(Permission is granted to print and redistribute this material
as long as this header and the footer at the end are included.)


brought to you by Kollel Iyun Hadaf of Har Nof
Rosh Kollel: Rav Mordecai Kornfeld

Ask A Question about the Daf

Previous daf

Rosh Hashanah 20

ROSH HASHANAH 19 & 20 (10, 11 Av) - dedicated by Rabbi Kornfeld's grandmother, Mrs. G. Turkel, to the memory of her husband, Reb Yisrael Shimon (Isi) ha'Levi Turkel, who loved Torah and worked to support Torah until his last breath. He passed away on 10 Av 5780.


QUESTION: The Gemara says that when Beis Din has a need to declare the new month on the thirtieth day since the last Rosh Chodesh (making the previous month a 29 day month), such as in order to prevent Shabbos and Yom Kipur from being on two consecutive days, they may coerce "the witnesses" who did not see the new moon to say that they saw the new moon.

If the witnesses did not see the new moon, then what are they doing in Beis Din? They are not witnesses if they did not see anything!


(a) RASHI implies that there indeed are witnesses who did not see anything. Apparently, people would volunteer in advance to come to Beis Din and serve as witnesses whenever Beis Din needed testimony from witnesses concerning Kidush ha'Chodesh. These people were not actually "coerced" by Beis Din to say something, but rather they said on their own what Beis Din needed to hear. The Gemara calls this "Me'Aymim Es ha'Edim," intimidating the witnesses, since the witnesses testified only upon the court's request.

(b) The RAMBAM (Hilchos Kidush ha'Chodesh 3:19) maintains that the reason Beis Din coerces the witnesses is not related to the considerations of preventing Shabbos and Yom Kipur from occurring consecutively. For such concerns, Beis Din does not interfere with the testimony of witnesses. Rather, the reason Beis Din would coerce witnesses to testify that they saw the new moon is in order to prevent the uncomfortable situation of losing witnesses who meant to testify about the new moon. The case when Beis Din coerces witnesses is when the witnesses came on their own accord to testify that they saw the new moon, but during the interrogation, Beis Din finds the testimony of the witnesses to be incongruent, either with each other's testimony or with the actual astronomical calculation of the appearance of the new moon. If the Beis Din must make the month Me'ubar because no witnesses could be found to testify about the new moon, it would be a "disgrace" (according to the popular Girsa in the Rambam). Therefore, Beis Din may coerce the witnesses to say the proper testimony and not say something incorrect, so that Beis Din can then declare the new month. Another case is when a second set of witnesses (Edim Zomemim) come to discredit the ability of the first set of witnesses to testify. The Beis Din may coerce those witnesses not to testify against the first set, in order that they may be Mekadesh the Chodesh on the 30th day of the month.

According to the Rambam's explanation, Beis Din is not telling the witnesses what they should say, but rather Beis Din is telling the witnesses what *not* to say.

According to the Rambam, though, what does Beis Din do if they see the need to make the month a 29-day month (so that Yom Kipur and Shabbos are not in proximity of each other), and no witnesses come to court? If there are no witnesses already in court, who will Beis Din coerce in order to proclaim the new month?

The CHAFETZ CHAIM (Ein Mishpat #2 in LIKUTEI HALACHOS) says that the Rambam holds like Rabeinu Chananel's explanation, that adjusting the month so that Yom Kipur does not fall next to Shabbos is not a serious concern of Beis Din. If so, Beis Din did not need to have witnesses testify when they saw that Yom Kipur would fall next to Shabbos.

(c) RABEINU CHANANEL says that the Gemara is discussing a case where there are witnesses who said that they saw the moon, but it was unclear and it might have been a cloud or some other object. In such a case, the Beis Din tells the witnesses to disregard the possibility that it was a cloud. Beis Din instructs them to say with confidence that they saw the new moon. According to Rabeinu Chananel, then, Beis Din finds witnesses who came to testify about something that they saw in the sky, but are not fully certain that it was the moon.

