Monday, January 22, 2007

Daf Yomi: Taanit daf 12: Eating After Waking Up Before A Fast

While preparing Rif Yomi, I came across the following interesting gemara and pesak halacha regarding someone who broke off eating and slept before a fast, whether he could resume, and under what circumstances:
{Rif Taanit 3b}
{Taanit 12a}
אמר רבא לא שנו אלא שלא גמר אבל גמר אינו אוכל
איתיביה אביי ישן ועמד הרי זה אוכל
התם שלא סילק
איכא דאמרי אמר רבא לא שנו אלא שלא ישן אבל ישן אינו אוכל
איתיביה אביי ישן ועמד הרי זה אוכל
התם במתנמנם
היכי דאמי מתנמנם אמר רב אשי נים ולא נים תיר ולא תיר כגון דקרו ליה ועני ולא ידע לאהדורי סברא וכי מדכרו ליה מדכר:

Rava {our gemara: Abaye} said: They only learned this {about being able to eat in the morning before a fast} where he did not finish {his meal} but if he finished it, he may not eat.
Abaye {our gemara: Rava} objected {from a brayta}: If he slept and arose, he may eat.
There, he did not remove {the table}.

Some {versions} say:
Rava said: They only learned this where he did not sleep, but if he slept, he may not eat.
Abaye objected: If he slept and arose, he may eat.
There {where he may eat} is where he {merely} dozed.

What is the definition of this dozing?
Rav Ashi said: {Taanit 12b} He is asleep but not asleep, awake but not awake, such that you call him and he answers, but it not able to respond with thought {to give a reasoned response} but if you remind him, he remembers {e.g. if you ask him while is is dozing like this, "Did you hear such and such?" he will be able to remember.}

הא מילתא לא אפסיקא בה הלכתא בהדיא אלא מדחזינא לגמרא דקא טרח לפרושי מתנמנם שמעי' דהלכתא כי האי לישנא בתרא
ועוד דאיסורא דרבנן הוא ולקולא עבדינן הלכך בין סילק בין לא סילק אם לא ישן אוכל ושותה עד שיעלה עמוד השחר הדין הוא סברא דילן

In this matter, the halacha is not explicitly set up. However, from the fact that we see that the gemara troubles itself to define "dozing," we deduce that the halacha is in accordance with this latter phrasing. And furthermore, it is a Rabbinic prohibition, and we conduct ourselves leniently. Therefore, whether they removed {the table} or did not remove, if he did not sleep, he may eat and drink until daybreak. Such is our opinion.

{Though see Ran as to the relation between the two leshonot and the subsequent context in which this dozing is done.}

ואיכא מרבוותא מאן דכתב הכין קי"ל דלגבי תענית דאכיל בלילי ומפסיק אם גמר וסילק לא יאכל אבל אם לא סילק ודעתו עוד לאכול אע"פ שישן עומד ואוכל
והאי מימרא דיליה אתי כלישנא קמא
And there is among the {post-Talmudic} Sages one who writes as follows, that we establish that by a fast in which he ate ate night and stopped, if he finished {his meal} and removed {the table} he may not eat, but if he did not remove and his thoughts are still to eat, even though he slept {entirely, not merely dozed}, he may arise and eat. And this statement of his goes according to the first phrasing.
So ends my citation of Rif.

There are slight girsological differences between the text in our gemara and the text in the Rif in terms of the disputants in the first lashon of the gemara. While our printed text has Abaye saying and Rava objecting, Rif here has Rava saying and Abaye objecting, just as in the second lashon. And Masoret haShas on the side of our gemara notes that old seforim have this text like the Rif, and so does the Rosh. (Bach also notes a change from ישן ועמד in the second lashon to אכל ועמד.) Etcetera.

I am not sure that I am persuaded by Rif's proof that we rule in accordance with the second lashon, based on the fact that Rav Ashi's took the trouble to define dozing. This is a twist on the general deciding principle (kelal horaah) that where the Amoraim have a give and take (shakla vetarya) within one opinion in a dispute, it is strong evidence that they felt the halacha is in accordance with that opinion, for otherwise they would not have gone to the bother. This instance is not really a give and take, but rather merely a definition of the term, but the proof follows the same general idea.

