In the Babylonian Talmud, authority comes in variety of flavors. Sometimes a tradition, heard from and cited in the name of a teacher, carries the day. At other times, logic wins. The behavior of a rabbi, the opinion of an expert, or the common practice of a community sometimes drive a discussion about law or ethics. But the trump, as anyone who has spent any time with the Bavli knows, is the Bible, especially the Torah. While it is certainly true that rabbis often turn and twist biblical verses as origami masters might, it is always better to have a verse on one’s side.
How, though, did the rabbis of late antiquity “know” the Bible? Did they have the whole thing memorized? Did they consult scrolls? Did their versions look like ours? Did they gravitate toward certain verses or sections, or steer clear of others? If so, why?
For me, these questions arose quite incidentally about a year ago in the context of an informal Talmud reading group. I figured that at least the empirical questions were easy to answer. Somebody, somewhere, must have compiled a list of the biblical verses in the Talmud and counted them up in various ways.
If such a study exists, though, I still cannot locate it. There are tools that indicate where in the Talmud a particular verse is discussed, but no charts, tables, and graphs that I could find helped very much when it came to quantifying the Talmud’s use of the Bible. So as a side project I began to assemble the data.
This turned into a more involved undertaking than I anticipated, but it is very close to completion. My crack research team – my son Dani Satlow and Elijah Petzold, a very talented Brown undergraduate – has now logged every biblical verse cited in the Bavli in a spreadsheet. The method for doing this was not perfect: we went copied the indices of each of the tractates published in the Schottenstein edition of the Talmud. We corrected obvious errors (mainly typos) as we went, but I suspect that the indices contain additional mistakes that are now incorporated into our spreadsheet (while undoubtedly introducing new ones of our own). Nevertheless, given the mainly quantitative goals of the project and the large numbers present, these errors should not significantly distort the results.
My next step is to figure out good ways to use this data (which I will make freely accessible, probably by the end of the semester), and here I welcome your advice. The three top questions on my list are:
- What is the most commonly cited verse in the Talmud?
- Are there verses, chapters, or books that the Talmud never cites?
- What is the density of biblical citations per tractate?
What would you like to know?
I generated the above image using Wordle, with random text from the beginning of the Talmud. Wordle might itself be useful for research; perhaps a future post on that.
Michael Satlow is a professor of religious studies and Judaic studies at Brown University and has been a mentor and sounding-board for the New Talmud Blog from the beginning. This post was crossposted from his own blog, Then and Now.
The Talmud Blog 
Neusner’s Talmud Bavli and Yerushalmi has a very good listing and index of Torah passages.
Jerry Rosenberg
Interesting. Looking at that image, one clearly sees that the source text was from the first perek of Brachos. I wonder what an image like that of the entire Shas would look like.
I think it would be interesting to break it down by stratum, although it may be too late in the game to do this. For example, in what attributive context are verses being presented? Depending on the methodological constraints one sets up, one could examine whether certain sages/stam-voices are more or less likely to quotes certain verses, or verses from certain contexts/books.
Also, I’ve always had a suspicion ever since I studied Chronicles that familiarity with Scripture might be related to liturgical issues; in other words, one might be more familiar with a verse, or group of verses if it crops up often in other contexts (to give an imperfect analogy: almost everyone I grew up with knows Psalm 130 by heart, but very few could even recite a single verse from Psalm 129). It might therefore be interesting to correlate the relative density of quotations from given Biblical contexts with whatever data we have on other contexts – especially liturgical, I suppose – in which these verses pop up.
In any case, this is a pretty exciting project!
The Chassidim would mock the Misnagdim saying that they love learning chezkas habatim because there is a very long stretch of no pesukim….
I find it fascinating to look through Y. Sh. Y. Chasida’s anthology פירושי החסידות לנ”ך, to see which books get the most coverage. (I have heard, unfortunately, that this anthology contains only a small percentage of what it could contain. Nonetheless, it could still be representative.)