QUESTION: The Gemara states that Beis Din has the right to intimidate witnesses to say that they saw the new moon even when they did not see it (see previous Insight). How can the Beis Din instruct people to lie? The Torah clearly prohibits lying, as it says, "mi'Devar Sheker Tirchak" (Shemos 23:7). Furthermore, there is a negative prohibition against giving false testimony (Shemos 20:13). How can Beis Din instruct the witnesses to transgress an Isur?


(a) According to the Rambam cited above (see previous Insight), Beis Din does not tell the witnesses to say that they something that they did not see. Rather, Beis Din tells the witnesses only *not* to say something that would ruin their testimony.

(b) According to the other Rishonim, such as Rashi, who understand that Beis Din actually tells the witnesses to lie, in this case there is no Isur of testifying falsely, because this testimony is not about one's fellow man. The Isur against testifying falsely applies only when a witness testifies against a person, to cause him harm, as it says, "Lo Sa'aneh b'Rei'acha Ed Shaker" --"Do not bear false testimony against your fellow man" (Shemos 20:13). (For the same reason, there is no requirement that the testimony of the new moon be "Edus she'Atah Yachol l'Hazimah.") (GILYONEI HA'SHAS)

Similarly, the TOSFOS HA'ROSH (Kesuvos 32a, DH sh'Ken Yesh) writes that it is permissible to accept false testimony for declaring the new month, because the Torah never prohibited false testimony in the case of Kidush ha'Chodesh. The source for this is the verse in the Torah (Vayikra 23) from which the Gemara (25a) learns that the Beis Din's declaration of the new month is valid and binding "even if it was done purposefully in error." Without testimony the court cannot declare a new month, but when there exists testimony for the new moon -- even it is false testimony -- they can declare a new month.

Regarding the Mitzvas Aseh to stay away from lies, that Mitzvah applies only when the lie will mislead someone or cause an undesirable outcome. If the outcome will be positive, though, and Beis Din sees that it is necessary, then it is not the type of lie that the Torah requires one to avoid. This is implied by the context of the verse (Shemos 23:7) which commands to stay away from lies -- the verse is talking about killing a person as a result of the lie.

Support for this view can be brought from the Gemara in Yevamos (65b) which says that it is "Mutar l'Shanos Mipnei ha'Shalom" -- "it is permitted to alter the truth for the sake of peace" (see also Bava Metzia 23b and Kesuvos 17a). If a lie cannot cause anyone harm, and can only benefit them, it is not prohibited by the Torah.


QUESTIONS: Rebbi Zeira makes two important statements regarding the new moon. First, he teaches that when the Beraisa of "Sod ha'Ibur" (the secret wisdom of lunar astronomy) says cryptically, "The Molad occurs before Chatzos, or the Molad occurs after Chatzos," it means that there is a difference between when the Molad occurs before or after Chatzos (midday). Rebbi Zeira explains that when the Molad occurs before Chatzos, the new moon can be seen right after sunset of the same day. When the Molad occurs after Chatzos, it is not possible to see the moon after sunset of that day. In addition, Rebbi Zeira says in the name of Rav Nachman that the moon fully disappears from view for a total of 24 hours at the time of the Molad: for people "here" (Rashi: in Bavel), the moon is not visible for 6 hours before the Molad and for 18 hours after the Molad, and for people "there" (in Eretz Yisrael), the moon is not visible for 18 hours before the Molad and for 6 hours after the Molad.

RASHI explains the statements of Rebbi Zeira in considerable detail, attributing his explanation to Rav Sa'adyah. Although Rashi explains this statements at length, the astronomy behind his explanations is very unclear.

