What Is the Bible’s Calendar?
As part of its introduction to the Pesach offering, Exodus 12:2 declares that the month in which the exodus from Egypt will occur should be counted as the first month of the year:
שמות יב:ב הַחֹדֶשׁ הַזֶּה לָכֶם רֹאשׁ חֳדָשִׁים רִאשׁוֹן הוּא לָכֶם לְחָדְשֵׁי הַשָּׁנָה.
Exod 12:2 This month is unto you the beginning of months; it is to you the first of the months of the year.
The text here does not define the month of the exodus, though other passages clarify that it is the month of Aviv (Exod 13:4, 23:15, 34:18, Deut 16:1). The word aviv refers to a stage of ripening of the crops (see Exod 9:31), suggesting that the month of Aviv is sometime in the spring. Yet, the Torah does not clarify exactly when this month, or any other month of the year, begins and ends.
This is rather strange. Severe punishments are prescribed for those who transgress the festivals; yet we are not told how the calendar works, and thus the dates when these festivals are supposed to fall. This ambiguity is highlighted by a debate between two medieval commentators about whether the Torah assumes a lunar or solar calendar.
A Solar Calendar or a Lunar Calendar: Yehudah Ha-Parsi versus Ibn Ezra
Yehudah ha-Parsi (Judah the Persian) is said to have argued that the calendar of the Bible was solar. He lived in or around the 9th century C.E. All that is known of him comes from Abraham ibn Ezra (1089–1167) in a handful of passages in his Pentateuch commentary (Introduction, Exod 12:2 [long recension], Lev 25:9, Num 3:39). For example, in his gloss on Exodus 12:2, ibn Ezra writes:
א”ר יהודה הפרסי: כי ישראל היו מונים כפי שנת החמה, וחדשיהן חדשי החמה, כמשפט הערלים. וראיתו: ושמרת את החקה הזאת למועדה (שמות י”ג:י’), כי שנת לבנה איננה שוה, כי ימי החריש והקציר תלויים בהליכת השמש לבדו, כפי נטותה לצפון או לדרום.
R. Yehudah the Persian said: “The Israelites used a solar calendar, and their months were solar months, like the non-Israelites do.” His proof is from [the verse (Exod 13:10)]: “Keep this statute (Matzot/Pesach) at its appointed time” since lunar years are not even, since the days of plowing and harvest are dependent only on the movement of the sun, on its moving northward or southward.
R. Yehudah Ha-Parsi would have had in mind something similar to the Julian calendar (the ancestor of our Gregorian calendar): a 365-day year with an additional leap day every four years.
Abraham ibn Ezra polemicizes vehemently against him, and attempts to demonstrate that the calendar of the Bible is lunar. Yet, even he falls back on the claim that the ultimate reason we know that the lunar calendar is correct is from oral tradition (Introduction, path #2):
כי אין בתורה חקי השנה מפורשים, ואיך נחשוב החדשים?… וזה לנו האות שסמך משה על תורה שבעל פה…
For the Torah does not have the rules of [determining] a year explicit, so how can we calculate the months?… Rather this demonstrates that Moses relied on the Oral Law…
In short, arguing that one calendar or another is the one used in the Bible is no easy task, since the Bible makes no explicit statement about its nature.
Earliest Sources for the Lunar Calendar
In Jewish tradition, the calendar is lunar, with the months beginning at the new moon. Years are made up of twelve such months, but sometimes a thirteenth month is added, to keep up with the seasons, ensuring that Passover is celebrated in the spring, Sukkot in the fall, etc.
Such a lunar calendar is already assumed by Philo of Alexandria (ca. 25 B.C.E.–50 C.E.) in his Special Laws (2, 41, 140–142), and of course, in rabbinic literature (e.g. Mishnah Rosh ha-Shanah). But long before them, it is already implicit in Ben Sirah (Sirach, early 2nd cent. B.C.E.), who describes the moon as “an indicator of times … a sign of the festival … after whom the month is named” (43:6-8 LXX); also “וּכְיָרֵחַ מָלֵא בִּימֵי מוֹעֵד” (“like the full moon in the days of the festival,” 50:6 Heb) implies a lunar calendar in which most festivals fall on the full moon.
