimonides, ibid. 5.1). Therefore, if the calendar was indeed established in the fourth century, when a rabbinical court with this authority still existed,[3] we can ascribe its establishment to the Sanhedrin which had the authority to determine the time of the festivals during the year.
There are indications in the Talmud that "today we know how to determine when the new moon will be" (Babylonian Talmud, Betza 4b and elsewhere), i.e., that they knew certain rules for setting the calendar. However, to our great surprise, nowhere throughout the entire body of Talmudic literature, neither in the Babylonian Talmud nor in the Jerusalem Talmud, is there a detailed list of rules for setting the calendar as we know them today--such as molad be-HR"D (in Tishre of the first year), or alternatively, molad VY"D (in Tishre of the second year), a nineteen-year cycle and the pattern of leap years--even though the editing of the Talmuds was completed after 359. This places one of the central commandments of the Torah--determining the time of the festivals--in the category of "mountains hanging by a thread," i.e., subjects whose sources are few yet whose halakhot are numerous (cf. Mishnah Hagigah 1.8).
2. Problems raised by the traditional date given for establishment of the calendar
Many noted scholars who investigated the calendar[4] have raised serious problems challenging the traditional dating of the calendar. Namely, there are several dated documents ranging from the year 359 to 839 that do not accord with our calendar. Among the evidence challenging the early date ascribed to the calendar are the following sources:
1. The data on several dated tombstones found at Zoar on the shore of the Dead Sea, from the fourth to the fifth centuries, appear to contradict our calendar.[5]
2. Several facts pertaining to the date of death of a rabbi in the early fifth century, appearing in a letter of Rav Sherira Gaon (from 988) appear not to match our calendar.[6]
3. An epistle from the Resh Galuta in Babylon, dating to 836, discovered in the Cairo genizah, on the basis of the determination of rabbis from the land of Israel sets the celebration of Passover that year on a date different from our calendar.[7]
4. Early works such as Pirkei de-Rabbi Eliezer (redacted c. 740) and Baraita de-Shmuel (redacted c. 777), appear unfamiliar with our calendar and mention cycles, dates for the new moon, and seasons that are different from those which we follow.[8]
5. The fact that there was a dispute between R. Saadiah Gaon from Babylon and R. Ben Meir, head of the Academy in the land of Israel, regarding the time of the holidays in the years 922-924 indicates a tradition that was not yet firmly established regarding the rules of the calendar. This tends to support the hypothesis that these laws had only gained new force shortly before the time of these rabbis.[9]
On the basis of these and other similar arguments some experts have tried to place the establishment of the calendar some time after 836, when the epistle from the Resh Galuta which contradicts our calendar was written. Others[10] found various ways of interpreting some of the documents mentioned and pointed to other eighth and ninth century documents from which one can conclude that our calendar was known before 836. They include:
1. The Aramaic translation of the Torah ascribed to Jonathan ben Uziel (apparently redacted c. 8th century) on Genesis 1.9, which explicitly mentions the appearance of the new moon in Tishre in the year before year one of the cycle according to our reckoning, and this piece of information is in complete agreement with our calendar.
2. A work by the Moslem mathematician and astronomer al-Khwarizmi, dating to 824, sets forth in detail the rules governing our calendar and remarks, "This [calendar -- Y. L.] was preserved by the select few, ... and is incomprehensible to the vast majority."[11]
3. Several other documents from the time of the geonim mention certain of the rules of our calendar, which it is argued were not founded until later.
Unfortunately, however, it is hard to establish the exact date of the relevant passages and to determine whether they were not perhaps added to the documents at a later date. Therefore, there is still much confusion regarding the date our calendar was finally set. Hopefully, as the study of the documents from the Cairo genizah progresses additional dated documents will be found that can shed light on this question.