(a) What does Rashi mean when he says (DH 24 Sha'os) that the people standing in the east can see the moon better when it is in the east, and the people standing in the west can see it better when it is in the west? The moon moves across the sky, and people in both the east and the west see it in both the east and west parts of the sky! Besides, living further east or west cannot possibly bring a person "closer" to the moon. No place on earth is significantly closer to the moon than any other due to the great distance between the earth and the moon (~250,000 miles)!

(b) What does Rashi mean when he says, repeatedly (DH Nolad, DH Kaf-Dalet, DH Mechasya, and again on 24a DH Kan), that at the time of the new moon, the moon *always* rises in the south-east and sets in the south-west? That is not true -- it depends on what season of the year it is! The sun and the new moon are always in close proximity. In the summer (in the Northern Hemisphere), the sun and new moon rise *north* of the midpoint of the eastern horizon, traveling in a curve across the sky, first towards the south, and then curving around (at midday) towards the north again, and then setting north of the midpoint of the western horizon. In the winter, the sun and new moon rise to the *south* of the midpoint of the eastern horizon, traveling again in a southerly curve across the sky and setting to the south of the midpoint of the western horizon. Why, then, does Rashi say that the moon *always* rises in the south-east and *always* sets in the south-west? (See also Rashi in Yoma 62b, DH Al Taba'as, and RASHASH there who asks a similar question.)

(c) Rashi writes that some people are able to see the moon six hours before or after the Molad. (People in the east can see the old, waning moon six hours before the Molad occurs, and people in the west can see the new, waxing moon six hours after the Molad occurs). The Rishonim ask that this is astronomically impossible! The moon cannot be seen for at least 18 hours after the time of the Molad! (Rambam, Hilchos Kidush ha'Chodesh 17:3 -- the impossibility of the six hour limit of visibility that Rashi describes has been confirmed by modern day astronomers.)

(a) HAGAHOS BEN ARYEH, as well as Hagaon Rav Yosef Eliyahu Henkin, zt'l, in LEV IVRA (p. 44,45), propose an ingenious solution that offers a simple, astronomically correct rationale for differentiating between the people of the west and the people of the east. (Hagahos Ben Aryeh were written by Rav Zev Lipkin, Rosh Beis Din of Telz and father of Hagaon Rav Yisrael Salanter -- the explanation referred to actually appears as a bracketed insertion in the Ben Aryeh, and it is not clear who added it.) The Ben Aryeh's explanation is explained in greater detail, with additional clarifications, in MAGID HARAKI'A (Rav Hasgal, presently in Kiryat Sefer, Israel) and KUNTRUS KAF-DALED SHA'OS (Dr. Nisim Vidal, former chief astronomer of the Royal Greenwich Observatory, and visiting professor at the Harvard University Center for Astrophysics), who explain at length the astronomical principles behind this explanation. We will present their explanation here in brief, based on three introductory remarks.
1. Although the sun and moon both travel in an east-west trajectory from our perspective (caused by the rotation of the earth), rising daily in the east and setting in the west, they also have a motion relative to each other. The moon travels slightly slower than the sun, constantly falling behind it more and more to the east. As the month progresses, the moon gets progressively farther away from the sun in the sky until passes the midway point, after which it starts to approach the sun from the other side, the west. Eventually it "catches up" with the sun, traveling west to east relative to the sun, and passes the sun in an easterly direction. This is what causes the changes in the appearance of the moon, as follows.

At the beginning of the month, at the time of the Molad, the moon and sun are at the same point in the sky (actually, the Molad is the point immediately after conjunction, or "Kibutz;" conjunction is when the moon and sun are at the same point). At that point, we cannot see the moon at all (it is between the sun and the earth, and all of the light of the sun that it reflects is on the side of the moon that faces *away* from the earth). As the days progress, the moon lags behind the sun, so that more of it becomes visible. A day or two after the Molad, we can see the moon "behind" (to the east of) the sun. Fifteen days after the Molad, the moon has lagged so far behind in the sky that it is now on the opposite side of the sky from the sun (that is, the earth is between the moon and the sun), and thus at night we are able to see the entire lit face of the moon (a full moon). As more days pass (the second half of the month), the moon's lag causes it to get *nearer* to the sun from the *other* direction (the direction in which the sun is traveling), so that when we look into the sky, we see the moon ahead (to the west of) the sun. Finally, at the end of the month, its lag causes it to be equal again with the sun, and the next Molad occurs. (See, or ask for, our Graphics section)

