Types of Mitzvot: 1. A Look at the Commandment of Fasting on Yom Kippur (Column 414)

A few weeks ago a request was posted on the site to write about types of mitzvot in Halakha. Ahead of Yom Kippur, I thought the commandment to fast gives us a good opportunity to begin touching on this topic. This column will deal with the prohibition of eating on Yom Kippur, and in the following columns I will continue the discussion on the difference between positive commandments (aseh) and prohibitions (lo ta’aseh), and finally distinguish among several kinds of positive commandments (which was the original question).

Is there a prohibition of eating on Yom Kippur?

In the verses of the Torah there appears a positive commandment to afflict oneself (“you shall afflict your souls”), but you will not find there a prohibition (lo ta’aseh) on eating. On the other hand, karet (spiritual excision) is indeed stated with regard to the fast (“and any soul that is not afflicted shall be cut off”). As is well known, all cases of karet in the Torah are imposed for a prohibition (with the exception of Passover and circumcision), and therefore it is quite clear that there must also be a prohibition here and not only a positive commandment. This raises the question of the source of that prohibition. As we shall see, the Gemara in Yoma 81a indeed wonders about this.

The source of the prohibition

The Mishnah there states:

If one ate and drank in a single lapse of awareness, he is liable for only one sin-offering. If he ate and performed work, he is liable for two sin-offerings. If he ate foods unfit for eating or drank liquids unfit for drinking, and [if] he drank fish brine or brine, he is exempt.

We see here that there is also a sin-offering for one who does not fast on Yom Kippur. This is an additional indication of the existence of a prohibition (since a sin-offering is certainly brought only for violations of prohibitions). Yet, as noted, such a prohibition has no explicit source in the verses. Indeed, the Gemara there remarks on this point:

Reish Lakish said: Why was no warning (azharah) stated regarding affliction? Because it is not possible. How would the Torah write it? If the Merciful One were to write “he shall not eat,” [that would imply] eating in the measure of an olive. If the Merciful One were to write “do not be un-afflicted,” it would imply “get up and eat.” Rav Hoshaya objected: Let the Merciful One write, “Beware lest you not be afflicted”—if so, there would be too many prohibitions. Rav Beivai bar Abaye objected: Let the Merciful One write, “Beware in the commandment of affliction”—if so, “beware” that relates to a prohibition will be a prohibition, and “beware” that relates to a positive commandment will be a positive commandment. Rav Ashi objected: Let it write, “Do not turn aside from affliction.” The difficulty remains. And the Tanna derives it from here…

Reish Lakish explains that no prohibition was stated regarding affliction because the Torah has no way to formulate such a prohibition. The Gemara then brings a baraita in which a derashah provides a source for a prohibition on failing to fast on Yom Kippur. The derashah is based on redundancy and superfluity in the verses, but we will not enter into it here.

Initial difficulties

The difficulty that arises here is twofold:

• Even if we are convinced that there is indeed a formulation difficulty, how can a prohibition be invented without a source? According to Reish Lakish, who says there is no Torah source for this prohibition, one cannot invent it. So how did Hazal establish that there is a prohibition here?

The existence of karet in the verses is not sufficient to say there is a prohibition, since the rule is that one does not punish unless a warning has been given (ein onshin ela im ken mazhirin). That is, even if the Torah explicitly states a punishment for a certain matter, if there is no warning, one cannot impose punishment for it.[1]

• Even according to the Tanna of the baraita quoted immediately afterward, who derives a prohibition for the fast from a doubled expression in the verses, at least according to Maimonides this is not acceptable. Maimonides rules in his Second Principle (Shoresh Sheni) that one cannot derive a prohibition’s warning from a derashah. He includes this under the Talmudic rule “one does not warn from an inference” (ein mazhirin min hadin). Nahmanides, in his glosses to that Principle, already objected and clarified that this rule applies only to a fortiori inferences (kal va-homer) and not to all types of derashot; this is the view of most Rishonim and emerges from many places in the Talmud. But Maimonides’ view is that the rule applies to all derashot and not only to kal va-homer.[2]