3. Astronomical evidence for the antiquity of our calendar
Astronomy is the originator of all calendars because every calendar is based on the motion of the heavenly bodies. Can we use astronomy to held determine when our calendar was set? Such evidence is immune to charges of being a "later addition" and hence would be very important. It is well known that the Jewish calendar is based on the "mean" time of appearance of the new moon[12] in Tishre (alternatively, Nisan), which determines when Rosh Ha-Shanah (alternatively, Passover) falls, and on "the true period of Nisan according to the sun" (i.e., the spring equinox), which determines whether a leap month must be added so that Passover will always occur in the spring month [=hodesh he-aviv]. It is also well known that our calendar is based on a 19-year cycle, known since before the Common Era, consisting of 12 ordinary years (of twelve months each) and 7 lunar leap years (of thirteen months), which on balance are a good approximation to nineteen solar years (measured from one spring to the next).
Therefore we may reasonably assume that the founders of the calendar, who set the lunar cycle in our calendar, began counting their nineteen-year cycles from a year in which the true period of spring (according to astronomy) coincided reasonably closely with the time for the appearance of the new moon of Nisan according to the calendar. Only in such a way could the occurrence of the actual spring equinox throughout the nineteen years of the cycle serve as a criterion for determining whether a leap month should be added to the lunar year. The rule is that a leap month must be added in years when the spring equinox would otherwise fall 16 days or more after the new moon ofNisan.
This criteria becomes distorted over time, the further we are from the date the calendar was established, because 19 solar years are actually a trifle shorter than the 235 lunar months in one 19-year lunar cycle of the calendar. Thus, as the years go by since the founding of the calendar, the spring equinox will no longer coincide with the appearance of the new moon in the first year of each cycle, but will move up approximately one day every 216 years. Therefore, to find astronomical verification that our calendar was founded in 359 we must examine whether the spring equinox actually coincided with the appearance of the new moon in Nisan in the first year of the cycle then being followed.
The year 359, which is 4119 by the Jewish calendar, fell in the 217th lunar month of our calendar. In the first year of this cycle, namely 4105 by the Jewish calendar, both the actual spring equinox and the "mean" appearance of the new moon of Nisan fell on the 29th of Adar, corresponding to March 20, 345, with only about a six-hour difference between the two.[13] Therefore, in terms of astronomy, the tradition that our calendar was founded near this time appears quite reasonable. In contrast, in 839--the first year of the 243rd lunar cycle, 4599 by the Jewish calendar, and close to the later date claimed for the establishment of the calendar--the actual equinox had already shifted about two days earlier than the new moon of Nisan. Therefore, it is not reasonable to assume that the 19-year cycle of our calendar was established close to this time.[14] For reasons of astronomy, a more suitable year than 4599 would have been chosen to be the first year of the 19-year cycle. In 1998, which was the first year of the 304th lunar cycle, the actual equinox fell more than seven days before the new moon of Nisan--a discrepancy which raises certain problems which we have discussed elsewhere.[15]
In conclusion, for reasons of astronomy alone it is reasonable to accept the tradition that our calendar, its pattern of leap years, and certain of its fundamental principles were founded close to 359, and not in the ninth century, as those who would date the calendar later maintain.
Of course, this finding based on astronomy does not resolve the problems raised by the documents listed in section 2 above. Therefore, it may well be that over the years the calendar was further refined, after its original founding in 359, until it reached its current form; alternatively, the details of the calendar may have been kept secret from the masses for fear of the authorities and of hostile sects, and therefore certain inaccuracies in applying it may have appeared in the documents mentioned above. Be that as it may, the astronomical findings indicate that we must treat with due respect Rav Hai Gaon's dating of our calendar since there are solid grounds for believing his opinion.[16]
[1] This responsum is recorded in Sefer ha-Ibbur (3.7) by Rabbi Abraham Bar Hiyya ha-Nasi of Barcelona, Spain, a work composed circa 1123; cf. Philipowski ed., London 1851, p. 97; cited in Otzar ha-Geonim, R.H. (ref. pg. 10b) par. 7, p. 16.
