Invoking Tradition

So, I recently described Rav Yehudah’s trip down to Pumbedisa, where he founded a yeshiva where learning was based on the dialectic method, the style of shaqla vetarya (question and answer) that typifies the gemara. So what was happening back in Eretz Yisrael?

To put the same question in a more straightforward way — after learning the first six mesechtos of Talmud Yerushalmi, I would like to discuss some differences in style I noted between the two talmuds.

The Talmud Yerushalmi, despite the name, was written in Northern Israel. In fact, you hear disdain for the “deromai”, the rabbis of the south, those in Judea — including Yerushalayim. As well as those in Babylonia. The work on it begins with Rabbi Yochanan and his student and brother-in-law Rabbi Shimon ben Laqish (“Reish Laqish”), in the very first generation after Rebbe compiled the mishnah. Rabbi Yochanan started out in Tzippori, which was the center of learning and housed the Sanhedrin. However, as a young rabbi, his shiur grew so popular that it caused friction with the older rabbis both due to competition for students and the number of disputes. So rather than hurt them, Rabbi Chanina was named in particular, Rabbi Yochanan moved to Teveryah. (I think something should be made of the point that R’ Chanina, the example of the older generation, is named in past tense, whereas R’ Yochanan has pretty much the future tense version of the same name. The Jerusalem Talmud is thus really a product of his school in Teveryah and the rabbis of either Katzrin (in the Golan) or perhaps Ceasarea.

(Side note: Rabbi Yochanan and Reish Laqish play a parallel role to the one Abayei and Rava do in the Bavli. They started the process that over a century later culminates in the talmud. More so than their being like Rav Ashi and Ravina, who close the vast majority of the text, with only light refinements by the savora’im. However, R’ Yochanan and Reish Laqish appear far more often, at least in the first six mesechtos of the Y-mi, than Abayei veRava do in the Bavli.)

One of the more pronounced differences one sees between the two talmuds is the greater reliance on tradition one finds in the Yerushalmi, and I intend to save other differences for another post.


I don’t just mean that because there was no Bavli-style shaqla vetarya, the Yerushalmi fell back onto relying more on sources. Rather, in Eretz Yisrael the feeling was that citations and word-of-mouth transmission is more reliable and more meaningful than relying on reasoning. And this mode was more viable for people who were actually living on the same land and using the same institutions as the tannaim. We saw that Rav Yehudah held it was prohibited for his students to move from Pumbedisa to Israel, as the loss of deep reasoned learning was too great. But also, that kind of reasoning was more critical in Bavel, where there was that extra discontinuity.

As a metaphor, I am reminded of Rabbi Yehudah haLevi’s description of Greek Philosophy:

(יג) אמר החבר: מה שאתה אומר נכון הוא בנוגע לדת המיסדת על ההגיון ומכונת להנהגת מדינה דת הנובעת אמנם מן העיון אך נופלים בה ספקות רבים ואם עליה תשאל את הפילוסופים לא תמצאם מסכימים על מעשה אחד ולא לדעה אחת כי דת כזאת בנויה על טענות אשר רק חלק מהן יכולים הפילוסופים להוכיח במופת ואלו על אחרות נתן להביא ראיות מספיקות בלבד ויתרן אין להביא עליהן אפילו ראיה מספקת אף כי להוכיחן במופת: (tr. R’ Yehudah ibn Tibon)

13. The Rabbi: That which thou dost express is religion based on speculation and system, the research of thought, but open to many doubts. Now ask the philosophers, and thou wilt find that they do not agree on one action or one principle, since some doctrines can be established by arguments, which are only partially satisfactory, and still much less capable of being proved. (tr. Hartwig Herschfeld)

Rav Yehudah haLevi says that the dependence on reasoning introduces uncertainty, and is a stop-gap  for those who do not have a tradition. Taking this from philosophy to be a metaphor for halachic styles, this parallels Rabbi Yochanan’s view, as opposed to Rav Yehudah’s down in Pumbedisa.


We see this also in terminology. The term used in Bavel for a volume of the talmud is masekhes, from the word that is also used to refer to the network of threads on a loom. Whereas in the Yerushalmi, the more common term is meikhlah, a measure or portion. (E.g. Shevi’is 10:3, Vilna 30b – R’ Yosi rules that someone who knows one meikhla who comes to a new town where they honor him for knowing two, must confess to only knowing one.) None of the implication of constructivism or filling in the holes in the network that one finds in meseches.

Another example is the difference in meaning of the keyword “ta’ama“. This appears so often in the Bavli, that I am still pretty consistently thrown by the Yerushalmi. When the Bavli asks “Mai ta’ama?” it is asking “What’s the reason?” And the answer will be the logic behind the ruling. However, in the Yerushalmi, “Mai ta’ama?” is answered with a citation of the verse that is the source of the law.