Maimonides and Saadia Gaon on the prohibition of eating on Yom Kippur

Maimonides, in the introduction to his Sefer Ha-Mitzvot (at the end of the Fourteenth Principle), addresses this point, and at the start sets out his general view and then qualifies it:

It is also proper to append the following introduction: Whatever carries liability of a court-imposed death penalty or karet is by necessity a prohibition (lo ta’aseh), except for Passover and circumcision, which incur karet even though they are positive commandments, as mentioned at the beginning of tractate Keritot. There is in no way any positive commandment other than those two whose violation incurs karet, all the more so a court-imposed death penalty. And whenever Scripture states that one who did such-and-such shall be put to death or incurs karet, we know that the act is in truth forbidden and that it is a lo ta’aseh.

Up to this point he writes what we have already seen above: there is no karet in the Torah for a positive commandment, except for Passover and circumcision (his source is tractate Keritot).

He now adds:

Sometimes the warning is stated in Scripture and the punishment is not, and sometimes the punishment is stated and the warning is stated, as with desecration of Shabbat and idolatry—He said (in the Ten Commandments), “you shall not do any work,” and “you shall not serve them,” and afterwards He required stoning for one who performed work (end of Parashat Shlach) or served (beginning of Parashat Shoftim). And sometimes the warning does not appear in Scripture as an explicit lo ta’aseh, but only the punishment is mentioned and the warning left unstated.

He does not bring Yom Kippur here, since in his opinion there is a warning there. But as we saw, at least according to Reish Lakish there is no explicit warning in the verses about eating on Yom Kippur. This raises the obvious difficulty:

But the principle among us (Yoma 81a; Sanhedrin 56b; Zevachim 106b–107a; Keritot 3b; cited in Prohibitions 90 and the end) is: Scripture does not punish unless it has warned, and it is impossible that there should be a punishment without a warning. Therefore they say everywhere: “We heard the punishment; from where [do we derive] the warning?”—Scripture therefore says such-and-such.

This is precisely the difficulty we raised regarding Reish Lakish’s words.

He now adds:

And when the warning is not explicit in the verse, they learn it by comparison (hekesh) from the Torah’s comparisons. As they said (Sanhedrin 85a) about the warning for one who curses his father or mother and one who strikes his father or mother (Prohibitions 318–319)—that the warning is not explicit in the verse, for it does not say “do not curse your father” and it does not say “do not strike your father,” but Scripture imposed the death penalty on one who struck or cursed. We therefore knew that these are lo ta’aseh, and we derived for them and their likes (Prohibition 26 and onwards; end of Prohibitions) the warning from elsewhere by way of comparison.

At times the warning is learned from a derashah and is not explicitly written in a verse. This, of course, takes us back to the baraita we mentioned above.

But here too the second difficulty we presented earlier arises, and Maimonides senses this and answers:

This does not contradict their statement (Pesachim 24a; Yevamot 22b; Makkot 5b, 17b; Zevachim 106b), “one does not warn from an inference,” nor their statement elsewhere (ibid.), “and do we warn from an inference?” For we only say “one does not warn from an inference” in order that we should not forbid something that has no specific prohibition by way of comparison. But when the punishment is explicitly stated in the Torah for one who performs that act, we know necessarily that the act is forbidden and warned against. We then derive by comparison the warning in order to uphold the principle that Scripture does not punish unless it has warned. Once the warning for that matter has been obtained, one who transgressed and did it incurs karet or death. Know this introduction and remember it along with the preceding principles for all that will be mentioned:

As noted, in his view a warning cannot be learned from a derashah; thus, seemingly, even where the punishment is explicit in the Torah there is no way to apply it and punish. He therefore explains that indeed a warning learned from a derashah cannot by itself serve as the basis for punishment—but this is only for the punishment of lashes, which is not explicitly written in the Torah.[3] For all other punishments, when the punishment is explicitly written in the Torah and there is no warning, the warning can be learned from a derashah.[4]