Identifying with the Babylonian Lunar Calendar
Earlier still, the postexilic books of the Bible adopt Babylonian month names; this implies that the Jewish calendar was identified with the Babylonian one, which was lunar. Thus, in a number of biblical verses in exilic or postexilic works, Babylonian months such as Nisan and Sivan are used and explicitly equated with the numbered months of the Torah. For example,
זכריה א:ז בְּיוֹם עֶשְׂרִים וְאַרְבָּעָה לְעַשְׁתֵּי עָשָׂר חֹדֶשׁ הוּא חֹדֶשׁ שְׁבָט…
Zech 1:7 On the 24th day of the 11th month, which is the month of Shevat…
Furthermore, the Passover Papyrus from Elephantine (419 B.C.E.) confirms that Passover was celebrated on the 14th, and Unleavened Bread, on the 15th–21st, of the Babylonian month of Nisan, indicating that his community used a lunar calendar. But even though the lunar calendar seems to have been standard among Second Temple period Jews, it was not without dissenters.
The 364-Day Calendar
The Book of Jubilees (2nd cent. BCE) upholds a calendar of 364 days, one day less than a solar calendar and about ten days longer than an average lunar year. Moreover, its months do not correspond to lunar phases, and do not begin at the new moon. It also explicitly polemicizes against those of who follow the lunar calendar
There will be people who carefully observe the moon with lunar observations because it is corrupt (with respect to) the seasons and is early from year to year by ten days. Therefore, years will come about for them when they will disturb (the year) and make a day of testimony something worthless and a profane day a festival. (6:36)
The Dead Sea Scrolls, dating from the 2nd century B.C.E. to 1st century C.E., only know of this 364-day calendar, and thus, at Qumran, all the biblical festivals are dated according to this non-lunar, and equally non-solar, schematic calendar. The origins of this 364-day scheme are unclear, but since the Bible does not explain how its calendar works, it would have been difficult for anyone to contradict the Jubilees-Qumran tradition.
Despite the lack of any explicit indication in the Bible, Jewish commentators attempted to deduce the nature of the calendar from clues they found in the text.
Genesis 1:14 states that the celestial bodies were created “וְהָיוּ לְאֹתֹת וּלְמוֹעֲדִים וּלְיָמִים וְשָׁנִים” (“to be for signs and for appointed times, days and years”). Which of these bodies were created for what, and what is meant by “appointed times” (מוֹעֲדִים) is unclear. Ibn Ezra suggests that this verse refers to festivals (על מועדי השם המקודשים), though in the context of the creation story, this is unlikely.
Other commentators, in contrast, suggest it refers to the signs used to determine lunar month. This is possible but not definitive since the term “appointed times” (מוֹעֲדִים) could also mean “seasons,” as it is often translated.
Another instance of “appointed times” (מוֹעֲדִים) deployed as a prooftext for the biblical origin of the lunar calendar is Psalm 104:19:
עָשָׂה יָרֵחַ לְמוֹעֲדִים
שֶׁמֶשׁ יָדַע מְבוֹאוֹ.
He made the moon for appointed times,
The sun knows its setting.
Genesis Rabbah 6:1 interprets the opening phrase as a reference to the lunar calendar (Theodor-Albeck ed.):
אמר רבי יוחנן לא נברא להאיר אלא גלגל חמה בלבד, אם כן למה נבראת לבנה למועדים, כדי לקדש בה ראשי חדשים ושנים.
R. Yohanan said: “Only the ball of the sun was created to shine light [upon the earth]. If so, why was the moon created? ‘For appointed times,’ in order to sanctify through it the new moons and years.”
But again, we must question whether the peshat of the verse really refers to festivals, or rather to some natural phenomenon.
Even Abraham ibn Ezra, who strongly supports the lunar calendar, admits that the context of the psalm as a whole militates against this verse having any connection to the calendar, suggesting that it instead refers to the lunar phases. Moreover, the second half of the verse functions as a poetic parallel to the first part, and thus the sun could also be taken as belonging to the first half, and may be seen as the subject of “appointed times” (מוֹעֲדִים).
In passing, the non-halakhic, aggadic nature of this midrash is noteworthy. Its purpose—unlike halakhic midrash—is to explain why the moon was created, not to prove that the calendar is lunar. In fact, such a proof is surprisingly absent in the entire corpus of rabbinic literature.
