Shiur #24: The Tale of Rabba bar Avuha and Eliyahu (Part II)

  • Rav Dr. Yonatan Feintuch


In the previous shiur, we discussed the encounter between Rabba bar Avuha and Eliyahu (Babylonian Talmud, Bava Metzia 114a-b). We dealt with the narrative itself and a literary analysis thereof. In this shiur, we will analyze the content of the narrative in light of its broader context in the sugya.

Generally speaking, we may say that this narrative is positioned on the border between Halakha and Aggada. In fact, the narrative itself may be split into two segments. The main content of the first segment of the narrative is halakhic, as the Amora Rabba bar Avuha asks the Prophet Eliyahu a number of halakhic questions and Eliyahu responds to them. This part of the narrative — the halakhic discussion between the characters — does have a literary framework with a certain plotline. This plot is, of course, incredibly minimalistic — an encounter in a cemetery — but based on the formal definitions of most literary researchers, this would still qualify as a story. Nevertheless, the essence of the content is halakhic.

As for the latter half of the narrative, it is of a more aggadic bent, telling the tale of Rabba bar Avuha’s visit to Paradise, also known as the Garden of Eden or the World to Come. Its message is aggadic as well, exploring whether it is worth it to withdraw from one’s account in Heaven for use in this world — even if one does so with prophetic authorization.

The transition between the two parts is effected by the third question posed by Rabba bar Avuha: his objection to Eliyahu’s presence in a cemetery. The question is wholly halakhic in context, but it advances the next part of the narrative. Even though this is not stated explicitly, the implication is that the cemetery is a waystation between this world and Heaven.

The integration of both parts of the narrative is a good example of the complexity of aggadic narratives that appear in halakhic sugyot in the Babylonian Talmud. The relationship between the halakhic discussion and the narrative is not always trivial, and sometimes it is difficult to define what exactly this relationship is. At the same time, even in such instances, we must attempt to draw on all of the connections between the narrative and the halakhic debate in order to give meaning to the appearance of the narrative in the sugya, beyond technical association. Nevertheless, we should attempt to understand these points without being forced into statements that cannot be anchored well in the text. Our aim in citing this narrative is to demonstrate the method for thoughtful consideration of a sugya of this type.

For the convenience of the readers, below is the narrative once again.

The Narrative

Rabba bar Avuha met Eliyahu standing in a non-Jewish cemetery.

He said to him:

1) “Is a means test to be applied in favor of a debtor?”

He replied: “We deduce [the law of] poverty [written here] from that of valuations. In respect to valuations, it is written (Vayikra 27:8), ‘But if he be poorer than your valuation […according to the means of him that vowed the priest shall value him],’ while of a debtor it is written (Vayikra 25:35), ‘And if your brother grow poorer […then you shall relieve him].’”

2) “From where do we know that a naked man must not separate [terumah]?”

“From the verse (Devarim 23:15), ‘That He shall see no matter of nakedness among you.’”

3) He said to him: “Are you not a priest? Why are you standing in a cemetery?” 

He replied: “Has the Master not studied the laws of purity? For it has been taught that R. Shimon bar Yochai says: Non-Jewish graves do not defile, as it is written (Yechezkel 34:31), ‘You are My sheep, the sheep of My pasture, you are adam’ — you are referred to as adam [to defile others in this manner], but non-Jews are not referred to as adam.”

He replied: “I cannot even adequately study the four; can I then study the six?”

He asked him: “And why?”

“I am too hard-pressed,” he answered him.

[Eliyahu] then led [Rabba bar Avuha] into Paradise and said to him: “Remove your robe and collect and take away some of these leaves.” So he gathered them and carried them off.

As he was coming out, he heard a remark, “Who would so eat his [portion in] the World [to Come] as Rabba bar Avuha has done?” So he scattered and threw them away.

Nevertheless, since he had carried them in his robe, it had absorbed their fragrance, and so he sold it for twelve thousand dinarim, which he distributed among his sons-in-law.