The reason I am not entirely persuaded is that a) I am not certain that Rav Ashi actually made his statement defining mitnamnem in this gemara, but rather there is strong reason to believe it was transferred from elsewhere and b) if the Savoraim indeed transferred it, their opinion does not carry the full halachic weight of the Amoraim, since Ravina/Rav Ashi were sof horaah, and c) it truly seems that the reason for this transfer is not because of specific halachic ruling in accordance with this opinion, but rather is a regular feature of the gemara in which it defines this term.

On to the details of what I set out above.

Our gemara defines dozing as follows: היכי דאמי מתנמנם אמר רב אשי נים ולא נים תיר ולא תיר כגון דקרו ליה ועני ולא ידע לאהדורי סברא וכי מדכרו ליה מדכר. If Rav Ashi set out to define it here, then one might effectively argue that this was because the halacha was in accordance with this particular lashon. However, this exact phrase actually occurs exactly five times in Bavli (see this Google search). In each case, it begins היכי דמי מתנמנם and proceeds to cite Rav Ashi for this exact definition.

When the exact same gemara occurs in more then one place, we usually assume that one of these places was the original location, the original context for the statement within a discussion amongst Amoraim, and then there was a transfer to the other sugya or sugyot. (I could see an argument to be made about Rav Ashi, as a redactor, that he placed his statement initially in each of these contexts. However, we claim transfer in so many other cases that it is safe to assume the same here as well.)

Let us then examine the statement in its various contexts throughout Shas.

1) One instance is in the 9th perek of Nidda, Nidda 63a:
איזהו רוק תפל: תנא כל שלא טעם כלום מבערב סבר רב פפא קמיה דרבא למימר כמאן דאמר לא טעם מידי באורתא אמר ליה רבא מי קתני בערב מבערב קתני לאפוקי היכא דקדים ואכיל
אמר רבה בר בר חנה אמר רבי יוחנן איזהו רוק תפל כל שעבר עליו חצות לילה ובשינה
למימרא דבשינה תליא מילתא
והתנן ישן כל היום אין זה רוק תפל ניעור כל הלילה הרי זה רוק תפל
התם במתנמנם
היכי דמי מתנמנם אמר רב אשי נים ולא נים תיר ולא תיר דקרו ליה ועני ולא ידע לאהדורי סברא וכי מדכרו ליה מדכר
Thus, the context is an apparent contradiction in defining rok tafel between a statement by Rabba bar bar Chana citing Rabbi Yochanan and a Mishna. (Perhaps we can say that the first generation Amora from Eretz Yisrael, Rabbi Yochanan, has a competing tradition, or perhaps we can harmonize as the gemara does.)

Note that the contradiction between these sources is done anonymously, by the setama digmara. The harmonizing detail proposed by the setama is mitnamnem, dozing. Then, perhaps it is Rav Ashi who pipes up, helpfully defining the term, or more likely it is the setama bringing from elsewhere Rav Ashi's statement defining the term. Since the context is not a discussion between Amoraim, I would interpret it as the latter.

2) A second occurrence is in the second perek of Megillah, Megillah 18b. The Mishna said that if one reads the megillah with long pauses, dozing in between, one fulfills the obligation. In the gemara, Rav Ashi defines dozing.

דף יז,א משנה הקורא את המגילה למפרע לא יצא קראה על פה קראה תרגום בכל לשון לא יצא אבל קורין אותה ללועזות בלעז והלועז ששמע אשורית יצא קראה סירוגין ומתנמנם יצא היה כותבה דורשה ומגיהה אם כוון לבו יצא ואם לאו לא יצא היתה כתובה בסם ובסיקרא ובקומוס ובקנקנתום על הנייר ועל הדפתרא לא יצא עד שתהא כתובה אשורית על הספר ובדיו:

דף יח,ב גמרא
מתנמנם יצא וכו': היכי דמי מתנמנם אמר רב אשי נים ולא נים תיר ולא תיר דקרו ליה ועני ולא ידע לאהדורי סברא וכי מדכרו ליה מידכר:

This is not in the context of a discussion of Amoraim, though of course Rav Ashi is an Amora. It is quite plausible that Rav Ashi is directly defining the term in the Mishna, and the reason no other Amora discusses this with him is that this is not a matter of dispute.

