Devarim (5765)

From the book Mida Tova: Articles on the Hermeneutical Principles by Rabbi Michael Avraham. Translated from Hebrew using gpt-5.4 (reasoning_effort=high, batch API).

---

With God’s help

Midah Tovah — Sabbath eve of Parashat Devarim, 5766

Questions:

1. What is the difference between halakha (Jewish law) and aggadah (non-legal rabbinic teaching)?
2. Why are there hermeneutical principles that are used only in halakha?
3. Are there hermeneutical principles that are used only in aggadah?
4. What is problematic about the study of aggadah?
5. The dispute between the sages of France and Spain regarding the halakhic implications of aggadah.
6. Is there room for halakhic ruling with respect to aggadic statements?

The principles:

Notarikon (acrostic interpretation). Gematria (numerical interpretation).

“And the Lord said to me: Do not harass Moab and do not provoke them to war, for I will not give you any of their land as a possession, for I have given Ar to the descendants of Lot as a possession.”
(Deuteronomy 2:9)

“… He said to him: ‘And the Lord said to me: “Do not harass Moab and do not provoke them to war.”‘ Did it occur to Moses to wage war without authorization? Rather, Moses drew a qal va-homer (an a fortiori inference) for himself. He said: If with regard to Midian, who came only to assist Moab, the Torah said, “Harass the Midianites and strike them,” then Moab themselves all the more so. The Holy One, blessed be He, said to him: Not what arose in your mind arose in Mine. I have two precious descendants to bring forth from them: Ruth the Moabite and Naamah the Ammonite.”
(Babylonian Talmud, Bava Kamma 38a)

“Another interpretation: ‘Take one vengeance against Midian and another against Moab.’ But was not Moab the first to devise evil counsel against Israel, as it is said, ‘Moab said to the elders of Midian’? So why were Midian treated differently from Moab, so that regarding Midian it says, ‘Avenge the children of Israel on the Midianites,’ whereas regarding Moab it says, ‘Do not harass Moab’? Rather, because David was destined to come forth from them. The Holy One, blessed be He, said to them: Wait with them. I have a treasure among them; I will take it from them, and afterward take vengeance upon them. And so we find that once David came, the Holy One, blessed be He, delivered them into his hand, as it is said, ‘He struck Moab and measured them with a line, making them lie down on the ground.’ But Midian were delivered into the hand of Israel, as it is said, ‘They arrayed themselves against Midian.’ And Moab’s counsel against Israel was only to kill them, as it is said, ‘Perhaps I will be able to strike them.’ But Midian’s counsel was only to cause Israel to sin; yet through them twenty-four thousand of Israel fell, and through them Israel became liable to destruction, for one who causes another to sin is worse than one who kills him. Moreover, the Holy One, blessed be He, wrote concerning them, ‘An Ammonite or Moabite shall not enter…’ Why are the males forbidden and the females permitted? Because Ruth was destined to come forth from them.”
(Yalkut Shimoni, Matot, sec. 785)

A. Summary of Last Year’s Article

In the Talmudic passage above, a qal va-homer is presented that could have served as the basis for Moses’ decision to go out to war against Moab. The comparison is between Midian, concerning whom the Torah commanded that they be harassed and struck, and Moab. The basis for the comparison is that Midian came only to assist Moab, and therefore war against Moab is clearly more justified than war against Midian.

We presented several perspectives from which specifically the war against Midian appears more justified, and some of them are rooted in rabbinic sources: Moab feared the Israelites, whereas Midian embroiled themselves in a quarrel not their own. The Holy One, blessed be He, wished to preserve Moab because Naamah the Ammonite and Ruth the Moabite were destined to emerge from them. Beyond that, Moab recruited a prophet to curse Israel, whereas Midian caused them to sin directly and actively. Moab wanted to kill Israel, while Midian wanted to make them sin; and, as is well known, one who causes another to sin is worse than one who kills him.

We saw two directions among the Sages, with roots already in the Torah itself, regarding which of the two was more severe. Those who hold that Midian’s sin was less severe do so because Moab were the initiators. Midian were merely the agents, and had they not done it, someone else would have.