The moment before the moon passes through the sun (before the Molad), it is immediately to the west of the sun. The moment after it passes through the sun (after the Molad), it is immediately to the east of the sun.

2. There are a number of ways of expressing the changing distance between the sun and the moon. Mathematically, it can be expressed as a matter of degrees around a circle (since the sun and moon travel around the earth in a circle, for our perspective). When the moon is on the opposite side of the sky from the sun (such as in the middle of the month, when the moon sets in the west at the time that the sun rises in the east), it is 180 degrees away from the sun. When the moon is one-quarter circuit away from the sun, it is 90 degrees from the sun, etc.

This distance can also be expressed in terms of the number of days or hours that have passed from the time of the Molad. This amount of time expresses how many hours or days it has taken for the moon to lag as much as it has behind the sun. For example, when we say that the moon and sun are fifteen days (half a month) apart, that means that the distance between the moon and sun is whatever distance the moon lags in fifteen days. (As we have seen, in terms of degrees, this means that the moon is 180 degrees away from the sun).

Since the moon travels 360 degrees from the sun (i.e. it meets the sun again after completing an entire circuit) in approximately 30 days (a lunar month), it travels 12 degrees in one day, or half a degree in one hour. When we say that the moon is one day (24 hours) away from the sun, that means it is 12 degrees away (because in one day, the distance between the moon and the sun increases by 12 degrees).

(NOTE: The distance *between the sun and the moon* as expressed by the amount of time that has passed since the Molad, as we have explained, is not to be confused with the distance that the *moon travels around the earth* each day. Since it goes around the earth (360 degrees) once a day, it travels 15 degrees every hour, or one degree every four minutes. However, the sun also travels about the same number of degrees around the earth in that amount of time, so the moon has not distanced itself from the *sun* during that time.)

3. The moon is not always visible. When it is close to the sun, it cannot be seen because the sun is so bright and the moon is so dim relative to the sun. How far away from the sun must the moon be in order to be visible? (In other words, what is the earliest time after the Molad that the moon can be seen, under the most favorable conditions?)

Rashi asserts that when the moon has lagged behind the sun for *6 hours* after the Molad and is *3 degrees* away from the sun, the moon can be seen, because the light of the sun is not strong enough to obstruct its visibility. This means that both 6 hours before and 6 hours after the Molad the moon may be visible, and in the twelve hours in between the moon will never be visible.

However, there is another factor which hides the moon from view: the rotation of the earth. Around the time of the Molad, since the moon is so close to the sun it rises and sets only shortly before or after the sun does. Throughout most of the night it is on the other side from the earth (like the sun itself) and therefore it is hidden from view from our perspective.

With these words of introduction, the Gemara can be explained as follows.

When the Gemara discusses the 24 hours during which the moon cannot be seen, it is referring back to the first statement of Rebbi Zeira, "Nolad Kodem Chatzos...." Rebbi Zeira meant by this that if the Molad occurs immediately before midday, the new moon can be seen right after sunset the same evening, since six hours have passed and the moon has distanced itself from the sun sufficiently to see it before it sets (a few minutes after the sun sets). If the Molad occurs *after* midday [by more than about 12 minutes], the new moon cannot be seen that evening after sunset, since less than six hours have passed since the time of the Molad before the moon sets. The Gemara is discussing a 12 hour day (such as at the time of the equinox, when the length of the day and night are equal).

