Skip to main content

Conclusion of Shabbat and Havdala

Rav Doniel Schreiber
Text file



By Rav Doniel Schreiber



Shiur #13: Conclusion of Shabbat and Havdala



I. Havdala and Melakha


            Shabbat ends with the onset of nightfall, since, according to the Torah, the new day begins at night.  (See Bereishit 1:5, Mishna and Gemara Chullin 83a, and Rashi ibid. s.v. Ma'aseh Bereishit; see also Pesachim 2a.)  Thus, in theory, according to Torah law (min ha-Torah) one should be permitted to perform melakha immediately at nightfall when Shabbat ends (MB 293:3 and 299:38).  In practice, however, this is not the case.  Even after nightfall one may not perform any type of melakha, be it a melakha forbidden by the Torah or by the Rabbis, until one has recited havdala (OC 299:10 and MB 299:32).


            Most poskim understand that this is a rabbinic prohibition based upon either one of two reasons:

1.  Prior to reciting havdala a partial (i.e. rabbinic) sanctity of the Shabbat prevails even after nightfall.  This creates the opportunity of escorting the king through havdala (MB 299:33,34 and Tzitz Eliezer v. 11, no. 34).

2.  The rabbis forbade melakha in order to remind one to fulfill mitzvat havdala (Arukh Ha-shulchan OC 299:21).


There is one opinion [Eshel Avraham (Butshetsh) mahadura tinyana 299:10] who believes that it is forbidden according to Torah law to perform melakha prior to havdala.  He understands that tosefet Shabbat min ha-Torah occurs automatically after nightfall.


            Before discussing the laws of havdala, it is necessary to address the issues concerning the onset of nightfall.


II. Determining Nightfall


1. Appearance of Stars


            All poskim agree that nightfall is determined by the appearance of three medium-sized stars in the evening sky (Bi'ur Halakha 261 s.v. She-hu and Bi'ur Halakha 293 s.v. Gimmel kokhavim ketanim).  [For further research, see Shabbat 35b; compare with Megilla 20b, Pesachim 2a, and Berakhot 2a-b.  See also Nechemia 4:15, Rambam Shabbat 5:4 (compare with Rambam Terumot 7:2) and Shulchan Arukh OC 235:1.].


However, in practice, we consider it to be night only when three SMALL stars have emerged.  This added stringency is because we are not experts in evaluating what is a "medium-sized" star; thus, we wait longer until even the small-sized stars have appeared.  In this way, we are certain that nightfall has arrived (OC 293:2, MB 293:3, and Bi'ur Halakha 293 s.v. Gimmel kokhavim ketanim).  Hence, one may not perform melakha prior to this time since it may still be daytime (MB 293:3).


2. Calculating Nightfall


A. Contradictory Gemarot


            Nonetheless, Rishonim and poskim have calculated the time at which three MEDIUM-sized stars emerge.  Two contradictory gemarot lie at the core of this calculation.  The gemara Pesachim (94a) states that the stars appear four "mil" (i.e., the time it takes to walk four mil) after sunset.  In contrast, the gemara Shabbat (35a) indicates that the period from sunset to nightfall is 3/4 of a "mil."  (See Tosafot ibid. s.v. Trei.)


It is unclear from the gemara how long a period a "mil" actually is.  See Shiur #3, MB 459:15, Bi'ur Halakha ibid. s.v. Havi, Bi'ur Ha'gra ibid. s.v. Ve-shiur, and Chok Ya'akov (ibid. s.v. note 10).  The view most widely accepted is that a mil can be walked in 18 minutes.  With this number in mind, the time of four mil between sunset and the appearance of the stars is 72 minutes, whereas the time of 3/4 of a mil between sunset and nightfall is 13 and 1/2 minutes.  How can these two contradictory conceptions concerning the timing of nightfall be reconciled?


B. Rabbeinu Tam


            Rishonim and poskim offer different resolutions to the contradiction in the gemarot.  Rabbeinu Tam (Tosafot, Shabbat 35a s.v. Trei, and Pesachim 94a s.v. R. Yehuda) asserts that there are two sunsets.  In the first stage, the sun sets, disappearing from our view; this is called "the beginning of sunset."  This phase lasts the time it takes to walk three-and-a quarter mil (i.e., 58 and 1/2 minutes) and this is still considered daytime (see Tosafot Menachot 20b, s.v. Nifsal).  In the next phase, a second "sunset" occurs when all the light in the sky disappears; this is called "the end of sunset."  This phase lasts the time it takes to walk three-quarters of a mil (i.e., 13 and 1/2 minutes) and is considered "bein ha-shemashot" (a period of time in which it is doubtful whether it is day or night).