The word ba’ei is used in the Bavli in the sense of “wants to know”, introducing a question. It sometimes appears as be’ai lemeimar, “wants to say”, introducting a suggested novellum. In the Yerushalmi, just the word ba’ei is used for both, and the “wants to say” sense is far more common. In light of their distrust of deductions about existing laws, it makes sense that the rabbis of Teveryah would be more prone to mark them. It also would explain why they didn’t find it confusing to use the same keyword for “R’ X wants to know, is the halachah A?” and for “R’ X wants to say the halakhah is A”.  The difference between not knowing at all, and suggesting something without a basis, apparently wasn’t seen as that being all that great. Since otherwise confusion would have driven one of them away from using the word “ba’ei“.

(Linguistic side note: In the Yerushalmi, it seems that words were pronounced with an emphasis on the later syllables, whereas in the Bavli, the emphasis was moved forward. So, the same amora we know in the Bavli as Rav Avin, the Yerushalmi calls Rabbi Bun. Note how in Bavel, the “-i” fell off the end of “Rabbi”, so that the amoraim there are called “Rav”, and in Israel, the leading letter fell off. Similarly, R’ Elazer becomes R’ Lazer, R’ Yehudah – R’ Yuda, or sometimes, with a more Greek ending, R’ Yudan, but in either case, the hei slurs away. More dramatic — and often confusing — is when the amora the Bavli refers to as Rav Ila is called Rabbi Lo, and one has to guess whether R’ Ila is being quoted, or Rebbe is saying something starting with the word “lo” [no]. We also see alefs simply dropped out of the spelling of words and other inconsistencies, but we don’t know how many of these are simply do to inferior transcription. Unlike what I noted about names, where there is a clear pattern.)

Quoting Style:

So, given the value placed on quotes and citations, the Yerushalmi sometimes quotes entire sections repeatedly. Not just repeating the one opinion, but including the whole dialog. Usually, this is at most two or three repetitions of the same dialog in the same mesechta, each time to bring out a different point related to the current topic. But a stark example, in Mes’ Shevi’is daf 10, an entire segment is repeated in two versions that are only trivially different, Rabbi Yaaqov bar Zevidei’s version vs. Rabbi Mana’s:

1) אמר רבי יעקב בר זבדי קומי רבי אבהו לא כן אמר רבי זעירא ורבי יוחנן בשם רבי ינאי רבי ירמיה רבי יוחנן בשם ר”ש בן יוצדק נמנו בעליית בית נתזה בלוד על כל התורה מניין אם יאמר עכו”ם לישראל לעבור על אחת מכל מצות האמורות בתורה חוץ מעבודה זרה וגילוי עריות ושפיכת דמים יעבור ולא יהרג הדא דתימר בינו לבין עצמו אבל ברבים אפילו מצוה קלה לא ישמע לו כגון לולינוס ופפוס אחיו שנתנו להם מים בכלי זכוכית צבועה ולא קיבלו מהן אמר לא מתכוין משמדתכון ולא איתכווין אלא מיגבי ארנונין כמה הם רבים רבנין דקיסרין אמרי עשרה דכתיב (דף י:) (ויקרא כג) ונקדשתי בתוך בני ישראל>רבי אבונה זעירא חמנוניה פרי חורי חמרא בשבתא רבי יונה ורבי יוסי הורין מפי לארסקינס בשובתא
2) אמר רבי מנא קשיתי קומי רבי יונה אבא לא כן אמר רבי זעירא רבי יוחנן בשם רבי ינאי רבי ירמיה רבי יוחנן בשם ר”ש בן יוצדק נמנו בעליית בית נתזה בלוד על כל התורה כולה מניין אם יאמר עכו”ם לישראל לעבור על אחת מכל מצות האמורות בתורה חוץ מן העבודה זרה וגילוי עריות ושפיכת דמים יעבור ולא יהרג הדא דתימא בינו לבין עצמו אבל ברבים אפילו מצוה קלה לא ישמע לו כגון לוליינוס ופפוס אחיו שנתנו להם מים בכלי זכוכית צבועה ולא קיבלו מהן אמר לא אתכווין משמדתון ולא אתכווין אלא מיכל פיתא חמימה כמה הם רבים רבנן דקיסרין אמרין עשרה דכתיב (שם) ונקדשתי בתוך בני ישראל

This shows the significance given to precise citation. Or, this segment from Maaseros 5:3, Vilna 24b. Notice how not only is the mishnah under discussion re-quoted before Rabbi Avohu’s comment is repeated, but R’ Avohu’s introduction, that sometimes the idea he is about to give is said in the name of Rabbi Lazer and sometimes in the name of Rabbi Yossi bei Rabbi Chanina is also repeated just a couple of lines later — all to make the two word point at the end “והוא שהחמיץ” that when Rabbi Yehudah says that water that was run through wine dregs that weren’t tithed and didn’t even increase in volume is obligated in maaser anway (because it was changed by the maaser-requiring dregs), it’s only if the mixture fermented.