This, of course, does not explain Reish Lakish’s view, for he maintains that there is no warning at all (unlike the baraita, which derives a warning from a derashah). On further thought, however, it seems possible to reconcile Reish Lakish’s position in a similar manner. In a case where there is a linguistic/formulation difficulty as he describes, one may infer from the fact that the Torah writes a punishment that there is also a prohibition. The outcome is similar to what we saw in Maimonides when one “warns by derashah,” except that here it would apply even when the Torah does not write any warning at all (not even by derashah).

Moreover, according to our approach it is possible that even the Tanna of the baraita, who derives the warning by a derashah, would not need the principle established by Maimonides above. If we assume that he too agrees with Reish Lakish—that is, that in his view there is no way to formulate a prohibition here in the straightforward language of the Torah—then in such a situation a warning by derashah might suffice even if the punishment had not been explicitly written in the Torah (though in that case there would only be lashes).

It should be noted that R.Y.F. Perla, in his commentary to Sefer Ha-Mitzvot of Saadia Gaon, indeed argues that according to Rasag there is no prohibition (lo ta’aseh) on eating on Yom Kippur, only a positive commandment (aseh). In his view this appears to be a third case of karet for a positive commandment, despite the Talmud explicitly stating that there are only two cases. In addition, we saw that the Mishnah in Yoma mentions sin-offerings for eating on Yom Kippur, and it is difficult to accept the notion of a sin-offering without a prohibition. His words are difficult, but the motivation is clear: the absence of an explicit warning in Scripture.

It may be that his intent is to argue that there is a prohibition (an issur) but that one is not flogged for it because there is no explicit warning. Note that the punishment here is not death or lashes but karet, and karet is meted out by the Holy One, blessed be He, not by the earthly court. One might say that just as God does not require formal forewarning (hatra’ah) to know that a person deserves punishment, so too He is not contingent on an explicit warning in Scripture. God knows what is in a person’s heart, and He can know when a person transgressed intentionally even without forewarning and without an explicit warning (that is, He knows when the offender knew the matter was forbidden even though there is no explicit warning in Scripture), and therefore He can punish even in the absence of a written warning.[5]

An additional proposal for explaining Rasag’s view

The question of the source is mainly technical. But on a deeper look, a more fundamental question arises. The Torah commands us to afflict ourselves, yet the content of that command is of a prohibitory nature rather than a positive one. It forbids me from eating and drinking and does not command me to do something. This raises the possibility that even though the Torah formulates it as a positive commandment to afflict oneself, in essence it is a prohibition against eating and drinking—only worded as a positive commandment.

This creates another way to understand the difficult position attributed to Rasag: perhaps, according to him, this commandment is indeed counted as a positive commandment because of its formulation in the Torah, but in essence it is a prohibition; hence the karet and the sin-offering.

Why is the fast on Yom Kippur a positive commandment rather than a prohibition?

This proposal raises a sharper and more fundamental question: what is the definition of an aseh and a lo ta’aseh?[6] Seemingly, a positive commandment is a command to do something, and a prohibition is a command not to do something (to refrain). According to that definition, it would indeed appear that the commandment of affliction is a prohibition and not a positive commandment, for the content of the Yom Kippur command is not an action but a non-action (to refrain from something). On the other hand, the Torah’s formulation indicates that it is a positive commandment and not a prohibition. If it is in fact a prohibition, why would the Torah choose a formulation that does not reflect the nature of the commandment? Moreover, the difficulty discussed in the Gemara, according to Reish Lakish, in formulating a prohibition in this case is not clear. Why shouldn’t it simply say, “Do not eat and do not drink,” like other prohibitions of eating? What, according to Reish Lakish, prevents formulating a prohibition here? One could simply write, “You shall not eat and you shall not drink,” as is written with other eating prohibitions.