However, one curious feature here that we must pay heed to is the introduction of the statement with היכי דמי. In fact, היכי דמי occurs systematically within this gemara in Megillah, in the analysis of this Mishna. Thus, on Megillah 18a:

קראה תרגום לא יצא וכו': היכי דמי אילימא דכתיבה מקרא וקרי לה תרגום היינו על פה לא צריכא דכתיבה תרגום וקרי לה תרגום:
which is an anonymous (and thus likely stamaitic) analysis of the Mishna.

Again on Megillah 18a, immediately following the above:
אבל קורין אותה ללועזות בלעז וכו':
והא אמרת קראה בכל לשון לא יצא
רב ושמואל דאמרי תרוייהו בלעז יווני
היכי דמי אילימא דכתיבה אשורית וקרי לה יוונית היינו על פה
א"ר אחא א"ר אלעזר שכתובה בלעז יוונית

In this instance, it is an anonymous analysis of Rav and Shmuel's analysis of the Mishna, which then leads into a named Amora's elaboration.

The next instance is on Megillah 18b, in the section we have seen already:
מתנמנם יצא וכו': היכי דמי מתנמנם אמר רב אשי נים ולא נים תיר ולא תיר דקרו ליה ועני ולא ידע לאהדורי סברא וכי מדכרו ליה מידכר
which is Rav Ashi's definition of the term mitnamnem, a term which occurs in the Mishna. It could well be that Rav Ashi stated this on this Mishna, and the setama adds the connective introductory glue of היכי דמי מתנמנם. It could also be that the setama wishes to define this term, just as it systematically defines other terms, and copies Rav Ashi's definitional statement from elsewhere.

The next instance immediately follows on Megillah 18b:

היה כותבה דורשה ומגיהה אם כוון לבו יצא וכו':
היכי דמי אי דקא מסדר פסוקא פסוקא וכתב לה כי כוון לבו מאי הוי על פה הוא אלא דכתב פסוקא פסוקא וקרי ליה
ומי יצא
והאמר רבי חלבו אמר רב חמא בר גוריא אמר רב הלכה כדברי האומר כולה ואפי' למ"ד מאיש יהודי צריכה שתהא כתובה כולה
אלא דמנחה מגילה קמיה וקרי לה מינה פסוקא פסוקא וכתב לה
This is once again a clearly anonymous, and thus stamaitic section, though the analysis is based on known statements of known Amoraim. And once again it is analysis of the Mishna.

Thus, it seems clear that the היכי דמי is part of the setama's systematic analysis of the Mishna. At first glance, this would seem to be disappointing. It would seem that this gemara in Megillah can not be our source text.

However, in fact the opposite seems the case. We have seen that the setama can be the connective glue introducing named Amoraim who are local to this gemara. Thus the setama analyzed Shmuel and Rav's statement in order to introduce Rav Acha's citatation of Rabbi Eleazar, which is clearly a local statement and analysis of the Mishna. So too, even if Rav Ashi actually defined it here as an analysis of the Mishna, the setama could introduce it with היכי דמי.

Why should it introduce it with היכי דמי? Simply put, the setama favors regularity and systematic approaches in analysis, and presents a systematic definition of the terms in the Mishna, introduced each time by היכי דמי, at times introducing the Amoraic statements already present and at times introducing its own analysis.

Indeed, the fact that the word היכי דמי occurs is good evidence that this gemara in Megillah is the source text for all. That is, the context of multiple היכי דמיs in analyzing the Mishna shows that this is a local feature, and thus it is a feature expected here. It is not out of place in any other gemara, such that we should ask why the words היכי דמי occur, but it stands to reason that its genesis was in this local gemara in Megillah. That it occurs in those other locations implies that it was transferred from this location, together with the phrase היכי דמי.

This is not to say that this is the only location where we have היכי דמי in context. Indeed, the next occurrence, in Yavamot, also has היכי דמי in proximity (though there are other reasons for thinking it did not originate there). However, the first instance, in Niddah, does not have היכי דמי in context, nor does the fourth occurrence, the gemara in Taanit, nor does the fifth occurrence, the gemara in Pesachim.

