Bar-Ilan University's Parashat Hashavua Study Center

Parashat Beshalah 5762/ January 26

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.
Prepared for Internet Publication by the Center for IT & IS Staff at Bar-Ilan University.
Inquiries and comments to: Dr. Isaac Gottlieb, Department of Bible,

Parashat Beshalah 5762/ January 26, 2002
Tu Bishvat (15 Shevat/January 28)

When Have "Most of the Rains Passed"?
Setting the Date of the New Year for Trees

Prof. Yair Goldreich
Department of Geography

The schools of Shammai and Hillel disagreed about the New Year for Trees, the former maintaining that it fell on the first of Shevat, and the latter on the fifteenth. The Talmud and its commentaries contain various botanical and climatic arguments concerning the significance of the New Year for Trees. In this article we shall deal with climatic factors which help to determine the date.

The Babylonian Talmud (Rosh ha-Shanah 14a) presents the following argument by the school of Shammai: "Rabbi Eleazar said, quoting R. Ushia, [we celebrate the New Year for the trees on the first] 'since most of the rains of the year are passed,'" adding, "even though most of the season of Tevet [winter] is still ahead of us." Tosafot and Rabbenu Hananel interpreted Rabbi Eleazar's remark as also pertaining to the stand taken by the school of Hillel. The Jerusalem Talmud presents a slightly different version of the determining factor: "Most of the rain of the entire year is done, and most of the season is over." The season referred to is that of Tevet or winter, extending from the shortest day (December 21) to the spring equinox (March 21). Since the Talmud uses the term 'season' (Heb. tekufah), and the seasons are based on the solar calendar, we must translate the Hebrew date, based on a lunar calendar, to the date used in the secular solar-based calendar.

If we take the average secular dates that correspond to the first and fifteenth of Shevat for the period of the current 19-year cycle of the Hebrew calendar (cycle 303), we come up with the dates of January 16 and 30, respectively. [Editor's note: Prof. Goldreich wrote this study in 1995.] The earliest these Hebrew dates fall with respect to the secular calendar is the 2nd and 16th of January (in 1995), and the latest (in 1987), January 31 and February 14, respectively. In other words, there is a maximal difference of one month. Taking the median dates instead of the average dates yields the same results in the secular calendar.

What is meant by "most of the rains have passed"

Rashi interpreted Rabbi Eleazar's remarks in the Babylonian Talmud (loc. sit.) as meaning that "most of the rainy days have passed, i.e., it is the time of revi'ah[1] when the sap begins to flow in the trees, and fruit begins to form," even though most of the winter (the season of Tevet) lies ahead. Rabbenu Hananel did not relate to the quantity of rain or to the number of actual days of rain, but to the number of days in the rainy season. He divided the rainy season into the period from the 17th of Heshvan (the time of the middle rains according to Rabbi Judah, Taanit 6a) until the fifteenth of Shevat, this period being three months minus three days. This period has more than the remaining days of the rainy season from the fifteenth of Shevat until the month of Nissan (in part). He adds, "One does not pay attention to the season" (Rosh ha-Shanah, loc. sit.).

There are ostensibly grounds for assuming that the above dates were chosen on the basis of measurement of the amount of precipitation, and that the difference of opinion between the schools of Hillel and Shammai is due to the fact that the mid-point of rainy season is not the same in all parts of the country. The Jerusalem Talmud says: "How much precipitation constitutes a period of rainfall (Heb. revi'ah)? According to Rabbi Meir, maleh keli mahazik shelosha tefahim [commonly thought to mean "a bucket-full containing three tefahim"]. Rabbi Judah says: the first period of rainfall, one tefah; the second, two tefahim; and the third, three tefahim" (Jerusalem Talmud, Taanit 10.3).

According to current units of measure, three tefahim is equal to 250mm. Could it be thar R. Meir, who lived in Beth Shean, held that the first period of rainfall is almost equal to the total average annual precipitation in this city (300mm.)? It would rather appear that the quantity of rain (the depth of water accumulated in a measuring vessel) was not the issue. Rather, it is as Prof. Judah Felix explains at length (in Ha-Haklaut be-Erez Israel be-Tekufat ha-Mishnah ve-ha-Talmud, Jerusalem-Tel Aviv, p. 345), citing tractate Taanit 55b: "How much rain has to fall for the public to cease fasting? Until it saturates [the soil] as far as the bend in the plow, according to Rabbi Meir. But the Sages say on parched soil, one tefah; on intermediate, two tefahim; and on tilled soil (which absorbs more rain), three tefahim." Thus, the word keli in the Jerusalem Talmud appears not to refer to a vessel for measuring rainfall, but to a tool-a plow blade. In other words, the phrase indicates a measurement of the dampness of the soil. Maleh keli, or perhaps "a full length of the tool," indicates that the end of the plow penetrates the soil as far as the cross bar that connects the blade and runs parallel to the surface of the soil.

Rabbi Menahem ha-Meiri (Beit ha-Behirah, Rosh ha-Shanah, beginning of ch. 1) relates more to the end of Rabbi Eleazar's remark: the fifteenth of Shevat falls in the middle of the season (according to the school of Hillel), and since half of the season of Tevet has passed, its strength is spent and the cold is not so bitter; so formation of fruits begins. In other words, the New year for Trees is not merely a question of plants having received sufficient water, but also of the coldest weather having passed and the warming of the season causing fruits to begin to the form, insofar as the main concern of the New Year for Trees is with the second tithe (ma'aser sheni) and orlah.