משנה המתמד ונתן מים במידה ומצא כדי מדתו פטור.  רבי יהודה מחייב.  …
גמרא א”ר אבהו זימנין אמר לה בשם רבי לעזר זימנין אמר לה בשם רבי יוסי בי רבי חנינא והוא שהחמיץ. תמן תנינן התמד עד שלא החמיץ אינו ניקח בכסף מעשר ופוסל את המקוה.  משהחמיץ ניקח בכסף מעשר ואינו פוסל את המקוה.  מתני’ דרבי יודה היא דתנינן תמן המתמד ונתן מים במידה ומצא כדי מידתו פטור רבי יודה מחייב.  אמר רבי אבהו זמנין אמר לה בשם רבי לעזר וזמנין אמר לה בשם רבי יוסי בי רבי חנינה והוא שהחמיץ.

The Ridvaz comments on this (Maaser Sheini 4:1, vilna pg 28a “תםן ללא יתיר פיריי…”:

פי׳ דהוא שייך למס׳ שביעית פ״ד דשם פריך מהברייתא ההיא לענין שביעית ומשני שם דתננן דלא יתיר פירוי כו׳ ואגב גררא דמביא הברייתא מביא גם כאן זאת כדרך הש״ס הזה  ככמה מקומות כידוע לדרגילין בו :

Meaning, that this is relevant to tractate Shevi’is ch. 4, because there it asksand it answers there… Because of “dragging”, that [once] it brought the beraisa it also brings here [this following discussion], as is the way in this talmud in numerous places, as is known to those who are used to it.

And then there is the opposite extreme — instead of repetitious verbiage in a desire to preserve the older discussion intact, the gemara will instead refer to opinions with the briefest of citations, leaving the commentaries scrambling to chase down what is being referred to. Sometimes, to wildly different conclusions. “A machloqes between what Rabbi Yehudah said on Sukkah and Rabbi Yosi said in Menachos.” Or, “since we see in the famous case of …” (without telling you the case), etc…

Similarly too, the Yerushalmi quotes the tannaim in the Tosefta regularly, far more often than the Bavli.

What one is clearly listening to is a group of rabbanim who follow very closely the exact wording of the statement. And knowing that they relied on studying that wording, they took caution to preserve it.

Things not said:

Therefore, the use of using reasoning to fill in gaps in our knowledge about the din under discussion occurs far less often. Many discussions end with an open question, rather than extrapolating from his words what a tanna probably would have said. Again, this isn’t only because the kind of reasoning Rabbi Yehudah developed was with him in Bavel. As we will see in another post, the Yerushalmi does engage in its own style of reasoning. (To give a teaser: there is more use of analogy to apply the parallel of a din to a different area of halakhah.)

Rather, I think this willingness to leave the question open is because they were very aware of the gap between the actual quote and a deduction. Also, tampering with the quote with suggestions contaminates the repetition process — they are actually detrimental to what the Yerushalmi is trying to do.

One amora who particularly suffered from this was Rabbi Yirmiyah. Every several pages or so, Rabbi Yirmiyah would propose some implausible case, something that would measure the limits of the just-quoted halakhah. In fact, he only arrives in Eretz Yisrael after being thrown out of the beis medrash in Bavel for doing this one time too many (Bava Basra 23b). For example, ֛Maaseros 3:4, 17b discusses how many fruit a picker may hold such that it still qualifies as a snack, as one may only snack in the field from food that hasn’t yet had terumah and maaser removed. R’ Yirmiyah asks about the case where the picker juggles the olives. Does the one in the air count as being “held”? In general, these questions don’t get answered.

Citation Culture:

Less directly connected, but I think part of this culture, is the Yerushalmi’s greater emphasis on finding sources in general.

For example, there is a halachic principle “ein sheliach lidvar aveirah — there are no messengers (or: proxies) for something that is a sin.” The Bavli’s explanation is logical, “the words of the Master, the words of the servant, which do you listen to?” Obviously, when given a choice between Hashem’s law and a person’s order, the Torah comes first. Therefore, someone who accepts such an appointment is culpable for making the wrong choice.

In the Yerushalmi (Terumos 35a), the source is a verse. “שליח לוקה והוא פטור דָּ֣ם יֵֽחָשֵׁ֞ב לָאִ֤ישׁ הַהוּ ולא לשולחיו – the messenger is punished and he [the sender] is not culpable. ‘[The murder] will be considered blood for this man’ — and not the one who sent him.”

