On Sefaria, an interesting Commentary

The other day when looking at the daily daf, I noticed a new commentary in the sidebar of Sefaria: Sha’arei Torat Bavel. And, while I disagreed with his conclusion in the specific comment discussed below, the nature of the commentary is good — comparison between parallel sugyot, emending texts, quoting books on Talmudic biography — the sort of thing I like to do.

In a recent post, I quoted Shevuot 4a:

אָמַר רַב אָשֵׁי: אַמְרִיתָא לִשְׁמַעְתָּא קַמֵּיהּ דְּרַב כָּהֲנָא, וְאָמַר לִי: לָא תֵּימָא רַבִּי נָסֵיב לַהּ אַלִּיבָּא דְתַנָּאֵי – וְלֵיהּ לָא סְבִירָא לֵיהּ;

Rav Ashi said: I said this statement of Rav Yosef before Rav Kahana, and he said to me: Do not say that Rabbi Yehuda HaNasi formulates the mishna according to different tanna’im but he himself does not hold accordingly.

and wrote:

(By the way, there were several Rav Kahanas, and my thoughts is that we’re dealing with fifth-generation Rav Kahana of Pum Nahara, who is Rav’s student and Rav Ashi’s teacher, rather than a sixth-generation Rav Kahana, who is Rav Ashi’s cousin. I think this because Rav Ashi does not merely say this to Rav Kahana and Rav Kahana then responds. That happened in the past. Now, in a single statement, Rav Ashi himself announces what Rav Kahana responded. That is what a student does for a teacher. I’ll pick up this thread in a separate post.)

So my understanding of this, and other, sugyot, is that Amora A repeated either a single statement of Amora B or IMHO more likely an encapsulated sugya, before his teacher, Amora C. Then, Amora C reacts. And Amora A repeats for us what Amora C’s reaction was.

Shaarei Torat Bavel writes otherwise:

(ד סע"א) אמר רב אשי אמריתא לשמעתא קמיה דרב כהנא ואמר לי לא תימא וכו'.—המלות "ואמר לי" מיותרות הן ויש למוחקן, כי כל הענין אמר רב אשי בעצמו לפני רב כהנא. וכן נראה מפירש"י שכתב: "אני פירשתי שמועה זו של רב יוסף לפני רב כהנא". וגם ברש"י יש למחוק מלת "ואמר". וכעין זה יש בב"מ צב א (וע"ש ברא"ש) וב"ב עו ב (ע"ש ברשב"ם), קיד א (ע"ש ברשב"ם) ובנדה נא ב (ע"ש בפרש"י) - כ"כ בס' דורות הראשונים ח"ב דף רסא.

והנה בב"ב עו ב הנוסח "אמר רב אשי אמריתא לשמעתא קמיה דרב כהנא ואמרית ליה טעמא וכו'".

ולפ"ז נ"ל דאין צורך למחוק כאן שתי המלות "ואמר לי" אלא לגרוס "ואמרית ליה", וכן יש להגיה ברש"י "ואמרית" במקום "ואמר". ודע שיש מקומות ג"כ, שרב אשי בעצמו לא אמר כל הענין לרב כהנא אלא רק השמועה הקודמת, ורב כהנא פירש השמועה, ובאלה גרסינן שפיר "ואמר לי" (עי' זבחים קו ב, כט א, ועי' מה שכתבנו שם; ועי' כתובות מ ע"א, צה ב, ב"ק טז א, כד ב, עא ב).

To roughly translate, with my interpolated comments in []:

The words “and he [Rav Kahana] said me” (on Shevuot 4a) are spurious and should be crossed out. For the entire matter [including what the revision / interpretation] was stated by Rav Ashi himself before Rav Kahana. And so it seems as well from Rashi’s commentary, in that he writes, “I explained this statement of Rav Yosef before Rav Kahana.”

A quick interjection, before commenting. The point seems to be that he sees the word פירשתי in Rashi, and thus assumes that version 2 of Rav Yosef, in which Rabbi Yehuda HaNasi has his own derashot, rather than version 1 in which Rabbi Yehuda HaNasi channels different Tannaim, Rabbi Akiva and Rabbi Yishmael, is the “peirush”. Therefore, according to this read of Rashi, Rav Ashi is saying that I gave the (re)interpretation #2 before Rav Kahana, so he should be the only speaker.

