Ch. 7 (1) Stages in the Redaction of the Talmud
I. Introduction
The previous chapter dealt with the question of the relationship between the plain meaning of a Tannaitic source and the ukimta in the Gemara. We argued that an ukimta is not necessarily intended to serve as an interpretation, and certainly not the only legitimate one, of the Tannaitic source, and that an ukimta might serve different purposes – even when it is clear that it does not accord with the plain sense of the Tannaitic source.
However, the ukimta phenomenon is not expressed only in relation to Tannaitic sources, as we also sometimes find it in the relationship between the Gemara and certain Amoraic statements. Here again, the question arises: Is it possible to interpret an Amoraic statement in accordance with its plain sense, even in cases where this interpretation differs from the Gemara’s explanation?
There are several parts to this question. First, it depends on the basic question of how the Babylonian Talmud came into being, for it is rooted in the assumption that the Talmud itself is built of several layers, separated by time. Second, the question must be examined on its own, as we did with regard to the Mishna. And finally, we must also ask the educational-methodological questions: Is there value to this type of study, and at what stage in the learning process should this issue be raised, if at all? This last question was in the past the subject of a public debate, and it deserves a discussion of its own.
As part of this discussion, we will relate more broadly to how the Talmud was redacted, including those layers that came later than the redaction, and the fundamental and halakhic significance of these layers.
II. The Redaction of the Talmud
How was the Talmud compiled? Neither in the Talmud itself, nor in all of Rabbinic literature, do we find any explicit references to this question, and it is only natural that speculations and assumptions have abounded on the subject. We will briefly review the various approaches that have been proposed.
The basis for the entire discussion is an obscure statement appearing in the Gemara:
Rabbi [Yehuda Ha-Nasi] and Rabbi Natan are the end of the Mishna; Rav Ashi and Ravina are the end of teaching [sof hora'a]. (Bava Metzia 86a)
What is the meaning of the phrase "sof hora'a" in reference to Rav Ashi and Ravina?
Rashi (ad loc., s.v. sof) explains as follows:
Sof hora'a – the end of all the Amoraim. Until their days, the Gemara was not [arranged] in a systematic order. Rather, when a question was raised in the beit midrash about the reasoning of a mishna, or about a practical case of civil law or ritual law, each sage expressed his opinion. Rav Ashi and Ravina arranged the teachings [shemu'ot] of the Amoraim who preceded them, and arranged them according to the order of the tractates, each one next to the appropriate mishna. And they raised valid objections and proposed appropriate solutions, they and the Amoraim who were with them, and they fixed them all in the Gemara, [introducing them with certain phrases,] for example: "He raised an objection against him," "they raise an objection," "they pointed out a contradiction," "they were asked," and the solutions for them, what was left by those who came before them. Those who raised objections and offered solutions before them did not fix them in the Gemara according to the order of the tractates of the Mishna arranged by Rabbi [Yehuda Ha-Nasi], and Rav Ashi and Ravina came and fixed them.