Evidence from Terminology: Chodesh versus Shanah
It has long been argued that the main Biblical Hebrew term for month, חֹדֶשׁ (chodesh), indicates a lunar calendar. The related term חָדָשׁ means “new,” and newness cannot apply to the sun, but only to the moon, which disappears and renews itself every month. Likewise, the less common Hebrew word for month, יֶרַח, is derived from the term יָרֵח, “moon,” proving, some state, that the biblical calendar was lunar.
On the other hand, Abraham ibn Ezra (Exod 12:2) points out that the term for year is שָׁנָה (shanah), meaning “repetition” or “change,” and that this only works for a solar year, in which the sun re-commences its cycle. The term shanah does not suit a lunar calendar, which can have twelve or thirteen months, and where the new year is not always a repetition of the previous one.
This leaves us with a complex situation. The two words in biblical Hebrew for month fit only with a lunar calendar while the word for year in biblical Hebrew, shanah, is only suited to a solar calendar.
Ibn Ezra makes this point explicit in his gloss on Leviticus 25:9:
ועוד כי פי’ חדש יכחיש הפרסי… דע כי אין ללבנה שנה כלל, רק בקשו המחשבים מספר חדשים קרובים לשנת החמה ומצאום י”ב כאשר אין לחמה חדש והמחשבים בקשו מספר לחדש שיהיה נחלק קרוב לחדש מימות חדשי הלבנה, על כן חדשינו הם ללבנה, ושנותינו ישובו בסוף לשנת החמה.
Moreover, the meaning of chodesh (new) contradicts [Yehudah] HaParsi… Know that the moon has no “year” at all. Only those who made calculations (for the calendar) wished for a number of months that would approximate a solar year, and they came up with twelve. At the same time, the sun has no “month,” and those who made calculations (for the calendar) wished for a number [of days] to comprise a month, such that it would approximate the days in a lunar month. Therefore, our months are lunar, and our years return in the end (i.e., through leap year adjustment) to the solar years.
We are thus forced to conclude that the Hebrew terminology does not favor the existence of either a solar or lunar year in ancient Israel.
Evidence from Biblical Narratives: The Flood versus Sinai
Some scholars have tried to deduce the nature of the biblical calendar from dates in certain narratives, but these proofs cut both ways.
The narrative of the Flood has been used, by some, as evidence that the calendar was lunar. According to Genesis 7-8 (following the Masoretic Text and the Samaritan Pentateuch), the Flood began on the 17th of month 2 (Gen 7:11; the LXX has 27th) and ended on the 27th of the same month the next year (Gen 8:14). It thus lasted one year and ten days.
As pointed out already in Genesis Rabbah 33, these ten days seem to represent the difference of 10–11 days between the solar year (c. 365 days) and the lunar one (c. 355 days).
ולא היה צריך קרייה למימר אלא בששה עשר יום לחדש יבשה הארץ ומה תלמוד לומר בעשרים ושבעה לחדש השיני יבשה הארץ, אלא אילו י”א יום שימות החמה יתרים על ימות הלבנה,
The verse should have said “on the 16thof the month the land dried up,” [which would have been 12 full months] what does the verse teach us by saying “on the 27th of the second month the land dried up”? These are the 11 extra days in a solar year that are not part of the lunar year.
The use of these two dates suggests, Genesis Rabbah argues, that the biblical calendar year was lunar. Since, for whatever reason, the flood had to last one solar year, ten days were added to the lunar, calendar year.
This argument may be attractive, but actually, the exact reverse can be inferred from the stay of the Israelites at Sinai. For they arrived at mount Sinai, after the Exodus, on the 1st of month 3 (Exod 19:1); and they left the next year on the 20th of month 2 (Num. 10:11). This suggests that they stayed at Sinai for one year less ten days. On this basis it could be argued that the biblical calendar was solar. Since, for whatever reason, the Israelites had to stay in Sinai for one lunar year, they left ten days before the end of the solar, calendar year.
The Earliest Month: R. Abraham bar Chayya’s Proof
To resolve this conundrum, a more adventurous inference is proposed by Abraham bar Chayya, who wrote perhaps in France, in 1123, the earliest surviving, complete treatise on the Jewish calendar, commonly called Sefer ha-Ibbur (The Book of Calendar Calculation). In chapter 2:5 of this treatise, he offers a new exegesis of Exodus 12:2, whose literary structure calls for creative exegesis, in rabbinic thinking:
הַחֹדֶשׁ הַזֶּה לָכֶם רֹאשׁ חֳדָשִׁים
רִאשׁוֹן הוּא לָכֶם לְחָדְשֵׁי הַשָּׁנָה.