One might infer from this that the prohibition on Yom Kippur imposes on us a positive duty to afflict ourselves and not merely a ban on eating, and therefore it is not an eating prohibition but a duty to be afflicted. But there is a logical difficulty here, since a fast is nothing other than a prohibition on eating. One might suggest that there is a duty to undertake active measures to arrive at a state of affliction, such as sitting in the sun or cold and the like. But the Gemara itself raises this possibility and rejects it. The Gemara in Yoma 74b writes as follows:

The Sages taught: “You shall afflict your souls”—one might have thought he should sit in the sun or in the cold so as to suffer; therefore Scripture says, “and you shall do no manner of work”: just as work is a matter of sitting and refraining, so too affliction of the soul is sitting and refraining. And say perhaps [only] that where he is sitting in the sun and it is hot for him, we should not tell him, “Get up and sit in the shade,” or where he sits in the shade and it is cold for him, we should not tell him, “Get up and sit in the sun”? [No—] comparable to work: just as with work you did not distinguish [between cases], so too with affliction you shall not distinguish. Another baraita: “You shall afflict your souls”—one might have thought he should sit in the sun or in the cold and suffer; therefore Scripture says, “and you shall do no work”: just as work is something for which one is liable elsewhere, so too affliction of the soul is something for which one is liable elsewhere. And what is that? Piggul and notar. Shall I bring piggul and notar, which are subject to karet, and not bring tevel, which is not subject to karet? Scripture says, “you shall afflict” and “you shall afflict your souls”—it includes. Shall I bring tevel, which is [liable to] death, and not bring neveilah, which is not [liable to] death? Scripture says, “you shall afflict” and “you shall afflict your souls”—it includes. Shall I bring neveilah, which is [forbidden by] a lo ta’aseh, and not bring hulin [ordinary food], which is not [forbidden by] a lo ta’aseh? Scripture says, “you shall afflict” and “you shall afflict your souls”—it includes. Shall I bring hulin, which is not [covered by a command of] “arise and eat,” and not bring terumah, which is [covered by] “arise and eat”? Scripture says, “you shall afflict” and “you shall afflict your souls”—it includes. Shall I bring terumah, which is not [subject to] “you shall not leave over,” and not bring holy [sacrificial] foods, which are subject to “you shall not leave over”? Scripture says, “you shall afflict” and “you shall afflict your souls”—it includes. And if you wish, say: it states, “and I will destroy that soul”—an affliction that is a loss of the soul; what is that? Eating and drinking. And if you would say [the verse] speaks of sexual relations, it states, “and I will destroy the soul”—an affliction involving loss of soul; and what is that? Eating and drinking. The school of Rabbi Yishmael taught: It says “affliction” here and “affliction” elsewhere; just as there [in the other place] “affliction” means hunger, so too here “affliction” means hunger. And shall we derive it from “If you afflict my daughters”? We derive the affliction of the public from the affliction of the public, and we do not derive the affliction of the public from the affliction of an individual. And shall we derive it from the affliction in Egypt, as it is written, “and He saw our affliction,” and we say: this refers to abstention from intercourse? Rather, we derive affliction by the hand of Heaven from affliction by the hand of Heaven, and we do not derive affliction by the hand of Heaven from affliction by the hand of man.

The Gemara insists on defining the duty of affliction as a passive “sit and refrain,” unlike all other eating prohibitions. This returns us to the two fundamental difficulties: why is this commandment defined as a positive commandment and not as a prohibition? And what was the formulation difficulty Reish Lakish described in stating a prohibition against eating on Yom Kippur?

Between fasting and refraining from eating

We are forced to conclude that fasting is not identical with not eating. If the Yom Kippur law were simply a prohibition on eating, there would be no problem formulating such a prohibition. All of the Gemara’s contortions arise because it seeks to formulate a prohibition on violating a fast. Of course, the positive command itself speaks of afflicting the soul and not of refraining from eating; one might have thought those are equivalent instructions. In the realm of a prohibition, it is quite clear that they are not.