We then raised the difficulty of how either side can ignore the opposing considerations. Even if one thinks that initiation is the more severe element, and that the fact that the Midianites involved themselves in another’s quarrel, caused Israel to sin, and the like is less important, still, to construct a valid qal va-homer one needs an unambiguous hierarchy. Any respect in which Midian turn out to be more severe constitutes a refutation of the inference.

We mentioned the mechanism, discussed by the medieval authorities, of neutralizing a refutation, but it is not relevant in our case. We then showed that in certain situations one may construct a qal va-homer even when there are two opposing stringencies that point in opposite directions.

The Talmud contains a clear case of this: the qal va-homer from an unpaid bailee to a paid bailee with respect to misappropriation of a deposit; see Babylonian Talmud, Bava Metzia 41b and parallels. The paid bailee is more severe, since he is liable for theft and loss. But the unpaid bailee is more severe in another respect, since he becomes liable for double payment when he falsely claims that the item was stolen. Even so, the Talmud says that liability for the principal amount without an oath is weightier than double payment that arises only with an oath. Therefore the paid bailee is considered more severe, and one may construct a qal va-homer from the unpaid bailee to him.

We mentioned that this problem exists in every disputed qal va-homer. Whenever one side raises a consideration that refutes the inference, the side making the inference must explain why that refutation does not destroy it. If the refutation is halakhic, one can say that one side does not accept the law on which it rests. But when the refutation is conceptual, that is much harder to say.¹

We noted that the Bava Metzia passage is exceptional, because the two stringencies are related to one another, and therefore they can cancel each other out. This is not the case with opposing refutations that lie on different axes. On that basis we also explained our midrash. There too, the refutations were related to one another, and therefore they can be offset. We even drew a halakhic implication from this, namely, that once King David was born, the command not to provoke Moab lapsed. We cited several opinions among the halakhic decisors on this matter.²

In the second half of the article we dealt with the question whether a person is judged on the basis of his future or according to his present state. We brought several examples, and our conclusion was that the Holy One, blessed be He, judges no one on the basis of his future. Judgment is based on the present state, and the forecast of the future serves only as an indication of the present. That is how we explained the consideration regarding Ammon and Moab as well. The presentation of the consideration on the basis of the future, namely Ruth and Naamah, is only an indication of a positive trait in the present, because of which they should be left alive.

At the end of our discussion we dealt with God’s knowledge of the future, and we saw that regarding events of human choice, the Holy One, blessed be He, relinquished foreknowledge of the future. In exceptional cases He takes the reins back into His own hands and dictates the future outcome. The author of Leshem calls this mode of providence “the governance of the Fearsome Plotter.”

B. On Halakha and Aggadah³

Introduction

We saw in last year’s article that the aggadic midrash regarding Ammon and Moab ends with halakhic conclusions. Can halakhic conclusions be derived from aggadic midrash (rabbinic interpretation)? And if so, what exactly is the difference between these two kinds of midrash? We saw a similar phenomenon in the midrash discussed in our article on Parashat Vezot HaBerakhah, 5765. In this year’s article we will try to broaden the discussion somewhat, and consider the relationship between halakha and aggadah.

Halakha and Aggadah

It is commonly assumed that halakha is a more precise and rigid domain, whereas aggadah is more flexible, less precise, and less unequivocal. Halakha certainly also depends on the reasoning of the halakhic scholar, but the dependence of aggadah on personal reasoning appears much stronger.⁴ For that reason, it is not customary to derive halakhic conclusions from aggadot.

These differences may be considered on two planes: the practical and the theoretical.

The Practical Plane

Practically speaking, students of Torah are less exacting when they deal with aggadah. All of us know this from the reality around us. There are several illuminating examples of this in the written sources, and one of the most striking is found in the introduction of Penei Yehoshua to his monumental commentary on the Talmud.