We see this more often in the greater emphasis the Yerushalmi gives asmachtos, finding references or mnemonics for rabbinic law in the Torah. For example, by Torah law, there is no minimum for terumah. However, according to Beis Hillel, the Rabbis set that the miserly must give at least 1/60th to a kohein, the norm would be to give 1/50th, and the generous would give 1/40th. According to Beis Shammai, the rabbinic range for terumah was instead 1/50th, 1/40th and 1/30th, respectively.  The Yerushalmi (Terumos 4:3, vilna 19b-20a) textually supports all six measures with asmachtos.

כתיב ששית האיפה מחומר החיטים וששיתם את האיפה מחומר השעורים יכול תורם חיטים א’ משלשים ושעורים אחד מששים ת”ל וכל תרומת שיהיו כל התרומות שוות שמואל אמר תן ששית על ששיתם ונמצא תורם אחד מארבעים בינונית אחד מחמשים א”ר לוי כתיב וממחצית בני ישראל תקח אחד אחוז מן החמשים כל שאתה אוחז ממקום אחר הרי הוא כזה מה זה אחד מן החמשים אף מה שאתה אוחז ממקום אחר הרי הוא כזה והרעה אחד מששים דכתיב וששיתם את האיפה מחומר השעורים. בית שמאי אומרים משלשים וששית האיפה מחומר השעורים בינונית מארבעים מן הדא דשמואל והרעה מששים מן הדא דר’ לוי דאמר רבי לוי בר חינא כל המוציא מעשרותיו כתקנן אינו מפסיד כלום מה טעמא ועשירית החומר יהיה האיפה מן החומר

Last, the Bavli will only cite a mishnah if it provides a clear source as to the law under discussion. In what I think is another instance of “Citation Culture”, the Yerushalmi will quote a mishnah that hints at the law, even if it is not usable as an indisputable proof. On Terumos 6:1, vilna 31b, Rabbi Yosi quotes a mishnah as saying “that which grows from [planting] terumah is terumah“. If someone plants terumah wheat, the entire resulting crop must be given to kohanim as terumah. (This is only rabbinically. Therefore, the Torahitic obligation of tithing still applies to the crop, and must be given by the kohanim.) Rabbi Yosi uses this mishnah to show that it’s specifically when one plants terumah itself. If someone consumes terumah and then has to reimburse the kohanim, “that which goes reimbursement for terumah is not terumah” as the mishnah speaks of terumah specifically, not reimbursements.

(One might see this as reading more into a source than what is there, thus potentially corrupting the purity of the transmitted quote. In short, as defying the entire thesis of this post. I rather see it as an implication inherent in the text, but one that simply doesn’t reach the unimpeachable source level of certainty, and thus in line with trying to find a connection to tradition for every existing law.)

Coming attractions:

In the first installment I argued that the style of dialectic (shaqla vetarya) was honed by Rav Yehudah, the founder of the yeshiva at Pumbedisa. In this essay, I tried to show how rarely the Yerushalmi engages in this argument style, instead preferring a dependency on existing sources. Next, I hope to illustrate the modes of reasoning one does find in the Yerushalmi insead.

You may also like...

No Responses

  1. avakesh says:

    Thank you for a very interesting discussion. How does the time passage between teh two Talmuds and the much lesser editing of Yerushalmi than Bavli impact on these points. In other words, could have the Bavli of R. Yochanan’s time looked exactly like the Yerushalmi?

  2. micha says:

    I suggested in the previous post in this series that the Bavli’s style of learning dates back to Rav Yehudah’s founding of the Yeshiva in Pumbedisa, and noted his dialectic style in sugyos like Bava Metzia 38b, Sanhedrin 17b and Chullin 110b, as well as a comment along this line by R’ Sheishes. So it would seem that the difference in learning style was around contemporaneous with R’ Yochanan and Reish Laqish, slightly later, but not much. Back in Neharda’ah, R’ Sheshes was known for what I call here the Y-mi’s “Citation Culture” (see how Rami b”R Chama compliments this about him in Bekhoros 52b). I would suggest that Neharda’a was a has-been town by then, and R’ Sheshes’s style therefore fell by the wayside in Bavel. (Technically, Neharda’ah was a region, Pumbedia is a town in that area. I mean here, the place in Neharda’a where Shemu’el’s academy was was where R’ Sheshes was a dayan, but by then the center of learning in the area was Pumbedisa.)


  1. November 7, 2010 – ל׳ במרחשוון תשע״א

    […] the Yerushalmi’s style of learning differs, as seen from a look at its first 5-1/2 mesechtos. Part I, Part II – […]

Leave a Reply

Your email address will not be published. Required fields are marked *