Another interesting explanation of the difference of opinion between the schools of Shammai and Hillel rests on social factors. According to the talmudist Prof. Louis Ginzberg (Al Halakhah ve-Aggadah, Devir 1960, p. 36; cf. also the reservations expressed by Felix on Ginzberg's ideas, Tarbiz 40, 1977, pp. 181-211), Beit Shammai represented the wealthy class, whose superior orchards began to form fruit two weeks earlier than the trees of farmers with poorer fields.

There are two methods for computing the mid-point of the rainy season in terms of quantity of precipitation: one is to take monthly quantities over many years, the other is to take the weighted average of rain for each month using harmonic analysis.

Comparing the results of both methods shows much similarity: in the hilly regions, the middle of the rainy season falls between January 20-25. The middle of the season can also be computed in terms of the number of days of rain, instead of the amount of rain. For computing the middle of the season in terms of number of days of rain (see table), we took data for the decade from 1938/9 - 1947/8 (Memuzaim Aklimiim Rav- Shnatiim [Cumulative Climatic Averages], Part I, (Rain), Ser. 1, 1948, Israel Meteorological Service, Ministry of Transport), this period not being identical to the period of the data used for computing the mid-season in terms of quantity of precipitation. A day of rain in this case was defined as a day on which there was at least 1mm of rain in a 24-hour period, from 8:00 to 8:00 according to the local time (in contrast to the standard definition of a day of rain, defined as one in which there was at least 0.1mm of rain, which is a less significant quantity with respect to the subject at hand).

Mid-Season According to Days of Rain at Various Weather Stations

Name of Station Median Date
Tel Aviv, Reading 15/1
Tel Aviv, Nahmani 16/1
Petah Tikva 21/1
Lod 21/1
Beit Jamal 26/1
Jerusalem 27/1
Beer Sheba 18/1
Jericho 27/1

If we compare these figures, based on days of rain, with the mid-season computed in terms of quantity of rain (using both methods), we observe that the mid-season in terms of days of rain falls considerably later in the coastal plain (10-15 days), and in the hilly areas only somewhat later (approx. 5 days).

Reconstructing the Rainy Season in the Time of the Tannaim

To draw inferences about the middle of the rainy season in the time of the tannaim from the data on the middle of the rainy season in our day, computed by various methods, we must first clarify two points: 1) Was the climate in the time of the tannaim similar to our climate today? 2) When did the New Year for Trees fall then in relationship to the solar year?

Reconstructing the climate of the past involves scholarship in a variety of disciplines, ranging from historical and archaeological evidence, to botanical, zoological and geomorphological arguments. Summing up the studies in these areas, we find that the quantity of precipitation (apparently also in terms of the course of the season) was similar then to now. This was not the case, however, with respect to the date of the New Year for Trees. In the time of the schools of Hillel and Shammai the Sanhedrin essentially determined this date (by scheduling leap years). Since the time when the calendar was fixed we have essentially been following the method of Rav Ada, setting leap years in a cycle of 19 years, so that the average length of a solar year is 365 days, 5 hours, 55 minutes and 25 seconds. The length of a tropic solar year[2] is 365 days, 5 hours, 48 minutes and 46 seconds. In contrast, the length of a solar year according to the system of Rabbi Samuel was taken to be exactly 365 days and 6 hours, similar to the length of a year as set in the Julian calendar by Emperor Julius Caesar in 46 C.E. Therefore, the calendar year being longer than a tropic year, we are about 13 days late in our era. According to Rav Ada's calendar, however, the cumulative discrepancy since the time of the tannaim would only be about seven days.

If we compute what must have been the secular date of the New Year for Trees in the time of Hillel and Shammai, assuming that the system developed by Rav Ada represents the natural continuation of the Sanhedrin's proclamations of the new months (except for the order of leap years), we must subtract about seven days from the date today. In other words, the New Year for Trees came out on the average to be January 9 according to the school of Shammai, and January 23 according to Hillel. If we compare these dates with the mid-point of the rainy season around Jerusalem (which yields the latest dates for any part of Israel) we find that on the average the 15th of Shevat comes out two days before the middle of the rainy season when computed by taking the mean precipitation, exactly in the middle when computed by harmonic analysis, and four days before the middle when computed in terms of half of the days of rain. Regarding most parts of Israel, however, it turns out that of all the approaches to determining when the New Year for Trees should be, including that of Rabbenu Hananel which takes into account the entire length of the rainy season and not the number of days of rain, the preferred approach is that of Hillel, namely that on the 15th of Shevat "most of the rains have passed" even though most of the winter season still lies ahead. Indeed, that is how the Halakhah ruled.

[1] Talmudic literature speaks of the rainy season having three periods of revi'ah or initial rainfall moistening the parched soil and making it arable: the first revi'ah being approximately the first two weeks of Heshvan, second being the third week of Heshvan and third being the last week of Heshvan or the beginning of Kislev. (Translator's note.)
[2] A tropic solar year is defined as the period elapsing from the time at which the sun crosses the equator one spring to the time it crosses the equator the following spring. A tropic solar year is approximately 20 minutes shorter than a sidereal year, which is the time in which the earth make a 360-degree revolution round the sun (with respect to the stars). The difference is due to precession, or the rotational movement of the earth's axis in relationship to the stars.