3) A third occurrence of the segment היכי דמי מתנמנם אמר רב אשי נים ולא נים תיר ולא תיר דקרו ליה ועני ולא ידע לאהדורי סברא וכי מדכרו ליה מידכר occurs in the sixth perek of Yevamot, in Yevamot 54a:

גופא אמר רב יהודה ישן לא קנה ביבמתו דאמר קרא יבמה יבא עליה עד דמכוין לה לשם ביאה
והתניא בין ער [בין ישן אימא בין ערה בין ישנה והתניא בין ער] הוא בין ישן הוא בין ערה היא בין ישנה היא
הכא במאי עסקינן במתנמנם
ה"ד מתנמנם אמר רב אשי נים ולא נים תיר ולא תיר כגון דקרו ליה ועני ולא ידע לאהדורי סברא וכי מדכרו ליה מדכר גופא

As I frequently note, the word גופא at the end of a segment is evidence that the analysis that preceded it was stamaitic. It signifies a return to the actual original Amoraic text of the gemara. Here, Rav Yehuda had a statement that a yavam cannot acquire a yevama in levirate marriage while asleep. The setama raises the issue that this appears to contradict a brayta, and they resolve and harmonize this (a common stamaitic approach) by saying that the brayta refers to a case of mintamnem. Then, Rav Ashi defines mitnamnem. If we accept that the setama is post-Rav Ashi, then this definition must have been brought in from elsewhere.

4) A fourth occurrence occurs in the gemara intially under discussion, in the first perek of Taanit, on Taanit 12b:

איכא דאמרי אמר רבא לא שנו אלא כשלא ישן אבל ישן אינו אוכל
איתיביה אביי ישן ועמד הרי זה אוכל
התם במתנמנם
היכי דמי מתנמנם אמר רב אשי נים ולא נים תיר ולא תיר דקרו ליה ועני ולא ידע אהדורי סברא וכי מדכרי ליה מדכר

Once again, this is an anonymous harmonization of Rava's statement with that of a brayta cited by Abaye. (Perhaps one can say that Abaye's was an effective disproof, or that this is a correct harmonization.) That this statement of Rav Ashi defines a term in the setama's harmonization implies that this is the setama transferring the definitional statement from elsewhere.

Perhaps we could say that it is not the setama harmonizing here but rather Rava answering Abaye, and Rav Ashi is giving a helpful definition. However, the fact that it is not preceded by אמר ליה, and all the more so because we have better candidate source texts, it is quite likely that this is indeed the setama transferring from elsewhere.

5) The final, fifth occurrence is in the tenth perek of Pesachim, on Pesachim 120a:

דף קכ,א משנה ישנו מקצתן יאכלו כולן לא יאכלו

דף קכ,ב משנה רבי יוסי אומר נתנמנמו יאכלו נרדמו לא יאכלו הפסח אחר חצות מטמא את הידים הפגול והנותר מטמאין את הידים:

דף קכ,ב גמרא ר' יוסי אומר נתנמנמו יאכלו נרדמו לא יאכלו
ה"ד נתנמנם
אמר רב אשי נים ולא נים תיר ולא תיר כגון דקרי ליה ועני ולא ידע לאהדורי סברא וכי מדכרו ליה מדכר
אביי הוה יתיב קמיה דרבה חזא דקא נמנם
א"ל מינם קא נאים מר
א"ל מינומי קא מנמנם
ותנן נתנמנמו יאכלו נרדמו לא יאכלו

Here, the Mishna sets out sleeping as an impediment to continuing the meal. Rabbi Yossi makes a distinction between nitnamnemu and nirdemu, which then would then prompt a definition of the terms.

To that end, there are two perhaps-identical definitions offered in the gemara. The first is offered in a direct manner by Rav Ashi, that it is נים ולא נים, "dozing but not dozing, etc." The second may be deduced from the anecdote related in the gemara about Abaye and Rabba:

אביי הוה יתיב קמיה דרבה חזא דקא נמנם
א"ל מינם קא נאים מר
א"ל מינומי קא מנמנם

"Abaye was sitting before Rabba. He saw that he was dozing (namnem). He said to him: Master is sleeping (na`im). He said to him: I am (merely) dozing (menamnem)."