At the end of column 192 I discussed the case of a minor who grew two pubic hairs in the middle of Yom Kippur. Later authorities formulate this as a “yeshiva-style” inquiry: is the commandment to fast on Yom Kippur defined over the entire day as one unit, or is it a separate obligation for each and every moment? The practical difference concerns a person who ate in the middle of the day, whether sinfully or due to illness: is there any point to completing the fast (for if the mitzvah is defined as an all-day obligation, he has already forfeited it when he ate)? I argued there that a minor who comes of age mid-day must complete the fast according to both conceptions. I explained that defining the fast as applying to an entire day is not a halakhic definition but a conceptual-practical one. That is, the intent is not that the mitzvah is defined as a single obligation for the entire day as a halakhic construct, but that the linguistic concept “fast” is, by its nature, a comprehensive definition over the totality of the day’s moments.

This can be explained as follows. The Torah’s mitzvah is to fast. But not eating is not, by itself, a fast. When a person does not eat for a single moment, we do not say that he is “fasting” in that moment; he is merely not eating. No one would say that a person is fasting between meals; he simply is not eating between them. A fast, by its very definition, is set over a period of time and not over a single moment of refraining from eating. The Torah (Leviticus 16:29) fixed that period as the tenth of the month, on which we are to afflict ourselves. From here we may infer that the period that defines a fast is from evening to evening, a full twenty-four hours. The implication is that if the minor was fasting rabbinically (chinuch) in the first half of the day and then came of age (grew two hairs), if he completes the fast he will have been in a state of fasting for the entire day (the first half rabbinically and the second half by Torah law). Since this requirement pertains not to the “wholeness” of the mitzvah but to the practical concept of “fast,” we may say that in fact he fasted; and therefore, in the moments after he came of age he fulfilled a Torah mitzvah (for the mitzvah can be defined moment by moment—but in each moment one must be fasting, not merely not eating).

From here we can understand why the Torah formulates its command as a duty to fast and not as a prohibition on eating. This is what troubled Reish Lakish, and therefore he concluded that there is no linguistic way to express a prohibition of violating a fast. If the Torah had formulated here a prohibition on eating, we would have thought it is a standard eating prohibition and interpreted it moment by moment, with the consequences described above. When the Torah commands us to fast, it must use the verb “to afflict,” and that is what creates the difficulty of formulating a prohibition.

This also clarifies why the formulation as a positive commandment is indeed a positive commandment and not a prohibition. There is a positive commandment to fast and a prohibition to violate the fast. The act of violating the fast is an active deed, and this is exactly what the Torah seeks to forbid (but it has no linguistic way to do so). On the contrary, the remaining question is why the command that does appear in the Torah is a positive commandment and not a prohibition, for it does not impose on us an act but rather an abstention (not to violate the fast).

This leads us to a different distinction between positive commandments and prohibitions, which will be the topic of the next column.

In the meantime, may we all have a good year and be sealed for good.

[1] For an explanation, see column 377.

[2] See at length in the book Ru’ach Ha-Mishpat, in the second gate, and in the book Yishlach Sharashav, vol. I, in the essay on the Second Principle. See also vol. II, in the essay on the Fourteenth Principle.

[3] Lashes are exceptional, learned from a general teaching that for every prohibition one who transgresses it is liable to lashes (“and the judge shall make him lie down and strike him”—see Makkot 13b). All other types of punishments must be explicitly stated in the Torah for each prohibition.

[4] For an explanation, see my articles cited there.

[5] The assumption here is that the need for a written warning in Scripture is analogous to the need for formal forewarning of the offender: to ensure that he is acting intentionally and knows that the matter is forbidden. The common approach does not understand it this way. It is generally thought that the need for a warning is substantive: without it, there is truly no prohibition. But when the punishment is explicitly written in the Torah, one may say that all would agree the need for a warning is only to establish culpability.

[6] See at length in the book Yishlach Sharashav, vol. I, in the essay on the Sixth Principle, and more briefly also in my essay on Torah and science.