Rava Injection or Deletion
Also, did Rashi confuse the scribes?
In yesterday’s daf, Nazir 12b, something curious happens with Rava’s analysis of the Mishnah. That Mishnah had discussed one person saying “It is upon me to shave half a nazir” with the implicit associated korbanot obligations, and another saying of the two korbanot, and another saying “And I, upon me to shave half a nazir”. Rabbi Meir and the Sages argue about whether each has to shave a full nazir, since there isn’t shaving half a nazir, or whether they can join together to shave a full nazir.
Upon this, in our printed texts, we have:
גְּמָ׳ אָמַר רָבָא: הַכֹּל מוֹדִים, כֹּל הֵיכָא דְּאָמַר ״חֲצִי קׇרְבָּנוֹת נָזִיר עָלַי״ — חֲצִי קׇרְבָּן מַיְיתֵי. ״קׇרְבְּנוֹת חֲצִי נָזִיר עָלַי״ — כּוּלֵּיהּ קׇרְבָּן בָּעֵי אֵיתוֹיֵי. מַאי טַעְמָא — דְּהָא לָא אַשְׁכְּחַן נְזִירוּת לְפַלְגָא.
GEMARA: With regard to this dispute, Rava said: All concede that whenever one said: Half of the offerings of a nazirite are incumbent upon me, he brings half of the offerings, since he vowed to pay only that amount. Also, everyone agrees that if he said: The offerings of half a nazirite are incumbent upon me, he needs to bring all of the offerings of a nazirite. What is the reason that he must bring all of the offerings of a nazirite? It is that we have not found such an entity as half a naziriteship. If one vowed to be half a nazirite, he is a full nazirite.
וְכִי פְּלִיגִי, בְּלִישָּׁנָא דְמַתְנִיתִין פְּלִיגִי. רַבִּי מֵאִיר סָבַר: כֵּיוָן דְּאָמַר ״הֲרֵי עָלַי״ — אִיחַיַּיב אַכּוּלֵּיהּ קׇרְבַּן נְזִירוּת, וְכִי קָאָמַר חֲצִי נְזִירוּת — לָאו כֹּל כְּמִינֵּיהּ. וְרַבָּנַן סָבְרִי: נֶדֶר וּפֶתַח עִמּוֹ הוּא.
And when they disagree, it is only in a case of one who used the precise wording of the mishna. Rabbi Meir holds that once he said: It is incumbent upon me, he is obligated in all of the naziriteship offerings, and when he later says: Half a naziriteship, it is not in his power to uproot his first obligation. And the Rabbis hold that it is a vow with its inherent opening. By saying he had only half the offerings of a nazirite in mind from the outset, he has nullified his own vow.
The beginning framing of the question, voiced by Rava, is actually in flux. We have, in our printed texts, Rava saying: all agree if he said half of the offerings of a nazir, he’s obligated in half, and if he said the offerings of half of a nazir, he’s obligated in all.
This text matches what Pseudo-Rashi has, with his hachi garsinan:
אמר רבא - ה"ג הכל מודים כל היכא דאמר חצי קרבנות נזיר עלי וכו' כוליה קרבן בעי לאיתויי מ"ט דהא לא אשכחן נזירות לפלגא כי פליגי בלישנא דמתני'
The וכו in the middle is an ellipses, so he is saying that, within Rava’s statement, it is half of the offering of a nazir [would be obligated to half, while the offerings of half a nazir] would be obligated to bring the entire offerings.
Based on the each dibbur hamatchil is Tosafot, they have the identical girsa. Still, Pseudo-Rashi is responding with hachi garsinan to fix some alternative girsa, or to establish this over another.
If we look at the two Talmudic manuscripts of this at the Hachi Garsinan website, Munich 95 and Vatican 110, we see something curious:
First, they have the (geonic) piska, quote of the Mishnah, while our printed texts lack it.
Now Vatican 110 is an “Ashkenazic manuscript from 1381”, from Germany or France, with the following colophon:
יהושעיה בן הרב ר' אברהם בן הרב ר' ברכיה בר' אברהם בר' יוסף ממשפחת יוסף המעוני כתבתי סדר נשים זה לר' ברכיה בר' מתתיה וסיימתיו בי"א יום לחודש שבט קמ"א לפרט, המקום יזכינו להגות בו בניו ובני בניו עד סוף כל הדורות. אמן אמן. ברוך הנותן ליעף כח ולאין אונים עצמה ירבה.
and Munich 95, the only complete Bavli we have, is written in Ashkenazi script, dated 1342, probably in France. The end includes enactments from Rabbenu Tam and Rabbenu Gershon. From Hachi Garsinan’s writeup:
Semi-cursive Ashkenazi script. Written in 1342, probably in France. The textual tradition is Ashkenazic, and follows the interpretation of Rashi in many instances. This manuscript contains many errors (Diqduqe Sofrim) Many names are not written precisely, and the textual tradition of the Talmud was edited in many places on the basis of parallel versions (Friedman).
Thus both of these seem to have been post-Rashi. Still, it is interesting to see what variant texts they have. Vatican 110-111 has:
The differences from the Vilna text are marked in red, and the big red circles mark an omission. The first omission is a big one, going all the way from “Amar Rava”. It doesn’t make much sense — why would he need to bring half a korban, because we don’t find half a nezirut? That answer would explain why someone would have to bring the full korbanot.
In Munich 95, we have:
again with the “amar Rava” omitted at the end. And this also makes little sense, for if he is bringing a [full] korban, why would the reason be that we find half a nezirut? We should emend this omission (marked by the second red circle) to that we don’t find nezirut in half.
It is disturbing that an entire section, with a named Amora, Rava is missing. The printed text (in Vilna with amar Rava, in Venice with amar Rav’ (with the apostrophe indicating Rava) got it from some textual tradition. And while I am OK with lengths of anonymous explication being added (from an original text that lacked it), and would say lectio brevior potior, the shorter text is stronger / more original, for texts with named Amoraim, I would say that, absent a transfer from some related sugya, the one with the name is stronger / more original. Fabricating from whole cloth a statement from an Amora is not something that the scribes would do.
Even though I wrote last night about Rava and interacting with the (early) Stamma,
and this looks like an alternation of Rava with Stamma, this isn’t the case here. It is whether this named statement of Rava ever occurred. And both Pseudo-Rashi and Tosafot have Rashi saying it, and the full text of our the printed edition.
In terms of editing errors, it is much easier to skip over a passage of text (haplography) than to introduce so much innovative text, with an ascription to a named Amora.
It seems that Pseudo-Rashi was responding to these corrupted texts in France and Germany, adding back “Rava” and a sensible explanation. Though either shorter version of the text, as long as you reverse the explanation in the mai taima, would also work.
I also wonder if the loss of Rava could have been sparked by Pseudo-Rashi’s emendation. After all, he begins with “Amar Rava” and then says a hachi garsinan, only establishing the girsa of what follows. A scribe might have misinterpreted that to mean that one should remove Rava from the equation. Also, Rashi used וכו as a shorthand for all the text I put in brackets above. And what happens with these brackets is that we have the position of the former and the explanation of the latter juxtaposed. This might match Munich 95 and Vatican 110, which have contradictory explanations of what they are trying to elaborate upon.