Rashi's words imply that before the time of Rav Ashi and Ravina, there were only shemu'ot – that is, Amoraic statements and short discussions about them, which dealt with questions that arose either as part of the study of the Mishna in the beit midrash, or in the wake of a case that was brought before the Sages for discussion. At the end of the Amoraic period, Rav Ashi and Ravina collected all of the original "shemu'ot," edited them, arranged them according to the order of the tractates, and added more objections and solutions. According to Rashi, then, the Talmud had editors, just as the Mishna did – namely, Rav Ashi and his disciple-colleague Ravina.[1] Indeed, elsewhere it is stated: "From the time of Rabbi [Yehuda Ha-Nasi] until the time of Rav Ashi, we do not find anyone who was supreme in both Torah and worldly affairs" (Gittin 59a). Rashi's position appears already in an early work, Seder Tannaim ve-Amoraim,[2] where it is briefly stated:
Wherever in all the teachings you do not find Abaye or Rava or other Amoraim, infer from this that Rav Ashi and Ravina established it. (Seder Tannaim ve-Amoraim, part II, no. 83)
However, Rav Sherira Gaon in his epistle[3] takes the statement in an rather different direction:
And in each generation of these generations, there were statements [shemu'ot] that the Rabbis learned and repeated over and over… And those statements that were learned by everyone, every Rabbi would offer his students different explanations of them and additional inferences. They would also introduce new positions, issuing practical rulings on questions that they were asked and they discussed. There were also disputes between each one and his partner. Then another generation arrived, and their hearts [=minds] grew smaller, and those things that were simple to the earlier Sages, and that they explained to their students, and they were like those explanations that everyone does not need to learn and set them in the Gemara – they were now in that generation in doubt, and they had to set them in the Gemara… And in every generation, the hearts grew smaller… And to the extent that the hearts grew smaller and doubts arose, those explanations of the early Sages that were not set down in their days, were set down now and learned, and the practical rulings that our early Rabbis issued were now learned in the Gemara… With these things the Talmud grew generation after generation, for every generation set down in the Talmud matters in doubt that arose for them, and rulings and questions that they were asked about… In this way teachings were added generation after generation until Ravina, and after Ravina this stopped, as Shmuel Yarchina'a saw in the book of Adam, in which it was written: Ashi and Avina were the end of teaching.[4]
According to Rav Sherira Gaon, the Talmud was a multi-generational work, which continued to expand over the course of several generations. In the first stage, the Talmud was quite limited, but due to generational decline, the explanations that had accompanied it without being part of it became essential, and thus they penetrated into the text and expanded it. This phenomenon repeated itself over the course of several generations, and together with these interpretive expansions, deeds performed by the Amoraim from the various generations were also added. This phenomenon eventually ended in the time of Rav Ashi and Ravina. According to this approach, the statement "Ashi and Avina were the end of teaching" is a historical account of when the phenomenon of the expansion of the teaching part of the Talmud came to an end,[5] not a declaration that Rav Ashi and Ravina were the editors of the Talmud.[6]
Modern Talmudic scholars have also been divided on this question: Some maintain, based on the words of Rav Sherira Gaon, that the Babylonian Talmud did not undergo any editing at all, for "there is no trace whatsoever in the words of the Gaon of the notion of arranging the Talmud in any sense, and no mention of Rav Ashi as arranger of the Talmud,"[7] and therefore the conclusion is "that the Talmud as a complete literary creation leaves no room for the assumption that there was some final arrangement of the Talmud, because the entire essence of the Talmud resting before us proclaims coming into being and development, and not final arrangement."[8] Most modern scholars, however, follow the approach of Rashi, that there was in fact significant editorial activity on the part of Rav Ashi and Ravina, which included the earlier stages of the Talmud.[9] This editorial activity was summarized by Y. N. Epstein as follows:[10]
The Babylonian Talmud as well is not entirely of a single shade: Its systematic arrangement began with the early Amoraim, who arranged the baraitot next to the mishnayot, explained the Mishna with short statements, and issued halakhic rulings ("These are the words of So-and-so, but the Sages say,"[11] "the halakha is in accordance with So-and-so,"[12] "the opinions should be reversed,"[13] "this is not an [authoritative] mishna,"[14] "I learn it as the opinion of an individual,"[15] and the like). They added laws that are not found in the Mishna, derivative laws, and set them in the Talmud" ("have you established this as part of your study").[16]
We see remnants of the arrangement of the Talmud already from the time of Rav Nachman,[17] Rav Yosef, and their disciples, Abaye and Rava; the arguments of Abaye and Rava;[18] the statements multiplied and the discussions grew; they brought statements and passages from the Land of Israel to Babylonia and from Babylonia to the Land of Israel; the yeshivot split apart, and each one taught the Talmud in its own way, according to its own opinion ("In Sura, they taught in this manner, whereas in Pumbedita they taught in that manner,"[19] and the like; "they said another version”;[20] "there are those who point out a contradiction";[21] "there are those who said it using this formulation";[22] "Rabbi Peloni taught thus, and Rabbi Peloni taught thus"[23]). And the "arrangers" (such as Rav Nachman bar Yitzchak,[24] "the head of the Kalla in Rava's yeshiva in Mechoza,[25] the head of the Sidra[26]) arranged it – until the time arrived for Ravina and Rav Ashi, "the end of teaching," "the arrangers of the Talmud," who collected all the material that preceded them, mostly in its [original] form (just as Rabbi [Yehuda HaNasi] did with the Mishna), explained it, completed it, and "arranged" it. (Mevo'ot le-Sifrut ha-Amoraim, Jerusalem 5723, p. 18)
It seems therefore that both according to Rashi and according to Rav Sherira Gaon, the Talmud is comprised of several layers; the Talmudic passages include a basic layer of earlier Amoraic statements, and the "shemu'ot" connected to them, and a later layer, which was added by the redactors of the Talmud.[27] In addition, there is also a Savoraic layer, which will be discussed in a later shiur.