The author of Penei Yehoshua, who at the time was still young, shortly after his first marriage, tells there of an explosion that took place in his city. A fire broke out as a result, his house collapsed, and several members of his family died beneath it. While he himself was trapped under the rubble, the rescuers, as they approached, also posed a danger to those buried below. At that point Penei Yehoshua vowed that he would no longer engage in aggadot, except in rare cases where he felt that he had reached the true meaning of Torah, but only in halakhot and in the give-and-take surrounding the interpretations of Rashi and Tosafot. Because of the uniqueness of the passage and the force of its language, we shall cite parts of it in his own words:

“One reason is that I accepted this upon myself as an obligation and vowed a binding vow in my distress, on the day of the fierce anger of the Lord, the third of Kislev in the year 5463, in the holy community of Lvov. I was tranquil in my home and flourishing in my halls, with companions and students listening to my voice, and suddenly, in an instant, the city was overturned into a ruin. No warning hands were raised and no cry of terror was heard; only the sound that the conflagration had broken out in one quarter, and the sight of the great blazing fire rising in our palaces and our windows because of several huge and terrible barrels filled with powder for burning, until houses were destroyed from their foundations, several great houses and fortified walls, lofty as heaven, were brought low to the dust, laid bare to their very foundations, and some thirty-six holy Jewish souls were killed. Among the dead too were members of my household: my first wife, of blessed memory, her mother, and her mother’s father. And the distress also reached my daughter, my little daughter, her mother’s only child, who was most dear to me.”

“And I too was among the fallen, cast from a high roof into a deep pit, and I came into the depths of the mire, into the lowest earth, truly as if inside a press, because of the crushing weight of wave after wave that fell upon me, and the beams of our house, heavier than the beams of an oil press, did not let me regain my breath. My hands and limbs were not in my control. I said: I am cut off in the midst of my days; I shall go, deprived of the rest of my years. I shall no longer behold any man among the inhabitants of the world. And I feared lest my house become my grave, like those stoned, burned, slain, and strangled; all four were upon me at once, and the law of the four death penalties had not departed from me…”

“In the end, I had already passed from certain mortal danger into mere uncertainty. Then, while I was still within the heap of rubble, I said: If God will be with me and bring me out of this place in peace, and build me a faithful house and enlarge my bounds with students, I will not withhold myself from the walls of the study hall, and I will diligently attend to the depths of the discussions of the Talmud and the decisors, and spend nights in the depths of halakha, even many nights on a single matter…”

“Before I had finished speaking these words to my heart, the Lord heard the voice of my affliction and made for me a path between the pillars, an actual passage, and I went out in peace, unharmed, without a bruise upon me. Then I knew with certainty that this was truly from the Lord, in a place where there is no rescue. From that time on I set these words upon my heart, that the main part of my study should be halakha in the discussions of the Talmud and the decisors, and that I should write nothing on matters of homiletics or other studies that are far from the center of true learning, except on rare occasions. Only whenever some new insight occurred to me in a Talmudic discussion, or in the interpretation of Rashi and Tosafot, and it seemed to me to accord with the true path of learning according to the way of our predecessors and teachers, that I would choose and bring near to writing in this book of memorials.”

In other words, Penei Yehoshua regarded the study of homiletic material as something far from the center of true learning. And indeed, anyone familiar with his book knows that Penei Yehoshua proceeds page by page through the Talmud, whereas among aggadic discussions one hardly finds his novel interpretations at all.

A similar case occurred to Rabbi Shlomo Cohen, one of the sages of Lithuania at the beginning of the twentieth century, and he too vowed not to engage in aggadot, in which people are not exacting about the true meaning of Torah.

The Theoretical Plane

It seems that the fact that work in aggadah is less precise is not merely a result of laziness or lack of attention. The very nature of this field dictates that it is less precise and less unequivocal than halakha.

We ourselves presented an example of this in our articles on Parashat Balak, 5765-5766, and on Parashat Vezot HaBerakhah, 5765, where we dealt with the principles of notarikon and gematria. In the article on Parashat Vezot HaBerakhah, 5765, we noted that there are hermeneutical principles used in aggadic interpretation that are not used in halakhic interpretation, such as notarikon and certain types of gematria. But we did not find the reverse phenomenon: principles used exclusively in halakhic interpretation and not in aggadah.

From this phenomenon one may infer that the standards demanded of halakhic midrashim are higher than those demanded of aggadic midrashim. In other words, the difference between halakhic interpretation and aggadic interpretation is that in aggadic interpretation we permit greater amorphousness, or less unequivocality. For these reasons, methods such as gematria and notarikon, which are very far from precise and determinate, and lie on the border of mere allusion, apply only in aggadic interpretation.