Tthe volume on Kethuvim breaks down as follows:
115 pages on Psalms (for 80 pages of Tanakh text in the Koren ed.): 144%
107 pages on Proverbs (for 26.5 pages of Tanakh): 404%
37 pages on Job (for 32 pages of Tanakh): 116%
28 pages on Song of Songs (for a bit over 5 pages of Tanakh): 560%
15 pages on Ruth (for 5 pages of Tanakh): 300%
21 pages on Lamentations (for 6 pages of Tanakh): 350%
50 pages on Koheleth (for 12 pages of Tanakh): 417%
33 pages on Esther (for 13 pages of Tanakh): 254%
11 pages on Daniel (for 24.5 pages of Tanakh): 45%
11 pages on Ezra-Nehemiah (for 39 pages of Tanakh): 28%
18 pages on Chronicles (for 102 pages of Tanakh): 18%
It’s not surprising that the megilloth get so much attention — after all, they are read liturgically on particular holidays or seasonal events, so there is opportunity for sermons. (It is interesting, though, how little attention Esther gets, since Hasidut is reputed to ascribe great significance to Purim.)
What is fascinating, though, is that there is so much interest in Proverbs.
(I flicked through the Nevi’im volume — the comments seem to be almost all on Haftaroth and other liturgical passages, e.g. the piece of Joshua 24 included in the Pesach Haggada. Makes total sense, of course.)
I’m reminded of the old yeshiva joke… “it’s a pasuk in Bava Kamma”
Gerald’s idea is a good one, especially when it comes to the Yerushalmi. Anyone want to do it? Crowdsource it?
Reb Chaim, I wonder too! All you have to do is cut and paste a copy of the Shas into Wordle. Then by all means share your results with us!
I like Ari’s suggestion. Once we have a good list of cited verses it can be crosschecked. If anybody has a digital file of “verses cited in liturgy” that could even be automated.
Jeffrey Hoffman wrote a DHL thesis at JTS on the Bible in liturgy. I’ll check with him and see if he has an index.
I don’t have an index, but I’ll add the following bibliographical note:
Richard Sarason, “Midrash in Liturgy,” in Encyclopedia of Midrash: Biblical Interpretation in Formative Judaism, ed. Jacob Neusner and Alan J. Avery-Peck; 2 vols. (Leiden: E.J. Brill, 2004), 1:463-492.
One problem is that we don’t know terribly much about the state of the liturgy at the time of Hazal. It seems that Palestinian liturgy contained far more Biblical verses than Babylonian; this was certainly the case in later centuries. Are more verses quoted in the Yerushalmi than in the Bavli?
In Shabbath 88b ff., there’s a long list of aggadic statements about Mattan Tora, all by R. Joshua ben Levi. (These follow a number of other aggadic statements about Mattan Tora, starting on 88a, not by R. Joshua ben Levi.) Although these statements quote passages from all around the bible, two Biblical hypotexts stand out particularly, for they are quoted several times: the Song of Songs, and Psalm 68. In many Jewish liturgical rites, including many (most?) Sephardic Rites today, Psalms 68 is used liturgically on Shavu‘oth. I think that it is quite likely that this could have been the case in the days of R. Joshua b. Levi, too (assuming that the connection between Shavu‘oth and Mattan Tora was secure by that time) — and that this whole list of statements could have been a long homily delivered on Shavu‘oth by R. Joshua ben Levi.
And what about the Song of Songs? Well, that could have also been part of the Shavu‘oth liturgy of the time, though later practice was to read it on Passover.
The Yerushalmi definitely has more of a tendency to learn halakhot from verses than the Bavli. I suspect that the Bavli stam shifts almost completely to sevara.
Thanks, Zohar! That’s fascinating parallel to the liturgical situation. Has anyone noted this before?
As I recall, David Stern is working on a similar project for midrashim. He gave a paper on it at Yad Ben Zvi over the summer. If one were to work on stratification in the Bavli, it would be interesting to merge this project with a midrashic index to get a broader picture.