(Translated by David Strauss)
[1] As Ravina himself said to Rav Ashi: "And I too am the master's colleague as well as his disciple" (Eiruvin 63a).
[2] Seder Tannaim ve-Amoraim is a work of unknown authorship from the Geonic period, but it stands to reason that it was composed by one of the Geonim of Sura. The book is divided into two parts: the first part deals with development of the Oral Law until the end of the period of the Savoraim, and the second deals with the rules of halakhic decision-making in the literature of Chazal.
[3] Spanish recension, pp. 62-69. We discussed Rav Sherira's epistle in the previous chapter, (shiur 40, note 8).
[4] Based on Metzger edition, paragraphs 72-78).
[5] Opinions differ as to the precise meaning of the phrase: "sof hora'a." See Sh. Z. Havlin, "Al ha-Chatima ha-Sifrutit ke-Yesod ha-Chaluka li-Tekufot be-Halakha," in: Mechkarim be-Sifrut ha-Talmudit, Jerusalem 5743, p, 161, note 57.
[6] Sources also disagree about the identification of Rav Ashi and Ravina. It would seem from Rashi that the reference is to the famous Rav Ashi who lived at the beginning of the fifth century C.E. and was a member of the sixth generation of Amoraim. In that case, Ravina, his disciple-colleague, was older than him. However, Rav Sherira in his epistle writes (para. 105): "And on Wednesday, the 13th of Kislev, in the year 811, Rabbana Avina the son of Rav Huna, who was the end of teaching, died." According to him, the Ravina in question lived later; he was a member of the eighth and final generation of Amoraim, and according to the Spanish recension, died in 811 to the count of contracts, which is the year 499 C.E. It stands to reason that Rav Sherira Gaon also identified "Rav Ashi" in a different way; indeed, when he mentions the well-known Rav Ashi in the chronology of the Amoraim, he notes that Rav Ashi served as Rosh Yeshiva for close to sixty years (p. 93), but does not mention anything else special about him – while regarding (p. 97) Rav Yose (in the Spanish recension) or Rav Asi (in the French recension), he writes: "And in his days was the end of teaching and the Talmud was completed," and the reference is to the Rosh Yeshiva of Pumbedita in the days of the later Ravina. Regarding the identification of Rav Ashi and Ravina, see also Rabbi Y. E. Halevi, Dorot Rishonim, III, Pressburg 5657, p. 18; Ch. Albeck, "Sof Hora'a ve-Siyyum ha-Talmud," Sinai – Sefer ha-Yovel, 5718, p. 77; A. Weiss, Hit’havut ha-Talmud bi-Shleimuto, New York 5713, pp. 248-253.
[7] Weiss (above, note 6), p. 249.
[8] Weiss (above, note 6), p. 245.
[9] For a concise summary of the various opinions on the matter, see Sh. Z. Havlin, Mesoret ha-Torah she-be-Al Peh, Jerusalem 5772, pp. 429-431.
[10] Along with the explanatory notes that I have added.
[11] Such statements are reported dozens of times in the name of early Amoraim, especially the following three: Rav (Eiruvin 41b, 48b; Ta'anit 18b; and elsewhere), Shmuel (Eiruvin 49a; Yevamot 67a; Ketubot 57a; and elsewhere), and Rabbi Yochanan (Shabbat 86b; Rosh ha-Shana 9b; Megilla 2a; and elsewhere).