This conclusion too may be considered on two planes:

1. In aggadic interpretation, precision is less important to us, since there is no danger of leading to sin, as there is with errors in halakhic interpretation.
2. In aggadic interpretation, it is not possible to be that precise, in the way that it is in halakhic interpretation.

These two planes correspond to the two planes of discussion presented above. The reason that learners are less exacting in aggadah, beyond the special and flexible nature of the field itself, is that the consequences of error in it are less severe.

A Qualification

As we concluded in the article on Parashat Vezot HaBerakhah, 5765, this distinction does not mean that the field of aggadic interpretation is completely lawless. There are indeed methods of aggadic interpretation; they are simply less precise than the halakhic ones. Some of them are listed in the baraita of the thirty-two hermeneutical principles of Rabbi Eliezer son of Rabbi Yose the Galilean.

The Mishnah in Avot 3:18, and likewise Avot de-Rabbi Natan 27, says:

“Rabbi Elazar ben Chisma says: The laws of bird-offerings and the gateways of menstrual impurity are the very body of halakhot; astronomical cycles and gematriot are desserts for wisdom.”

On the one hand, this statement hints at the vagueness of gematria. On the other hand, gematria is regarded as a dessert of wisdom. Thus we are not dealing with total amorphousness. There are certainly rules for the proper use of gematria; see examples in our article on Parashat Vezot HaBerakhah, 5765. Therefore it counts as a form of wisdom.⁵

We also saw a similar qualification in the words of Penei Yehoshua above. On the one hand, he says that in homiletic and aggadic writings one does not aim at the truth. On the other hand, where he feels that he has in fact arrived at the truth, he does not refrain from writing. There is truth in aggadah as well; it is just that the learners do not always succeed in aiming at it.

“One Does Not Learn from Aggadot”

In light of our discussion, we can understand the rule, “One does not learn from aggadot,” found in Jerusalem Talmud, Peah 2:4. In Encyclopedia Talmudit, s.v. “Aggadah,” at the beginning of chapter 4, several Geonic explanations are cited:

“Words of aggadah are not like words of received tradition and halakhic teaching. Rather, each expounds what arose in his heart as something possible,⁶ and they are matters of estimation.⁷ Therefore one does not rely on them, and the rule is: one does not rely on aggadah.⁸ And they said: ‘One does not learn from aggadot.’ From them you should learn only what commends itself to reason.⁹ One does not bring proof from aggadah,¹⁰ nor raise objections from aggadah.¹¹“

Because of the lack of precision found in the domain of aggadah, both in practice and in theory, it is reasonable that we cannot derive practical halakha from it. Some attribute this not to the imprecision of the field itself, but to the ambiguity of its language, which often uses parables and riddles. More on this below.

There are, however, qualifications to this rule as well. For example, where aggadot and midrashim do not contradict our Talmud but merely add to it, some say that one may indeed learn from them and rely on them.¹² Others have written that one does not learn from aggadot even if they do not contradict what is said in the Talmud.¹³

How are we to understand this dispute? Presumably, those who hold the first position maintain that aggadah is a precise domain, though one with its own rules, and therefore in principle one can learn from it and rely on it. This approach, associated with Rabbenu Tam, is indeed characteristic of the Tosafists. They tend to treat aggadah as a precise field, to raise objections to it from parallel passages, and to analyze it with tools similar to those used in halakha.¹⁴ Some have connected this issue with the dispute between the sages of Spain and Andalusia, who tended to regard aggadah as a more problematic domain, and the sages of France and Ashkenaz, whose stance finds expression in the approach of the Tosafists.¹⁵

Even according to the Tosafists, however, where there is a halakhic conflict between what emerges from aggadic passages and what emerges from halakhic passages, the halakhic passage has the advantage, since halakha is after all its principal subject. Here the preference does not stem from distrust of aggadic midrashim, but from a natural and clear hierarchy between halakhic and aggadic discussions. Therefore, according to this approach, where there is no contradiction to a halakhic passage, one does in fact rely on and learn from aggadot.