            At the conclusion of the second phase (i.e., bein ha-shemashot), three medium-sized stars appear, indicating nightfall.  Thus, according to Rabbeinu Tam, the total elapsed time between sheki'a and nightfall is 72 minutes.  In this fashion, R. Tam resolves any contradiction between the two gemarot.  The gemara Pesachim is referring to both stages of sunset while the gemara Shabbat is referring only to the second stage.  See MB 261:20,23, and Bi'ur Halakha s.v. Mi-tchila and s.v. She-hu.


            It is interesting to note that while the opinion of Rabbeinu Tam leads to an extra stringency at the conclusion of Shabbat, it also theoretically leads to a leniency at the onset of Shabbat.  According to R. Tam's calculation, bein ha-shemashot begins much later, therefore one would in theory be allowed to perform melakha much later on Friday than according to other opinions.  In practice, however, we do not rule this way.  We are stringent regarding the onset of Shabbat on Friday and one must be very careful not to do any melakha after sheki'a, i.e. once the sun can no longer be seen.  See MB and Bi'ur Halakha ibid.]


C. Vilna Gaon


            However, this position of Rabbeinu Tam is disputed by many Ge'onim, Rishonim and poskim, including the Vilna Gaon (Bi'ur Ha-gra, OC 261:2, s.v. She-hu).  According to them, only one stage of sunset exists.  They rule that the disappearance of the sun from the horizon (i.e. sheki'a) marks the onset of bein ha-shemashot.  According to this position, bein ha-shemashot lasts the time it takes to walk three-quarters of a mil (i.e., 13 and 1/2 minutes).  At the conclusion of this time period, three medium-sized stars appear; this is nightfall min ha-Torah for all purposes.  This is the position reflected by the gemara Shabbat.  The gemara Pesachim, however, which cites a four-mil period, is relevant for other purposes; it refers to the time between sunset and a later time at night when all the stars appear.  See also MB ibid. and 233:14.


D. Accounting for Change in Latitude and Time of Year


            Nevertheless, according to many poskim, whether one understands that there is one sheki'a or two, all agree that the bein ha-shemashot period of three-quarters of a mil (i.e., 13 and 1/2 minutes) varies according to the time of year and one's location on the planet.  This 13 and 1/2 minute bein ha-shemashot time period was mentioned in the Talmud only for the latitude of Babylonia during the months of Tishrei and Nissan when days and nights are equally long.  European lands, however, which are closer to the North Pole, have a much longer time period of bein ha-shemashot.  Therefore, one should be very particular not to perform melakha after Shabbat, even if much time has passed since sunset, until three small stars appear.  Bi'ur Ha-gra ibid., MB 261:23 and Bi'ur Halakha s.v. Mi-tchilat.


E. Supplemental Indicators


            Acharonim point out that other conditions must also exist for three small stars to indicate nightfall.  According to the Likutei Pri Chadash and R. Akiva Eiger, the sky must cease to glow with a light resembling daylight.  However, the Vilna Gaon writes that even if the sky is aglow, three small stars indicate nightfall as long as the sun has set and all redness has disappeared from the sky (see SSK vol. 2, 58:2 who rules that it is proper to conduct oneself according to this view).  MB 293:4.  For further research, see Shiurim Le-zekher Abba Mari z"l, vol. 1, by Maran Rabbi Joseph B. Halevi Soloveitchik, pp. 91-112).


            The Tif'eret Yisrael has a different condition.  He rules that three small stars indicate nightfall only when one can also discern three medium-sized stars.  This may be to ensure that we have a point of comparison to determine that the small stars are indeed the small-sized variety and not large ones.  See MB ibid.


III. Tosefet Shabbat


            In addition, we are stringent to refrain from melakha until three CLOSELY-SPACED small stars become visible (OC 293:2).  This is due to the requirement of tosefet shabbat that part of the weekday be annexed to Shabbat (see shiur #3).  By waiting for these stars to appear one has automatically added time onto Shabbat (MB 293:5).  However, if one sees three small stars that are not closely spaced and then waits briefly afterwards, one has also fulfilled the mitzva of tosefet shabbat (MB ibid.).  There is an opinion that one should refrain from melakha until the sky is filled with stars to satisfy all opinions with regard to tosefet shabbat (SSK vol. 2, 58:2, note 8).