Thus, this anecdote in its way shows the meaning of the terms na`im and mitnamnem, which is then of great utility in understanding the Mishna. The gemara makes this clear by stating immediately thereafter ותנן נתנמנמו יאכלו נרדמו לא יאכלו.

Could this be the source text of the transfer, and all other sugyot destinations? Perhaps. Rav Ashi is giving his definition directly on a Mishna, just as he was in Megillah. And this particular gemara seems a vortex for defining mitnamnem, so Rav Ashi's statement would be quite at home here.

On the other hand, while the gemara in Megillah gives ample basis to expect the words היכי דמי, such that the statement looks quite local there, this gemara in Pesachim does not have other examples of היכי דמי in immediate context.

Furthermore, Rav Ashi is later than Abaye and Rabba. If his statement were local to Pesachim, why lead off with his statement? Better to begin with the earlier statement of Abaye and Rabba, which might well even serve as a basis for his statement.

Finally, what is the need for two definitions here. Abaye and Rabba, with this incident, gave an excellent definition of mitnamnem, and Rav Ashi's statement really just echoes it. It is dozing but not entirely dozing. Certainly it is true that Rav Ashi's statement gives more detail than that, but it really seems that Abaye and Rabba's definition should suffice, and indeed, initially did suffice, to define mitnamnem in Pesachim.

I wrote above that this gemara in Pesachim seemed a vortex for definitions of mitnamnem. A vortex is "a place or situation regarded as drawing into its center all that surrounds it." Indeed, once we have one definition in Pesachim, that of Abaye and Rabba, it is possible that the setama saw fit to engage in inter-sugya borrowing to transfer Rav Ashi's statement from Megillah, in order to shed further light on the topic at hand.

In sum, of the five sources, only Megillah and Pesachim seem likely candidates for the original location of the source-text. The gemara in Taanit appears to be a transfer, and a transfer by the setama digmara, and so I am unconvinced by Rif's argument that the fact that Rav Ashi or the gemara bothered to define mitnamnem, we rule in accordance with the second lashon.

And what of the fact that the setama digmara troubled itself to bring this definition here? I am still not persuaded, firstly because the setama is post-Talmudic and thus not necessarily binding halachically, but secondly and more importantly, I am not convinced that the reason for troubling to bring a definition here is that they held conclusively like the second lashon.

Rather, it seems to me that the gemara is consistent in bringing Rav Ashi's definitional statement wherever the word mitnamnem occurs. Thus, Rav Ashi's statement occurs five times in Shas, and the word mitnamnem occurs in 6 different locations: the five we mentioned (though in Pesachim the word is nitnamnemu/nitnamnem, leading with a nun rather than a mem) plus one instance in Berachot 3b:

ודוד בפלגא דליליא הוה קאי מאורתא הוה קאי דכתיב (תהילים קיט) קדמתי בנשף ואשועה וממאי דהאי נשף אורתא הוא דכתיב (משלי ז) בנשף בערב יום באישון לילה ואפילה אמר רב אושעיא אמר רבי אחא הכי קאמר <דוד> מעולם לא עבר עלי חצות לילה בשינה. רבי זירא אמר עד חצות לילה היה מתנמנם כסוס מכאן ואילך היה מתגבר כארי רב אשי אמר עד חצות לילה היה עוסק בדברי תורה מכאן ואילך בשירות ותשבחות.

Why does the gemara not give Rav Ashi's definition here? There are a few possible reasons. Firstly, it is clear that David did not entirely sleep from context. Second, it gives its own definition here of היה מתנמנם כסוס. Thirdly, in this instance Rav Ashi himself holds that he was not mitnamnem but rather was learning Torah, so it might appear awkward to offer Rav Ashi's definition immediately before Rav Ashi's denial. Finally, an exact definition is not required here since this is narrative aggada rather than halachic material that might (but might not) require technical definition if we were to rule in accordance with it.

In all other contexts, though, mitnamnem is defined by Rav Ashi's statement. This suggests to me that the motivating factor is simply to transfer and apply a known Amoraic statement consistently to other sugyot because of the ambiguity of the word mitnamnem, rather than a concern prompted by a knowledge that we rule in accordance with a specific opinion.

No comments:


Blog Widget by LinkWithin