By contrast, those who hold the opposing approach maintain that aggadot are indeed imprecise, whether because of subjectivity or because of ambiguity of formulation, and therefore even when there is no contradiction between them and what emerges from the Talmud and from halakhic discussions, one should not derive halakha from them.¹⁶

This dispute also connects to our earlier distinction between two planes of imprecision. The Tosafists would locate the amorphous nature of aggadah only on the practical plane. In their view, learners do not always exercise sufficient care in aggadic passages, but in principle aggadah itself has no problem of precision. One might have attributed the problem instead to the ambiguity of aggadic language, but it seems that the Tosafists treat aggadic midrash just as they treat halakhic midrash, and therefore it appears that in their view the problem lies only with the learner. Those who hold the opposing position, by contrast, probably think that aggadah contains an inherent imprecision.

Halakha in Aggadah

Maimonides, in three places in his Commentary on the Mishnah, repeats the principle that one does not issue halakhic rulings in aggadic matters that have no practical consequences. This is his formulation in the Commentary on the Mishnah to Sanhedrin 10:3, and likewise to Sotah 3:3 and Shevuot 1:4:

“We have already mentioned to you several times that whenever there is a dispute among the Sages that does not depend on practical action, but only on the establishment of an opinion, there is no place to rule the halakha like one of them.”

Several other medieval authorities and Geonim wrote similarly.¹⁷

Why indeed should one not rule in matters of thought, worldview, or morality? It seems that in aggadah each person may expound according to his own understanding. At first sight, one would have expected the explanation to be that the field is not precise and determinate enough to permit fixed rulings. But we did not find that explanation in the commentators. The common explanation is that the matter does not involve prohibition or permission. In his comment on Shevuot there, Maimonides adds that the matter is entrusted to God alone, though his precise intention is not entirely clear to us.

It would seem that these explanations mean the following: one cannot compel a person to think something that he does not believe. In the practical sphere, a person can act contrary to his own understanding and outlook, and submit himself to the established halakha. But in the sphere of thought, if someone thinks that the Holy One, blessed be He, has bodily form, and we issue a ruling that He has no bodily form, that ruling is of no use whatsoever. That person still thinks he is right. Since the ruling concerns the act of thought rather than a practical act, his own conclusion necessarily remains operative within his thought.

It may be for this very reason that such matters were not classified in terms of prohibition and permission. If a person thinks incorrectly, then at most he has erred in his reflection, but he has not violated a prohibition, at least within certain limits, and this is not the place to elaborate.

What Is “Aggadah”?

It is important to note that the very term “aggadah” is not always clear. Usually it denotes everything that is not halakhic discussion. But in the contexts cited above, terms such as “haggadah,” “aggadah,” or “homily” sometimes refer to material presented before the public, in the sense of what is called in yeshivot today a “talk,” as distinct from a formal lesson.¹⁸ Such material is meant to attract the heart of the audience; see Babylonian Talmud, Shabbat 87a; Mekhilta, Beshalach, parasha 5; Babylonian Talmud, Yoma 75a; Babylonian Talmud, Chagigah 14a; and likewise Me’or Einayim, Imrei Binah, chapter 15. Therefore its purpose is not always to be precise and to hit the truth exactly.

Some commentators in fact distinguish between aggadot that appear in the Talmud and midrashim outside the Talmud,¹⁹ or other homiletic works.²⁰ Others distinguish between midrashim that are anchored in exposition of scriptural verses, that is, aggadic midrash,²¹ and aggadic discussions in general.²² In every context, we must examine carefully the meaning of the terms used in the discussion.