The Narrative in Its Broader Talmudic Context

At the beginning of the narrative, Rabba bar Avuha asks whether means-testing (literally siddur, arrangement) for debtors is appropriate. As we have noted previously, this question stands at the center of the discussion of the halakhic sugya that precedes the aggada, and it constitutes the essential connection between the narrative and the sugya.

There are a number of other connections between the story and the sugya, although they are more general. The sugya deals with a debtor who is in default and whose collateral is being taken away. The segment of the sugya that is adjacent to the narrative and that is directly tied to it thematically is the discussion of means-testing for loans, valuations (arakhin) and dedications to the Temple (hekdesh). In each of these cases, a person lacks the capacity to fulfill a monetary commitment. The question discussed in the sugya is what necessities such a pauper is entitled to keep.

When it comes to arakhin (a type of vow in which one obligates himself to give the valuation of a certain person to the Temple, based on a set table of age and gender), the consensus in the sugya is that means-testing is appropriate, as the Torah itself mandates such a procedure (Vayikra 27:8):

But if he be poorer than your valuation, he shall stand before the priest; according to the means of him that vowed the priest shall value him.

Thus, the priest has an essential role in the domain of arakhin.

In this narrative, Rabba bar Avuha is revealed to be a pauper, and he encounters Eliyahu, who (at least according to the story) is a priest, and Eliyahu tries to help Rabba out. As previously noted, this is the lone Talmudic narrative in which Eliyahu is presented unambiguously as a priest. This connection between the story and the sugya is not an obvious one, as the priest’s role in means-testing is not explicitly mentioned and exists solely in the background, as it appears in the passage about arakhin only (the above verse is never quoted in full).

Another connection is Rabba bar Avuha’s robe, which plays an important role in the second part of the story, as we saw in the previous shiur. In the halakhic discussion of collateral, the borrower’s garment occupies a pivotal place.[1] In one of the passages dealing with collateral, the Torah describes it as the minimal possession of the impoverished borrower:

If you take your neighbor’s cloak as a pledge, return it by sunset, because that cloak is the only covering your neighbor has. What else can they sleep in? When they cry out to me, I will hear, for I am compassionate. (Shemot 22:25-26)

The Torah uses a number of similar terms here: salma, kesut, simla. The term for Rabba bar Avuha’s robe or cloak is gelima, which in Talmudic parlance is the most basic garment. It is the shirt off one’s back, as it were; more precisely, it is “the gelima from on one’s shoulders.” In the context of repaying a debt, the term is frequently used.[2]

It is possible that there exists an additional, indirect link between the narrative about Rabba bar Avuha’s poverty and the halakhic discussion preceding the narrative, as the debate there is between two Amora’im from the Land of Israel, Ilfa and R. Yochanan. In another well-known Talmudic tradition (Babylonian Talmud, Ta’anit 21a), these Amora’im contend with serious economic privation as they study.

Structure of the Sugya and the Editorial Voice of the Gemara

The sugya in which the narrative is situated deals with taking collateral, and a secondary question that develops in it is the propriety of means-testing for debtors. This question is addressed and answered in the affirmative, initially in the baraita (113b), which compares the law of loans to the law of arakhin[3] due to the textual analogy of references to poverty. Subsequently, this rule is quoted in the presence of R. Nachman, which leads to an anonymous discussion of a matter that is not related to means-testing:

A teacher recited before R. Nachman: Just as assessment is made in the case of valuations, so is it also made for debtors.

He said to him: If we even sell [his property], shall we make an assessment for him?!

But do we really sell [his property]? Did we not learn…

The question, “But do we really sell [his property]?” is asked by the editorial voice of the Gemara, and it weakens R. Nachman’s challenge to siddur for a debtor, as derived from arakhin. In other words, this question militates for means-testing for loans. This means that the editorial voice of the Gemara minimizes the opposition to siddur, as derived from arakhin.