Since the Molad depends on the positioning of the moon relative to the sun (and not to a particular spot on earth), it occurs at the same instant in time the world over. For some places on earth, that instant will be in middle of the day (i.e. while they are facing directly towards the sun and the sun is overhead), while for others it will be in middle of the night, and for yet others it will be at the beginning of the day or the night. The specific case the Gemara is discussing (in Rebbi Zeira's second statement) is one in which the Molad is just before midday in Israel. For one who lives further east, for example in Bavel, the time of day in which the Molad occurs is not before midday, but shortly after midday (since the sun already passed overhead earlier in his more easterly time zone) , or about 12:30 PM.

(When we refer to different times, such as 11:59 in Israel and 12:30 in Bavel, we are not referring to the time as based on the standard time zones used today, but on actual sun time for each place. That is, if it is a 12 hour day, the sun will set in 6 hours from now in Israel and 5 1/2 hours from now in Bavel).)

Since in Bavel the Molad is half an hour after Chatzos, the moon will not be visible that evening (six hours will not have passed from the Molad before moonset, which is approximately 6:12 that evening, about twelve minutes after sunset). However, the Molad did occur more than six hours from *sunrise* that morning. Therefore, that morning the old moon was visible in the east, right before sunrise (i.e. to the immediate east of the sun), when the moon is three degrees away from the sun. The first time the people in Bavel will be able to see the new moon *after* the Molad is approximately 18 hours after the Molad, that is, when it rises again the morning after the Molad. This is what the Gemara means when it says that "for us [in Bavel], the old moon is covered for 6 hours, and the new moon is covered for 18 hours."

In contrast, in Eretz Yisrael, since the Molad occurred just *before* noon, the new moon *will* be visible just before sunset, 6 hours later (and it will remain visible until it sets, a few minutes after the sun). However, the *old* moon will not be visible in the morning, since it is within 6 hours of the Molad. The old moon will only be visible before sunset the evening before the Molad, when it is approximately 18 hours (9 degrees) away from the Molad. (The old moon will set *before* the sun, approximately 36 minutes before sunset). This is what the Gemara means when it says that "for them [in Eretz Yisrael], the old moon is covered for 18 hours and the new moon is covered for 6 hours."

All of the words of Rashi throughout the Sugya are easily understood based on this brilliant explanation.

(b) We asked, why does Rashi write that the new moon is first visible after the Molad "in the south-west corner of the sky" and the old moon is last visible before the Molad in the "south-east corner." It is true that if the Molad is close to midday, the new moon first appears in the west (i.e. when it is setting), and the old moon last appears in the east (i.e. when it is rising). But why does Rashi say that it is in the *south* part of west and the east? Rashi explains in a number of places (see 24a, DH Kan) that the sun does not always rise and set at the same place along the horizon. It moves along the horizon, rising and setting more to the south each day in the winter and more to the north each day in the summer. (Our discussion here, as well as in all the Sugyos dealing with astronomy, is limited to the northern hemisphere.) Since the moon orbits the earth the same way as the sun does for our perspective (i.e. on the ecliptic plane), shortly before or after the Molad, when it is very near to the sun, it should be seen setting approximately in the same place that the sun sets, and therefore its inclination to the north or south of the horizon should also vary according to the season. It should rise and set on the northern side of the horizon in the summer and the southern side in winter. Why does Rashi write that it is always in the south?

There are two possible ways of understanding the words of Rashi.

1. Professor Vidal explains as follows. As we mentioned earlier, when the Gemara said that "if the Molad occurs before Chatzos, the new moon will be seen before sunset," it must be talking about a 12 hour day, in which the sun sets exactly 6 hours after midday. Such a day occurs twice a year, on the vernal equinox and the autumnal equinox (Sept. 21 and March 21). At the equinox, the sun sets exactly at the midpoint of the western horizon, approaching the horizon at an angle from south to north. In such a situation, the new moon -- which is visible right before it sets -- will always be visible when it is slightly south of the midpoint. That is what Rashi means when he says that the new moon is in the "south-west" corner when it is first seen. The same is true for moonrise before the Molad; the moon will be seen rising slightly south of center of the eastern horizon.