[12] Such statements are also reported dozens of times in the name of early Amoraim, and here too, mainly in the name of the three Sages mentioned in the previous note: Rav (Yevamot 37a, 60a; Ketubot 6a; and elsewhere), Shmuel (Yevamot 18b, 108a; Ketubot 6a; and elsewhere), and Rabbi Yochanan (Berakhot 13b, 14a; Yevamot 42b; and elsewhere).
[13] This expression is usually associated with Rabbi Yochanan (Berakhot 17b, 49b; Eiruvin 99a; and elsewhere), and sometimes also with Rav (Ketubot 24a; Nedarim 61b; Kiddushin 64b).
[14] We discussed this expression in the previous chapter (section III), and it too usually appears in the name of Rabbi Yochanan (Shabbat 106a; Yevamot 43a; Chullin 82a; and elsewhere).
[15] This expression is unique to Rabbi Yochanan (Pesachim 84a; Chullin 55b, 134a).
[16] The Gemara (Eiruvin 32b) records a discussion between the Amoraim about how to interpret the mishna there. After Rav Nachman confirms the interpretation of his students, they ask him: "Have you established this as part of your study?" and he answers them in the affirmative, and the Gemara brings as a proof the statement of Rav Nachman in the name of Shmuel, who agrees with what the students said. What this means is that the matter was included as a fixed element in the text that was studied as an expansion of the mishna. This is one of the examples brought by Rav Sherira Gaon, in the passage cited above, to show how the Talmud kept expanding in size over time.
[17] As stated in the previous note.
[18] The Gemara (Sukka 28a) mentions the concept of "the arguments [havayot] of Abaye and Rava," which seems to refer to "objections and resolutions and doubts that were not clarified to them" (Ritva, ad loc., s.v. davar katan). Epstein assumes that the reference is to a familiar concept that was known even before the redaction of the Talmud.
[19] Megilla 18b; Gittin 35a; Bava Batra 142b; and elsewhere. This expression attests, of course, to the existence of a tradition of learning that differed slightly between the various schools. Thus, for example, it is stated in Megilla 18b: "Rabbi Abba said in the name of Rabbi Yirmeya bar Abba who said in the name of Rav: The halakha is in accordance with Rabbi Muna. And Shmuel said: The halakha is not in accordance with Rabbi Muna. In Sura, they taught in this manner, whereas in Pumbedita they taught in the following manner: Rav Kahana said in the name of Rav: The halakha is in accordance with Rabbi Muna. And Shmuel said: The halakha is not in accordance with Rabbi Muna."
[20] Eiruvin 103b; Sukka 11a; Yevamot 35a; and many other places. In these cases as well, as in the cases brought in the previous note, the existence of different "versions," even without mention of a particular school, attests to different traditions that existed before the redaction of the Talmud.
[21] Rosh ha-Shana 27b; Sukka 16a; Ketubot 27a; and elsewhere.
[22] Nedarim 34a, and a similar expression in Shabbat 40a: "Rabba taught this ruling of Rav in the following formulation"; Pesachim 107a: "Ameimar commenced this ruling of Rabba in the following formulation"; and elsewhere.
[23] Shabbat 129a; Eiruvin 26b; Ketubot 100a; and many other places.
[24] Rabbi Nachman bar Yitzchak stated: "I am neither a scholar, nor a visionary, nor unique [in this ruling]; rather, I am a teacher and a [systematic] arranger [ve-sadrana]," as Rashi explains there: "Ve-sadrana ana – I regularly arrange rulings before my masters." Many have concluded from here that Rav Nachman bar Yitzchak, an Amora of the fourth and fifth generation, had a special role in the arrangement of the Talmud. To this we can add the fact that many passages contain a mnemonic device given by Rav Nachman bar Yitzchak, in order to remember the positions of the Sages on various topics. For example, "And your sign is samekh samekh” (Shabbat 66a, where the sign indicates that it is Rabbi Yose who forbids [oser], since the letter samekh appears in both words: Yose, oser). See also Shabbat 60b, 66b; Ta'anit 10a; and elsewhere.
There are, however, those who reject this conclusion and maintain that the phrase gamrana ve-sadrana has no bearing on the issue of editing, but rather stands in contrast to chakhima'a, that is, a person who is fluent in the traditions; see, for example, A. Weiss, Mechkarim be-Talmud, Jerusalem 5735, pp. 51-54; a summary of the discussion can be found in Havlin (above, note 5), p. 126, note 45.