IV. When does Shabbat end in practice?


            The position commonly followed with regard to the end of Shabbat is the Vilna Gaon's.  Taking into consideration the change in latitude, the varying opinions with regard to a "mil," and the mitzva of tosefet shabbat, it is widely accepted that Shabbat ends in the New York area, for example, about 42 minutes after sunset, while in the Jerusalem area about 40 minutes after sunset.  There are other opinions, however, in this matter.  For instance, Moreinu Ha-rav Aharon Lichtenstein shlita informed me that the Rav zt"l, Maran Rabbi Joseph B. Halevi Soloveitchik, ruled that Shabbat ends in the New York and Boston area 30 minutes after sunset.  [The Rav zt"l, though, was personally stringent in accordance with Rabbeinu Tam.]  Others require 50 minutes while some suggest even 60 minutes.


            Some poskim are even more stringent and rule that we must take into account the position of Rabbeinu Tam that true nightfall does not occur until 72 minutes after sheki'a (sunset).  According to this view, one may not perform melakha until 72 minutes after sheki'a.  [See OC 261:2, MB 261:20,23, and Bi'ur Halakha s.v. Mi-tchila and s.v. She-hu.]  Whether this calculation changes with time of year and location is a subject of debate (see MB 261:23 and Bi'ur Halakha s.v. Mi-tchilat).  Bi'ur Halakha (261, s.v. She-hu) and R. Moshe Feinstein zt"l (IGM OC, vol. 1, no. 24) both indicate that one should ideally wait 72 minutes regardless of the time of year and location.  In addition, one should add a few minutes to this calculation to fulfill the mitzva of tosefet shabbat.


            One may rely on a clock or timepiece to indicate the time that the above conditions for nightfall will exist (MB 293:7).



V. Melakha and Havdala


1.  Scope of prohibition


            The Rema (OC 299:10) cites that some poskim, such as Rabbeinu Yerucham, are of the opinion that since the prohibition to perform melakha after nightfall, prior to havdala, is rabbinic in origin, the Rabbis were lenient with regard to performing melakha that does not involve labor and trouble (MB 299:38).  Thus, according to this opinion, one would be permitted to kindle a light or carry an object from one domain to another after nightfall prior to havdala (Rema ibid.).  However, the Rema (ibid.) rejects this leniency in practice, as do many other poskim, since there seems to be no source for distinguishing in this manner between the melakhot.  See MB 299:39.


            Moreover, some question the Rema's interpretation of Rabbeinu Yerucham, claiming that Rabbeinu Yerucham instead meant to be stringent.  According to this approach, Rabbeinu Yerucham forbade all performance of melakha prior to havdala, while allowing only non-laborious melakha after havdala recited in the amida; one would be permitted to perform all melakha only after reciting havdala over a cup of wine.  See Sha'ar Ha-tziyun 299:51 in the name of Derekh Ha-chaim.


            Thus, in practice one may perform melakha after nightfall only once one has recited the one of these havdala formulas: 1. The amida (ata chanantanu); or 2. "Barukh ha-mavdil bein kodesh le-chol" (Blessed is He who distinguishes between the sacred and the mundane); or 3. Havdala over a cup of wine (see next shiur).  Yet, the Mishna Berura cites that when not reciting havdala over wine (i.e. options 1 or 2 above), a scrupulous person should still not perform any melakha until after he has recited ve-ata kadosh (said after the amida).  This stringency does not apply, however, to rabbinic prohibitions, nor to melakha required for the public welfare (MB 299:40).


            It is important to note that the prohibition to perform melakha prior to reciting havdala devolves on each individual and is not communal in orientation.  Thus, one may recite havdala after nightfall and perform melakha even if the rest of the Jewish community has not yet done so.  Similarly, one may not perform melakha after nightfall prior to havdala even if the rest of the community has already recited havdala.  [MB 263:67.]


2. A Leniency


            There is one leniency prior to havdala after nightfall, however; one may discuss business or plans for the week, or anything else which might otherwise fall under the category of not being proper Shabbat talk ("m'm'zo chafatzekha ve-daber davar" - this will be discussed in detail in a future shiur).  One may also ask a non-Jew or a Jew who has already recited havdala to perform a task involving melakha.  [See OC 263:17 and 299:10, MB ibid., and Shulchan Arukh Ha-rav 299:15.]


            In the next shiur we will discuss the laws relevant to the various recitations of havdala and the requirement of "melave malka," escorting the Shabbat queen with a meal.


This website is constantly being improved. We would appreciate hearing from you. Questions and comments on the classes are welcome, as is help in tagging, categorizing, and creating brief summaries of the classes. Thank you for being part of the Torat Har Etzion community!