Afterwards, the question is raised anew as a scholarly query, and it features a debate from the sages of the Land of Israel, which is developed by the editorial voice of the Gemara, asking whether siddur can be applied to hekdesh based on an analogy to arakhin and personal loans:

The scholars queried: Is a means test to be applied in favor of a debtor? Do we deduce [the law of] poverty [written here] from that of valuations or not?

Come and hear, for Ravin sent word in his letter: I asked this thing of all my teachers, and they gave me no answer thereon. But in truth, the following problem was raised: If one vows a hundred [drachma] for the upkeep of the Temple, is a means test applied in favor of such a person?  R. Yaakov, on the authority of Bar Pada, and R. Yirmeya, on the authority of Ilfa, said: [It follows] a fortiori from an ordinary debtor. If no assessment is made even for a debtor, to whom it [the pledge] is returned, then in regard to hekdesh, where it [the pledge] is not returned, surely, there is no assessment!

But R. Yochanan ruled: It is written (Vayikra 27:2), “A vow by your valuation” [to equate the two]. Just as a means test is applied for valuations, so also for [a vow to] hekdesh.

And the other? That is to teach the judgment [of a limb] according to its importance. Just as in valuations [a limb] is judged according to its importance, so also for [a vow to] hekdesh.

But let there be an assessment for a debtor, a fortiori from valuations: If an assessment is made in the case of valuations, where it [the pledge] is not returned, then surely there should be an assessment for a debtor, where it [the pledge] is returned!

The verse writes (ibid. v. 8), “But if he be poorer than your estimation” — “he,” but not a debtor.

And the other? This teaches that he must remain in his poverty from beginning to end.

Now, in the case of [a vow to] hekdesh, let it [the pledge] be returned, a fortiori from a debtor: If it [the pledge] is returned to a debtor, for whom there is no means test, surely it is returned in the case of [a vow to] hekdesh, seeing that an assessment is made there!

The verse states (Devarim 24:13), “[You shall restore to him the pledge as the sun sets,] that he may sleep in his cloak and bless you [and it shall be righteousness for you before the Lord your God],” thus excluding hekdesh, which needs no blessing.

Does it not? But it is written (ibid. 8:10), “And you shall eat and be full, and you shall bless the Lord your God for the good land he has given you.”

Still, the verse says, “And it shall be righteousness for you before the Lord your God.” Hence, it [the law of returning] holds good only for him [the creditor] for whom the act of righteousness is necessary, thus excluding hekdesh [as a creditor], which does not require [the merit of] righteousness.

From the preceding discussion, it arises indirectly that according to one of the views from the Land of Israel, that of Ilfa, there is no siddur to be made for personal loans. However, the question itself is not explicitly answered. The editorial voice of the Gemara adds and tries to derive means-testing for debtors from arakhin, with an argument a fortiori from Ilfa, although this argument is rejected due to the limitation of “he,” while “the other” explains the “he” in a different way. The impression that remains is that the anonymous discussion in the Gemara undermines, to a certain extent, Ilfa’s view, according to which it is clear that there is no siddur for a debtor, as it tries to direct his view towards a derivation of loans from arakhin.

To summarize, both parts of the halakhic sugya raise opposing arguments about the question of means-testing for a debtor, and both parts conclude without coming to a clear verdict about the matter. Nevertheless, one gets the impression that the Gemara at least favors deriving means-testing for a debtor from arakhin, even if it does not explicitly say so in the view of the Amora’im.[4] After the two parts, our narrative is cited, which opens with Rabba bar Avuha posing the question of the sugya to Eliyahu: “Is a means test to be applied in favor of a debtor?” The response is in the affirmative, and Eliyahu justifies this by way of the derivation from arakhin.

How the Narrative Arrived at Its Current Composition

Our analysis of the narrative in the previous shiur raised the idea that it is difficult to find a link between the first two questions in the narrative and the rest of it, in terms of content. This difficulty gives us good reason to analyze how the narrative arrived at its current composition, with all of its constituent parts. This is no easy task, as this tale has no parallels in the literature of Chazal, making it difficult to recreate the sources from which it was drawn and how it came to be. Nevertheless, we will raise some hypotheses aimed at explaining the narrative’s inception and the integration of these first questions within it, even though there is no way to determine which theory is correct.