[2] Abaye passed away in 338, and Rava in 352; according to Iggeret Rav Sherira Gaon, Levin ed., Haifa, 1921.
[3] The institution of Nasi in the land of Israel was abolished circa 429, by order of the Byzantine rulers; cf. M. Avi-Yonah, Bi-yemei Roma u-Vizantiyon, Jerusalem 1970, 5th ed., p. 197. The date the Sanhedrin was abolished remains uncertain.
[4] Several important names of researchers into our calendar (through present-day scholars) include: Azariah de Rossi ["of the Edumim"] (16th century), Raphael Halevi of Hanover (18th century), Hayyim Zelig Slonimsky, Hirsch Mendel Pineles (19th century), Hayyim Yehiel Bornstein, Tz. H. Jaffe, R. Menahem Kasher, Umberto (Moses David) Cassuto, and Avraham Arye Akavya (20th century).
[5] On the Zoar tombstones, cf. L. Sukenik, Kedem 2 (1944), pp. 83-88; also cf. Y. Naveh, Tarbiz 64 (1995), pp. 477-497.
[6] Ts. H. Jaffe, Korot Heshbon ha-Ibbur, Tel Aviv 1931, p. 51.
[7] Iggeret Rosh ha-Golah; Cairo genizah document T-S 8 G7, published by Y. Mannen, London 1922. Cited in Otzar ha-Geonim, R. H. (ref. p. 20a), par. 42, p. 35.
[8] Pirkei de-Rabbi Eliezer, chapters 6-8, Warsaw ed., 1852; Baraita de-Shmuel ha-Katan, ch. 5, Jerusalem ed., 1932.
[9] Hayyim Yehiel Bornstein, Mahloket Rav Saadiah Gaon u-Ven Meir, Warsaw 1904.
[10] Menahem M. Kasher, Torah Shelemah, 13, Parshat Bo, New York 1950; Prof. E. Beller, "Matai Nosad ha-Luah ha-Ivri?" lecture given at the Seminar on the Hebrew Calendar, Bar Ilan University, December 1998.
[11] Ts. Langerman, "Aimatai Nosad ha-Luah ha-Ivri?" Asuppot 1, Jerusalem 1987, pp. 159-168.
[12] On the actual appearance of the new moon, its mean appearance, and the appearance of the new moon according to the calendar, cf. Y. Lewinger, "Sheloshet Sugei ha-Moladot," Sinai 118 (1996), pp. 71-82.
[13] Tekufat Nisan le-Rav Ada, or the spring equinox according to Rav Ada, is essentially the actual spring equinox in 345 which was rounded so that, by calculating backwards, it would fall exactly at the beginning of the fourth day of the week (the day on which light was created) in the year 1 of the Hebrew calendar and would also come out a round hour (23:00, the time in the evening that the day was considered to begin) in the year D-K-H (meaning "minute"; numerical value=109). The new moon of Tishre of the following year, 110, was also calculated to occur at a round hour (13:00).
[14] In 839 the mean molad and the mean solar spring season (when the mean angle of the sun is 0 degrees from the point of spring) indeed coincided by a fairly close approximation. However this fictitious point always occurs in Nisan two days after the actual astronomic point. Therefore it has nothing to do with day and night being of equal length, which is the only significant point in determining when Passover falls. Cf. Y. Lewinger, "Ha-Hodesh ha-Zeh--Hodesh ha-Aviv," Daf Shavui 167, 1997, Bar Ilan University.
[15] Y. Lewinger, "Ha-im ibarnu et shnatenu (5757) lashav?" Daf Shavui 190, 1997.
[16] This article summarizes the lecture given at the 1998 Conference
on Torah and Science, held at Bar Ilan University.
Prepared for Internet Publication by the Center for IT & IS Staff at Bar-Ilan University.