I wonder if there is any comparable index for Syriac works. I know Joel Thomas Walker has a list of cited passages in his edition of the Acta Mar Qardagh, and I’m pretty sure (although I don’t have it in front of me) that Sebastian Brock has a list of cited passages in his edition of Mar Ma’in. I bet someone has done an index for the scholastic texts, Aphrahat, etc. Might be interesting to compare to the Bavli.
…. and of course there is Piyyut that stands in between talmudic/midrashic literature and the liturgy. Here one finds biblical verses that are quoted verbatim (mostly in the so-called Sharsheret Pesuqim [chain of verses]) but also all over the poems in more fluid forms. I am embarking these days on a digitalized database of (early) piyyut so perhaps some day in the future we will be able to run queries like that one.
These comments point to the next stage of most humanities computing projects: having datasets “talk” to each other. Now that we are well along on developing digital corpora, the question is how those corpora interact with each other to produce exciting new research.
In any case, just for the fun of it I’ll start selectively releasing results of the project in my blog. How many times does the Bavli cite Scripture? For the answer, go here: http://74.220.215.212/~mlsatlow/?p=331
P.S.
I highly recommend this (rather long) post by Stanley Fish on digital humanities and interpretation:
http://opinionator.blogs.nytimes.com/2012/01/23/mind-your-ps-and-bs-the-digital-humanities-and-interpretation/
מאמר מעולה
תודה!
Fascinating. I would just point out that “chapters [that it never cites]” would be misleading, as of course they had no common chapters with us, and it isn’t even completely clear that their parshiyot were identical with the masora.
To add to the yeshivishe jokes, we shall see if it’s true that by looking up the pesukim and the one before and after if you can really learn Tanakh.
R’ Hillman wrote an essay about this, published in the last last issue of Yeshurun.
it isn’t even completely clear that their parshiyot were identical with the masora.
Do you mean “parshiyot” as in paragraphs, or as in weekly Sabbath portions?
Regarding the latter, almost all the Babylonian Sabbath portions were identical to ours. (A major exception is פרשת השכם — that is, the first four parashiyyoth of Exodus were שמות, וארא, השכם, בשלח, rather than שמות, וארא, בא, בשלח. Although many studies have been written which mention this tangentially, as part of their respective fields of research, Yosef Ofer has a comprehensive discussion of it in his book המסורה הבבלית לתורה.)
If you mean “parshiyoth” as in paragraphs, then there was no set list of “the Masora”, just various individual traditions. In Tiberias, they didn’t even attempt to standardize this;* in Babylonia, they attempted to make authoritative lists, but it seems that they never really came to a standardization. (I think there’s an article about it in the recent volume of articles in memory of Yisrael Yeivin.)
*At least, this is what Rav M. Breuer and Dr. M. Cohen write, and it seems to be a fact. Raises enormous problems in theology of halakha, because a scroll of the Torah or of Nakh in which the paragraphs have been broken incorrectly is pasul. And yet we have no ancient list of the standard paragraphs! Eventually, Maimonides solved this problem by fiat, by declaring that the paragraphs in the Aleppo codex would henceforth be the standard. Remarkably, the Jewish communities around the world accepted this. This solved the practical halakhic problem, but not the theological one (i.e., that the Torah is supposed to be a copy of some authoritative text from Sinai, and the Nevi’im / Kethuvim copies of authoritative texts from antiquity.)
What about the infamous book כל המקראות שבתלמוד הבבלי?
http://www.bookgallery.co.il/content/hebrew/bookpageschema.asp?bookpageid=68445
Also Aharon Heyman’s ספר התורה הכתובה והמסורה lists the uses of each verse in all of rabbinic literature.
Cool. What’s the controversy?
Pingback: The Bible in the Bavli: Some First Numbers- Guest Post by Michael Satlow | The Talmud Blog
Pingback: An Index of Biblical Verses in the Liturgy « Menachem Mendel