As we have noted, the first question appears in the preceding halakhic discussion: “Is a means test to be applied in favor of a debtor?” Moreover, both the formulation that Rabba bar Avuha employs and Eliyahu’s response recall the Gemara’s own phrasing:

The scholars queried: Is a means test to be applied in favor of a debtor? Do we deduce [the law of] poverty [written here] from that of valuations or not? (114a)

The phenomenon in which a question asked anonymously by the Gemara is voiced by a character in a story, while the two sides presented by the Gemara are expressed as well, is not unique to our narrative.[5] Still, as we have noted, our narrative has no full parallels in other sources, which means that we can at most say that it is highly reasonable that this formulation of the question is the reason that this question is integrated into the sugya.

In any case, we may present two possibilities for how the narrative came to be. One possibility is that this tale was in existence, in one form or another, long before it was inserted into this sugya, independent of the question of siddur for a debtor (which was added later). Adding this scholarly query allows the narrative to fill a role in the sugya, as we shall explain below.[6] Alternatively, it may be that the question of means-testing for debtors existed in the narrative for some reason before it was integrated into the sugya, and for this reason it was introduced here.[7] However, the precise formulation of the question, which is identical to the formulation and development of the query in the Gemara’s editorial voice, would still have been influenced by the sugya here.[8]

As we have previously stated, it appears that the narrative fulfills a role in the sugya. In this narrative, Eliyahu unequivocally answers the central question of the sugya, “Is a means test to be applied in favor of a debtor?” His affirmative response buttresses the view that siddur for personal loans is appropriate, as derived by textual analogy from arakhin. This view, as we have shown above, is a position to which the editorial voice of the Gemara inclines, but the matter is not conclusively settled in the sugya itself.

In terms of the connection between the aggada and the sugya, in the narrative we are exposed to an additional role of the aggadic tale in a sugya: strengthening the halakhic position of one view discussed in the halakhic sugya, even though the debate does not reach a definitive conclusion.

We cannot come to a definitive conclusion concerning if and to what extent this narrative has halakhic weight. This is a complex and convoluted question requiring a separated discussion, as it depends on the various views expressed by the Rishonim and the halakhic decisors concerning aggada. Our sugya does not explicitly address the halakhic weight of rulings expressed by the Prophet Eliyahu, and it is difficult, even in this specific case, to reach a definitive conclusion about the specific status of narrative within the entirety of the halakhic debate in the sugya, as viewed by the editors of the Gemara.[9] However, it appears that the very fact that the narrative is cited indicates something about the view that the editors of the sugya seek to support, just as we may maintain concerning the anonymous discussions in the Gemara that appear in the halakhic segment.


Translated by Yoseif Bloch


[1] This is discussed, in a general manner, in the Jerusalem Talmud as well; see Pe’ah 8:7-8, 21a-b.

[2] See, for example, Bava Kama 11b; Bava Batra 44b, 157a.

[3] The law of siddur in the context of arakhin is specified in Mishna Arakhin 6:3; cf. Gemara ad loc. 23b.

[4] Cf. Nedarim 65b; the gemara there tries to derive from the mishna there something about the question of siddur for a debtor, which indicates that the question has not been resolved.

[5] See, for example, the narrative of the officials, Bava Kama 38a.

[6] A similar claim may be raised concerning the second question, about a naked man separating teruma, which also has no connection to the rest of the story, as we noted. However, this question cannot be explained in any way due to the adjacent sugya. According to this option, it may be that we can explain that this question (which does not appear to be a new creation of the authors of this story, as we have seen that it exists in the sugya in Terumot in the Jerusalem Talmud) is interwoven in the narrative due to one of the reasons mentioned in our literary analysis: for structural/ literary reasons, either in order to create a triad or in order to illustrate the ignorance of Rabba bar Avuha in three different Orders of the Mishna. However, this is at most speculation.