This month is unto you the beginning of months;
it is to you the first of the months of the year.
Abraham bar Chayya proposes that the repetition in the verse indicates not only that the Passover month is the first of the year, but also that in the year of the Exodus, this month fell earlier than in any other year.
כל ראש חדש שיהא בתוך ימי החדש הזה ראשון הוא לכם לחדשי השנה, ואם יהיה קודם לימי החדש הזה אינו ראשון אלא החדש הבא אחריו אשר יהיה בתוך ימי החדש הזה. ומשם רמז לעבור השנה.
Any Rosh Chodesh (New Moon) that falls within the days of this month [the agricultural or solar month of Aviv] is unto you the first of the months of the year, but if it falls before the days of this month, it is not the first [month] but rather the month that follows, whose [new moon] falls within the days of this month. From there is an allusion to the intercalation of the year.
This only works, however, if the calendar is lunar. For as we have seen, in a lunar calendar the year can be of twelve or thirteen months. This means that sometimes the year can start earlier, and sometimes later.
By the Middle Ages, the Jews had adopted a 19-year cycle of twelve and thirteen-month years, at the end of which the lunar year returns to the same position in relation to the sun. Assuming that the same cycle was used in biblical times (which is very unlikely), Abraham bar Chayya calculates that in the year of the Exodus, Nisan would have fallen the earliest in the whole cycle. This was the meaning of the “beginning of months … the first of the months of the year.”
This calculation fails, however, because Abraham bar Chayya chooses the wrong date for the exodus—2448 from the creation—whereas the year was 2450 (according to our count of years), as he acknowledges himself later in the same book (chapter 3:8). In that year, the first month would have fallen rather later.
But Abraham bar Chayya’s inconsistency, the intricacies of his calculation, and its inherent problems, should not concern us here. Suffice it to note that through this ingenious exegesis, Abraham bar Chayya claims to prove that the 19-year cycle was used, and that the calendar was lunar. The reader can decide.
Archaeological Evidence for Lunar Calendar
An entirely different approach to the problem is to look for archaeological evidence about calendars in this period. A number of very small artefacts, discovered in the area of ancient Judah and dating from Iron Age II (early first millennium), have been convincingly identified as calendars.
These pocket-size, animal-bone plaques are all perforated with three rows of ten holes, which were probably designed for tracking the days of the month: a peg would be inserted and moved along the holes on a daily basis, so as to keep track of the current date. One such plaque from Aroer, contains an additional twelve-hole row, for tracking the twelve months of the year.
This common artefact might tell us something about the calendar of ancient Israel or the Hebrew Bible. Thirty peg holes are well suited for a lunar calendar. Solar calendars can have 31-day months, but the lunar month never has more than thirty days.
Lunar Calendars Were the Standard
The Jewish tradition that the Torah’s calendar was lunar finds support, finally, from the mere fact of the Torah’s silence. This silence is inherently significant. As far as we know, in the second-first millennia B.C.E., the calendars of the whole of the Ancient Near East, from (pre-Achaemenid) Persia and Mesopotamia to the Levant, were all lunar. Perhaps it is precisely because the lunar calendar was the norm in Near Eastern culture that the Torah does not bother to mention it.
TheTorah.com is a 501(c)(3) nonprofit organization.
We rely on the support of readers like you. Please support us.
April 3, 2019
November 22, 2020
Prof. Sacha Stern is Professor of Jewish Studies at University College London, where he heads the Department of Hebrew and Jewish Studies, and is the Principal Investigator for the Fritz Thyssen Foundation research project “Qaraite and Rabbanite calendars.” He holds an M.A. in Social Anthropology from UCL (1988), and a D.Phil. in Jewish Studies from Oxford. Stern is the editor of the Journal of Jewish Studies and among his many publications, he is the author of Calendar and Community: A History of the Jewish Calendar, 2nd Century BCE to 10th Century CE (2001), Time and Process in Ancient Judaism (2003), and Calendars in Antiquity: Empires, States, and Societies (2012).
Essays on Related Topics:
Previous in the Series
Next in the Series