However, it is not clear according to this explanation why Rashi writes that it is in the south-west corner "l'Olam" -- "always." Second, why is that called the south-west "corner" ("Keren")? The term "corner" implies, as Rashi himself later (24a) says, the *furthest* point to the south at which the sun sets, and not just slightly south of center. Third, Rashi cites a source for his statement that the moon is always in the south-west -- the Gemara later (24a). Rashi there (DH Kan) repeats this assertion that the new moon is always in the south-west, and he clearly states that this applies both in the summer and in the winter months!

2. A second possible interpretation of Rashi could be that "the south-west corner" does not refer to the south-west of the world, but rather it is a term describing the moon's location relative to the sun's location. Whenever the new moon is seen in the west (close to sunset) it is to the south of the sun. Since the sun is traveling from south to north as it sets in the west, the new moon --which is further to the east on the same orbit, setting after the sun -- is always to the south of the sun. Similarly, the old moon, when seen over the eastern horizon, is always to the south of the sun, because it rises before the sun and is thus ahead of the sun, traveling towards the south.

According to this interpretation, when Rashi uses the term "south-west" with regard to the moon both here and on Daf 24a it is not consistent with the way he uses the term "south-west corner" with regard to the sun (on 24a), because there it clearly means that the sun is south of the *horizon*. (See Insights to 24a.)

(c) We asked that astronomically, it is impossible to see the new moon less than 18 hours after the Molad (Rambam, Hilchos Kidush ha'Chodesh 17:3; see also Ba'al ha'Me'or here -- the same applies to seeing the old moon six hours before the Molad).
1. Dr. Vidal suggests that although it is true that under normal circumstances it cannot be seen before 18 hours, nevertheless, under perfect viewing and atmospheric conditions, it is possible that one who knows exactly where and when to look will be able to see the moon earlier.

Professor Vidal points out (Kuntras Kaf-Daled Sha'os, p. 9) that almost every year, the record is broken for the earliest time that the moon is seen after the Molad. In Teves of 5757, the new moon was seen by the naked eye only 14 hours after the Molad (and with a telescope, 12 hours after the Molad). Rashi is saying that the Chachamim had a tradition as a Halachah l'Moshe mi'Sinai that the very earliest that the new moon could possibly be seen is six hours after the Molad, even under the most ideal conditions. That is how Rashi understands Rebbi Zeira's statement in the name of Rav Nachman. It is not a statement of the average time of visibility of the new moon, but rather it is a statement of the extreme limit of how early the new moon can be seen, so that we can prove witnesses to be liars if they claim to have seen it earlier.

2. Alternatively, it may be suggested that according to Rashi, the Molad referred to in our Gemara is not the true astronomical Molad. Rather, it is the "average Molad" that the Rambam describes in the beginning of Hilchos Kidush ha'Chodesh (based on which we reckon each lunar circuit as 29.5 days and 793 parts of an hour). This Molad does not take into account inconsistencies in the speeds of the earth and the moon at different times in the month or year. Although these inconsistencies balance out over the course of the year, depending on the month they can cause the true Molad to actually occur six to 14 hours before or after the average Molad. Since we are trying to disprove the testimony of witnesses, we cannot ignore the possibility that our calculation of the average Molad was not the time of the true Molad, and the Molad may have occurred 14 hours earlier. Therefore, we believe the witnesses as long as they claim to have seen the new moon at least six hours after the Molad (which could actually be 20 hours after the Molad, taking the variations in orbital speeds into account). It is entirely reasonable for the new moon to be seen that long after the true Molad. (M. Kornfeld)

Next daf


For further information on
subscriptions, archives and sponsorships,
contact Kollel Iyun Hadaf,