Another common denominator between these three questions that may explain their being gathered into one narrative is that these are the only instances in which a scholar asks Eliyahu halakhic questions and the prophet responds by citing biblical sources. One additional tradition in which Eliyahu makes three halakhic statements is the narrative in the Babylonian Talmud, Berakhot 3a, about R. Yosei in a ruin. However, as opposed to the story in Bava Metzia, the scholar in that case does not ask Eliyahu questions, but rather derives three laws from a dialogue between him and Eliyahu (without any biblical citations).

In all of the other narratives in the Babylonian Talmud, scholars consult Eliyahu about aggadic and philosophical matters. Some are Tanna’im (R. Yosei, one question, Yevamot 63a; R. Natan, one question, Bava Metzia 59b), some are Amora’im (R. Beroka Choza’a, one question, Ta’anit 22a; R. bar Avuha, one question, Megilla 15b; R. bar Shila, three questions that lead from one to the next, Chagiga 15b; R. Evyatar, three interconnected questions, Gittin 6b; R. Yehoshua ben Levi, series of [more than three] questions, Sanhedrin 98a).

There are additional descriptions of encounters between Eliyahu and various scholars in which they speak to each other, without the format of a question-and-answer session, e.g. Yoma 19b, Ta’anit 24b, Chagiga 9b, Ketubot 106a, Gittin 70a, Kiddushin 40a, Bava Metzia 84a and 85b, Sanhedrin 97b. Compare this to Bereishit Rabba 35:2 (ed. Theodor-Albeck, pp. 328-329), in which Eliyahu and R. Yehoshua ben Levi are studying together and have a question about the view of (the long-dead) R. Shimon bar Yochai, so they seek him out to consult him.  

For further analysis of these encounters between Eliyahu and various scholars, see Yona Fraenkel, “R. Yehoshua ben Levi,” in his book Sippur Ha-Aggada: Achdut shel Tokhen Ve-Tzura, pp. 273-294.

[7] For example, the partial reasons that were raised as hypotheses in the literary analysis; see the previous footnote.

[8] There is, of course, a third possibility – that the narrative as a whole was created and developed as part of the development of the sugya. This possibility, which indeed cannot be dismissed absolutely, still seems less reasonable, as the lion’s share of the narrative and its central themes are not at all tied to the topic of the sugya, and it is unclear what the authors of the sugya would gain from crafting such a tale.

[9] We should note that although various Talmudic statements indicate that a prophet cannot decide halakhic questions, there are a number of traditions in which Eliyahu is the source for halakhic information – e.g. the partial parallel to our narrative in the Jerusalem Talmud Terumot, as well as the story of R. Yosei in the ruin in the Babylonian Talmud Berakhot 3a. This matter requires a separate and wide-ranging analysis. The same is true of the question of how much weight to give to the narrative in the sugya of means-testing for debtors, as discussed by various commentators and halakhic decisors. In brief, we will note that some of the Rishonim who address our sugya cite Eliyahu’s words as a halakhic source in the full sense of the term, in keeping with their general views about aggadic material in the Talmuds. Thus, for example, Rabbeinu Tam (as cited in the previous shiur) emends the text to read, “From where do we know that a means test is to be applied in favor of a debtor?” — motivated, inter alia, by halakhic reasons, as we saw there. It appears that the motivation to change the formulation shows that Rabbeinu Tam treats aggadic material no less seriously than halakhic material. In contrast, Ramban (Chiddushim ad loc. s.v. ve-ha) challenges Rabbeinu Tam, arguing that altering the text would not change the ruling; nevertheless, he does not in any way question the halakhic value of what is conveyed in the narrative. An exhaustive and comprehensive analysis of these questions must take place in another forum, as part of the general discussion of the application of aggadot in the Talmud for halakhic rulings, a matter of some dispute among Geonim, Rishonim, and Acharonim.