Nahmanides’ Words in the Disputation²³

One of the most famous debates between a Jew and a Christian was the disputation between Nahmanides and the apostate Pablo Christiani. In the course of the debate, that apostate tried to prove from the Talmud and the midrash that the Messiah had already come, that is, that “that man” was the Messiah. In the course of the disputation, Nahmanides made very sharp statements about aggadah, and stated that he does not necessarily believe it: either because it contains a hidden secret,²⁴ or because at times it is simply not correct.²⁵ The next day Nahmanides wrote words of explanation, and this is his language:

“Know that we have three kinds of books. The first is the Bible, and all of us believe in it with complete faith. The second is called the Talmud, and it is an explanation of the commandments of the Torah… and we believe in it with regard to the interpretation of the commandments. We also have a third book, called Midrash, meaning sermons. It is as though a bishop were to stand up and deliver a sermon, and one of the listeners liked it and wrote it down. As for this book, one who believes in it does well, and one who does not believe in it suffers no harm…”

At first sight, Nahmanides seems to be saying that he does not believe the books of Midrash. Yet it appears that he ascribes the problem to the transmitters, that is, to “one of the listeners,” and not to the problematic character of the midrashim themselves. Beyond that, he distinguishes between kinds of books rather than between fields of study. The distinction is between the Talmud and books of Midrash copied down by listeners. We already saw such a distinction above in the commentators.

However, if we look closely at his words, he is careful to say that even the Talmud is believed by us with respect to the interpretation of the commandments. This implies that in the aggadic parts we do not have full confidence even in what is said in the Talmud itself.

The commentators disagree as to the meaning of these words.²⁶ Some explain them as defensive remarks forced upon Nahmanides by the circumstances of the disputation, while his true opinion was different. Others argue that Nahmanides truly stands somewhere between the French and the Spanish schools, and that at least regarding some midrashim he does not accept them. It seems that the truth lies somewhere in the middle. There is indeed a certain ambiguity and flexibility in midrash, or at least in some midrashim; yet in a fundamental sense it is also clear that midrash is part of the Oral Torah.

Summary: Halakhic Conclusions from Aggadic Midrashim

We have thus learned that the question whether halakhic conclusions may be derived from aggadic midrashim is a matter of dispute. In a good number of cases, medieval and later authorities do derive halakhic conclusions from aggadic midrashim. The Tosafists do so quite frequently, and Nahmanides does so from time to time as well. The sages of Spain, following the Babylonian Geonim, do so much less often.

The problematic nature of this practice lies on several levels:

1. The midrashim are often formulated ambiguously, by way of parable and riddle.
2. The domain of aggadah has an inherent ambiguity and flexibility.
3. There is a lack of practical exactness on the part of those who study this field, mainly because it is not as consequential, since error in this field does not lead to transgression.

As for the definition of aggadic midrash and halakhic midrash, with which we opened this chapter, the common denominator of all views seems to be that all such passages are certainly aggadic midrashim. A midrash that does not deal directly with halakhot, even if some people derive various halakhot from it, is an aggadic midrash. Only afterward does a dispute sometimes arise as to whether halakhic conclusions may be drawn from it. The definition of the nature of the midrash is based solely on its explicit content, not on the conclusions that may be inferred from it.

In light of our discussion, it is clear that the midrash we dealt with in last year’s article is an aggadic midrash. The same holds for the midrash discussed in our article on Parashat Vezot HaBerakhah, 5765; see there at the end of the article. The halakhic conclusions drawn from it were conclusions that we ourselves derived.

What emerges from our discussion is that it is not clear whether all the medieval authorities would agree to such a move.

It should, however, be noted that where what is at issue is a central principle in the midrash, and not a fine detail that appears within it, many commentators would agree that one may derive halakha from that midrash. This is especially so when the midrash is a case of “speaking innocently,” that is, when the halakhic principle is learned only implicitly, as a self-evident assumption, and not when it is stated explicitly. Ironically, specifically when aggadic midrash explicitly expresses a halakhic position, its weight seems to be lower.

One further point should be added. When some midrash appears difficult in light of established halakha, it is important to reconcile the two, and perhaps a halakhic conclusion may even be drawn from that reconciliation; see the article of Rabbi Zalman Nehemiah Goldberg, where he elaborates on this at the end of chapter 4.

All of this applies, at least, to rabbinic midrash found in the Talmuds. As for midrashim that appear in other sources, beginning with the classical books of Midrash and extending to more external sources, deriving halakhic conclusions from them becomes increasingly problematic. In the final analysis, deriving halakhic conclusions from aggadic midrashim is problematic, but it should not be rejected categorically. As for our own conclusions from the midrash, the matter remains for the judgment of the reader.

Footnotes