To continue my rough translation:

And also in Rashi, one should cross out the word ואמר [which would mean, “and he, Rav Kahana, said to me]. And in similar fashion there is in Bava Metzia 92a [with Rav Ashi reciting before Rav Kahana] (and see there in the Rosh) and in Bava Batra 76b — see there in the Rashbam; Bava Batra 114a — see Rashbam; and Niddah 51b — see Rashi. So is written in the Sefer Dorot Harishonim, volume 2, page 261a [though I don’t see it there; maybe there’s an OCR error]

And behold, in Bava Batra 72b, the nusach is “Rav Ashi said: I said this statement / shmayta before Rav Kahana, and I said to him the reason / ואמרית ליה, etc.

And based on this, is seems to me [the author of Shaarei Torat Bavel] that we don’t need to erase the words ואמר לי, “and he said to me”, but rather to emend it to ואמרית ליה, “and I said to him”. And so should we correct in Rashi, with ואמרית instead of ואמר.

And know that there are places, as well, in which Rav Ashi himself does not say the whole matter to Rav Kahanah, but rather only the preceding statement / shmua, and Rav Kahana explains the shmua, and in those it is totally fine and good that we are gores ואמר לי [and he, Rav Kahana, said to me]. (See Zevachim 106b, 29a, and see what we wrote there.)

I definitely agree that there is fluidity, in which ואמר לי and ואמרית ליה could turn into one another. The easy way is not that letters were added or omitted by the copyist. Rather, it is

אמר לי →

א”ל →

אמרית ליה

That is, it is a contraction followed by an erroneous contraction.

I think that Dorot HaRishonim is wrong, and is emending in error. This is because he misunderstood Rashi. What Rashi was saying was that the first explanation of Rav Yosef, in which Rabbi Yehuda HaNasi merely channeled the two earlier Tannaim, was not a solid quotation but a peirush. It was Rav Ashi’s interpretation of the idea he had heard. And therefore, fifth-generation Rav Kahanah wasn’t arguing, chas veshalom, with a third-generation Amora, but merely revising the interpretation. But he surely did respond to Rav Ashi. I discussed this at some length in yesterday’s post, where I wrote:

When Rav Ashi relates that he said this “shmayta” before Rav Kahana, we see that Rav Shteinsaltz rendered shmayta as “statement of Rav Yosef”. Then, I guess, Rav Kahana disagrees with Rav Yosef’s statement. Artscroll has it as “this version of Rav Yosef’s teaching”, and then has Rav Kahana saying “do not explain Rav Yosef to mean”. That is, he isn’t arguing but reinterpreting. Given that Rav Kahana is a later Amora than Rav Yosef, the latter is more “frum” in perspective, but I’d incline towards the former.

Also, as the author of Sha’arei Torat Bavel writes, there are certainly sugyot in which the teacher, Rav Kahana, does respond to Rav Ashi, the student. I see no reason to emend this one, and to emend Rashi to match. It is all a big misunderstanding of Rashi’s point.

Here is Sefaria’s writeup of this Talmudic commentator:

Sha’arei Torat Bavel is an early 20th-century commentary on the Babylonian Talmud by Lithuanian scholar Rabbi Zev Wolf Rabinowitz. The work includes traditional rabbinic commentary, incorporating and discussing the positions of earlier commentators and insights from contemporary works on science and medicine when relevant topics are addressed in the Talmud. In the work’s commentary on aggadic sections of the Talmud, Rabinowitz incorporates philosophical and ethical discussions and expresses love for the people and land of Israel. Sha’arei Torat Bavel was published by the author’s son, Yitzchak Rabinowitz, and edited by Professor Rabbi Professor Ezra Tzion Melamed.

Finally, let us peek at Hachi Garsinan, so that we can spot how the variants deal with the existence of non-existence of amar li / amrit leih.

Printings have Amar Li:

For manuscripts:

Florence messes up the name so it is before Rav Huna; but adds the preceding word ki and follows with amar, so “when I said it before Rav Kahana, he said”.

Munich has ve’a’ li, presumably for ve’amar li.

Vatican 140 omits it the phrase, so maybe this supports the reading, or at least allows it. But the juxtaposition, without a switch in speaker, is jarring. I said the shmayta before him. Don’t say this!

Vatican 156 has at least the am’ short for amar, and a marginal or above the line insertion of li.

We can imagine how לי followed by לא could cause omission of one word or insertion of the other.