[25] See Bava Batra 22a.
[26] The Gemara in Chullin 137b mentions the term reish sidra, which in Babylonia refers to the Rosh Yeshiva (the reference there is to Rav).
[27] We will not go into detail here on a question that has been discussed extensively in the last generation, regarding the time of "stama de-Gemara," the anonymous frame of the Gemara. D. Halivni (in many of his writings, and in detail in the introduction to his book, Mekorot u-Mesorot on tractate Bava Batra, Jerusalem 5768, pp. 1-144) has argued that the stama de-Gemara was not composed by Rav Ashi, Ravina, and the members of their generation, but rather at a much later time, between the middle of the sixth century and the middle of the eighth century. Halivni innovated the term "Stamaim" for those who created the majority of the Talmudic discussion. According to him, in this way it is possible to understand the various difficulties found in the ukimtot in the anonymous frame of the Gemara, which, in his opinion, stem from two reasons: "A very large time gap between the Amoraim and their Torah, and the Stamaim who explain it; as well as the lack of a tradition regarding the handing down of the Talmudic discussion" (p. 2). Elsewhere he writes (p. 15): "This is only because the forced explanations in the words of the Amoraim stem from the fact that the Stamaim did not have sources in their entirety or in the correct order; and to complete them and fix them, they had to give forced explanations." Halivni has also argued that the well-known statement, "Rav Ashi and Ravina are the end of teaching," "does not reflect a historical fact, but rather words of admiration of a student who naively thought that with the passing of his master teaching has come to an end" (p. 6). According to him, the Stamaim parallel the period of the Savoraim, and the term "Savoraim" relates to the last of the Stamaim.
Halivni's approach has been fairly widely accepted in the world of scholarly Talmudic research, but it faces difficulties from several directions, and various scholars have challenged it. First, his argument is contradicted by all the sources that we saw above, first and foremost of which is the epistle of Rav Sherira Gaon, who assigned the end of the period of the redaction of the Talmud (to the exclusion of the additions of the Savoraim) to the days of Rav Ashi and Ravina, and there is no historical basis upon which the hypothesis of "Stamaim" can rest. Second, as noted by Y. Brody ("Stam ha-Talmud ve-Divrei ha-Amoraim," in Iggud – Mivchar Ma'amarim be-Mada'ei ha-Yahadut, I, Jerusalem 5668, pp. 213-222), there are many proofs that there are early "anonymous passages" that the Amoraim were familiar with and related to. So too, the assumption that the gap between the plain meaning of a certain Amoraic statement and its interpretation based on the stama de-Gemara proves distance in time is also unfounded, for such phenomena may also occur at the same time at which the words were said. And furthermore, there is no proof from this of a uniform time for all of the stama. Brody rejects Halivni's approach, arguing: "While there are many anonymous passages that do not precede Rav Ashi, there are also quite a few anonymous passages concerning which it is possible, in my opinion, to 'prove' that they were formulated already in his time, at the latest, whereas regarding the overwhelming majority of anonymous passages, it is impossible to decide even about this question… Whenever we start studying a passage, we must maintain maximum interpretive openness and examine impartially and without prejudices the full range of possible relationships between its components, both the anonymous ones and those attributed to particular Sages. In many cases we will have to be satisfied with noting the possibilities and we will not be able to reach a clear determination, but it is better that we teach ourselves to say 'I do not know," than to follow blindly magical formulas that offer overly sharp and simple solutions" (ibid., pp. 224-227).
This dispute has been discussed at length in various articles. See, for example, Sh. Y. Friedman, Sugyot be-Cheker ha-Talmud ha-Bavli, New York and Jerusalem 5770, pp. 57-135.
In any case, for our purposes, this dispute, with all its importance not only from a historical perspective, but also in principle and conceptually, is not particularly relevant. All acknowledge the existence of different layers and of a certain time gap between the Amoraic statements and the stama de-Gemara, and the need to address the phenomenon is essential, whenever it occurred and whatever name we give to it.
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!