From Oral to Written Talmud - Some Thoughts Based on Professor Danzig's Article.

I just finished reading the very interesting article in Bar-Ilan Annual volume 30-31 (2006) by Professor Neil (Nachman) Danzig of JTS – From Oral Talmud to Written Talmud: On the Methods of Transmission of the Babylonian Talmud and its Study in the Middle Ages. I am no specialist in the field but I am always interested in reading about the latest scholarly developments in this area. After all an understanding of how contemporary Halacha has developed and the sources it uses is intimately tied with how we relate to it and understand it. Here are a summary and selection of the various conclusions Professor Danzig presents in his article and my thoughts as I read it.

Until the time of the Geonim, the Talmud was not written down. It was memorized and repeated orally in the Beit Midrash to be expounded upon and analyzed. The Talmud remained mostly an oral transmission until the 8th – 10th centuries. It is only during the Geonic period that tractates were put to paper (parchment) mainly to be used outside the Yeshivot of Bavel while in the Yeshivot themselves oral study and transmission continued in spite of the availability of manuscripts.

As exact memorization was so crucial, mnemonic devices were developed. We also find customs for different prayers and behaviors, some with magical connotations, with the goal to sharpen and preserve the ability to memorize. One of those is the Hadran we say when we finish a tractate. Professor Danzig elaborates in the article on the methods used for memorization.

A hierarchy developed within the Yeshivot among those who specialized in memorization. They had different specialties though it is not clear what exactly the function of all of them was. There were Tannaim (already referred to in the Mishna) who memorized Mishnayot, Braitot, Sifre and Sifra, Amoraim, Rabanan De’siyuma and Girse’ani who were involved in memorizing the Talmud although it is not always clear what the responsibility of each group was. The latter, the Girse’ani, apparently were responsible to make sure that the versions transmitted were accurate. It was however inevitable that various versions, with slight differences developed. Exactly were there was leeway in the versions is not clear but it appears that it was mostly in what is referred to as the Stama in the Gemara. It is the glue that holds together the discussions of the Gemara and where each Girse’ani could use his own words in expressing the same idea. We however sometimes find differences in Halachik rulings though that is rarer. (For an example, see Rambam Hilchot Malveh Veloveh 15:2 where he apparently records such a case which ended up in the text of the Gemara – see further.)

Apparently, these memorization specialists were authenticated by the Geonim and the head of the Yeshivot and only those approved by them were considered reliable. The others were referred to as Tarbitza’i and their versions dismissed when questions arose. This formal system was crucial to maintain a reliable tradition. It was successful because of this and also the concentration and numbers of people devoted to the maintenance of the system.

As the Jewish communities flourished in places far from Iraq, those communities no longer had the structures established in the Yeshivot of Sura and Pumpeduta which allowed for the reliable oral transmission. As a consequence of this dispersion, there also was a slow reduction of Talmidim in the classic Yeshivot in Bavel, and we read complaints by the latter Geonim of the dearth of dedicated learning and the decline of the generations. The Geonim therefore reluctantly allowed for the writing of the Talmud. The written version was not considered as reliable as the oral system and thus we find letters asking the Geonim for correct readings. As there was more than one oral version, more than one ended up in writing which tells us that the variations we find in different texts may reflect different original versions and not only copying errors or voluntary insertions. (See my reference earlier to Rambam.) This also explains the original reluctance of writing the Talmud. A written text lends itself to amendments while oral transmission in a strictly regulated way is seen as more authentic. (I found interesting that the word Girsa, which I always translated version, really means verbatim repeating or memorization.)

Until the 8th century however, only sections were written as they were requested by the communities. There is no reliable record of a written copy of the whole Talmud. It is only after that time that the rapid dissemination of the complete written Talmud is recorded. It also seems that concurrently with the writing, there was a transition from an oral based to a manuscript-based system. Already during the early Rishonim in Europe, learning was from manuscripts and their unavailability at times, would be seen as a barrier to learning.

Interestingly, though counterintuitive, oral tradition was seen as more reliable than the written text. (I think I recall Maharatz Chayes already discussing this in one of his writings though I would be hard pressed to find it again – I read him about 20 years ago). As people learned from manuscripts, they made interpretive notes and emendations when they encountered a difficult text. These eventually found their way into the Gemara itself sometimes distorting the original idea. Finding a reliable manuscript became a challenge and we find many Rishonim struggling with this.

Based on this presentation, I think that from our contemporary perspective, we need to think of textual variants as not only possible unauthorized inclusions of individual notes and interpretations into the original text or scribal errors but also as variants of the original oral version, at the time it was put into writing. Finding out what type of variant reading a particular version is would help us understand which practice is based on an old tradition and variation thereof and which should be seen as an error. Textual analysis coupled with an in depth logical analysis of a Halacha, is the only way to find the authentic original intent of the editors. If we think about it, the oral system had the advantage that it was not set in stone. In other words, as we see many times in the Gemara, when a statement from an earlier generation was presented in the Beit Midrash, it was dissected and analyzed and many times a corrected version was offered which now became the new memorized text. There is much more flexibility to correct an oral text than when we are confronted with a written one. It is more difficult to deny what one sees than what one hears. Understanding how the written text came to be allows for an open-minded analysis of the subject and a reassessment. The Rishonim were aware of this and did a lot of the work for us. We therefore find them, many times, referring to the reliable and authentic Gemarot of Spain (Ramban many times in his writings especially in the Milchamot), the Gemara written in the handwriting of Rabbeinu Gershon Me’ or Hagolah and many other such instances.

We are indebted to the Rishonim for this work, especially since after the traumas we suffered in Europe over the centuries, with the burning of the Talmud and other sefarim many times over[1] and the censorships we no longer have many variants that could shed light on some difficult texts. The advent of print with its concomitant sometimes ignorant and uninformed choice of corrupt manuscripts as source exacerbates the problem for us.

I have not done justice here to Professor Danzig’s excellent article which I highly recommend to anyone interested in the subject. As usual when I read something I inject my own ideas and prejudices, pick on things that touch me ignoring other important points and I am sure I did not always report the author’s exact meaning. I just wanted to jot down some thoughts I came away with after reading and share it.








[1] There is a record of 24 carts loaded with manuscripts being burned in one session in France (Shevilei Haleket).