Mishpatim (5764)

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 — Eve of the holy Sabbath, Parashat Mishpatim 5765

Questions

1. What is a warning?
2. What is the difference between Kelsen and the author of Sefer HaChinukh: does the Torah contain rules that direct behavior as well, or only rules of adjudication?
3. What is the relation between warning and punishment?
4. Why does one not derive warnings or punishments from inference?
5. What happens when the punishment is stated explicitly in Scripture but no warning is given?
6. What is the relation between cursing one’s father and mother, cursing a judge, a prince, a deaf person, and—on an entirely different plane—God?
7. Four logical models for a complex “common denominator,” in the context of warning and more generally.

The Hermeneutical Principle

binyan av (derivation from a paradigm case). The common denominator.

A diagram is attached by fax.

A. One Does Not Derive a Warning from Inference

“One who curses his father or his mother shall surely be put to death.”
— Exodus 21:17

“‘Shall surely be put to death’—by stoning… We have heard the punishment; from where is the warning? Scripture says: ‘You shall not curse a judge.’ If your father is a judge, then he is included in ‘You shall not curse a judge’; and if he is a prince, then he is included in ‘You shall not curse a ruler of your people.’ If he is neither a judge nor a prince, but merely an ordinary person, then you may derive a binyan av from the two of them: a judge is not like a prince, and a prince is not like a judge; their common denominator is that they are among your people, and you are warned against cursing them; so too your father, who is among your people, you are warned against cursing him. But perhaps their common denominator is that they are prominent among your people, and it is their prominence that gives rise to the warning against cursing them; would you say the same of your father? Therefore Scripture says: ‘You shall not curse the deaf.’ Scripture speaks of the most downtrodden among people. Thus you derive a binyan av from all three. A judge is not like a prince, and a prince is not like a judge, and these two are not like a deaf person, and a deaf person is not like these two; their common denominator is that they are among your people, and you are warned against cursing them; so too your father, who is among your people, you are warned against cursing him.”
— Mekhilta of Rabbi Ishmael, Mishpatim, Tractate Nezikin, parashah 5, on the phrase “shall surely be put to death”

Introduction

The midrash (rabbinic exposition) quoted above seeks a warning for the offense of cursing one’s father and mother, whose punishment is stoning. In the end, the warning is found through a binyan av from the combination of three sources: judge, prince, and deaf person. This midrash raises several fundamental issues that deserve separate discussion: warning and its meaning, warning and punishment derived from inference, and the structure of the exposition, which is a complex common-denominator argument from three verses rather than two.

Warning and Its Meaning

The term “warning” in rabbinic language means the violation of a prohibition, that is, a negative commandment. See, for example, Maimonides’ wording at the beginning of Root 8 in his Sefer HaMitzvot (Book of the Commandments). The Sages take it as a simple premise that one cannot punish for any offense, even if the punishment is written explicitly in the Torah, unless there is an additional source, called a “warning,” that establishes the act as forbidden—that is, as an offense. This is what the Sages mean by the principle: Scripture does not punish unless it has first warned. See, for example, Babylonian Talmud, Yoma 81a, and parallels. This is the source of the expression—though it does not actually appear in rabbinic literature—”one does not punish unless one first warns.” Because of this assumption, in many places where a punishment appears without a warning, the Sages ask, as in the midrash above: “We have heard the punishment; from where is the warning?”

The need for a warning is explained in Sefer HaChinukh, mitzvah (commandment) 69, as follows:

“For the mere mention of the punishment in a commandment, without a warning, would not suffice us… For if the divine restraint were not conveyed to us, and Scripture merely said that whoever does a certain act will be punished thus, it would imply that anyone who wished could accept the punishment, disregard the pain, and transgress the commandment, without thereby acting against the will of God, blessed be He, and His command. The commandment would then become something like a business transaction: whoever wishes to do a certain act may pay such-and-such and do it, or may shoulder such-and-such suffering and do it. That is not the intention of the commandments. Rather, God, for our good, restrained us from certain matters, and regarding some of them informed us of the punishment that reaches us immediately, apart from the violation of His will, which is harder than all.”

The author of Sefer HaChinukh explains that if the Torah stated a punishment without a warning, we would think that the punishment is only a kind of technical stipulation—”something like a business transaction,” in his words. Whoever wants to commit the offense may do so and accept the punishment, but the act itself would contain nothing contrary to the will of the Holy One, blessed be He. The Sages teach us that this conception is mistaken.

In jurisprudence, one distinguishes between rules of adjudication, meaning punishment, and rules of conduct, meaning warnings and commandments. In philosophy of law there are approaches—positivist ones, such as that of the Austrian legal thinker Kelsen—that maintain that there is no place in law at all for rules of conduct. This conception also fits the liberal assumption that the legislator cannot dictate to the citizen what he may or may not do. According to this approach, the law, by definition, addresses only the governing authorities. For example, in Israeli criminal law there is no clause that prohibits murder or theft as such, but only a specification of the punishment for those acts.¹ As noted, in halakha (Jewish law) the situation is different: there are rules of both kinds, and in fact one cannot punish without a warning.

Warning and Punishment from Inference

There are two rules that pertain to these two kinds of laws; see the discussion in Babylonian Talmud, Yevamot 22b:

1. One does not punish on the basis of inference.
2. Even one who disagrees and maintains that punishment may be derived from inference will still agree with the rule: one does not derive a warning from inference.

That is, with respect to punishment there is a dispute whether it may be learned from the kind of reasoning called din, but with respect to warning it is clear that it cannot be learned from that sort of reasoning.

The accepted interpretation of the term din here is qal va-homer (an a fortiori inference). Thus it is written as a simple and agreed matter in Encyclopedia Talmudit, s.v. “One Does Not Derive a Warning from Inference.” However, Maimonides, in Root 2, maintains that the term din here means all the hermeneutical principles by which the Torah is expounded. In his view, one does not derive warnings at all by means of derashot (exegetical inferences). Nachmanides, in his glosses there, discusses this point at length in disagreement with Maimonides and raises many difficulties from the Talmud, but this is not the place to enter into them.

It should be noted that according to most views, the very existence of a prohibition can certainly be learned from inference. After all, the hermeneutical principles are the source of many prohibitions. A warning from inference is unacceptable only for purposes of punishment; that is, one cannot punish on the basis of a warning learned from inference. Thus both punishment from inference and warning from inference deal, in one way or another, with punishment. We should note that Maimonides’ view here is different, for according to him, in Root 2—see last week’s sheet—even the prohibitions themselves that are learned through the thirteen hermeneutical principles are not Torah law but rabbinic law. In other words, on his view the rule that one does not derive warnings from inference means that no biblical prohibition is created at all, not merely that one cannot punish for such an offense.

There are several qualifications to these rules. First, in a number of Talmudic discussions it is clear that warnings may be derived from heqesh (textual juxtaposition), for example Babylonian Talmud, Pesachim 24b, and likewise from gezerah shavah (verbal analogy), Babylonian Talmud, Zevachim 106b, and Rashi there, s.v. “atia.” As for mah matzinu (derivation from a precedent) and ma ha-tzad (the common-denominator inference), later authorities disagreed over whether warnings may be derived from them; see Knesset HaGedolah, Hoshen Mishpat 359, Hagahot Beit Yosef 3, and Encyclopedia Talmudit there, note 56.

A Temporary Return to Our Midrash: A Case in Which the Punishment Is Explicit in the Torah

Maimonides, in his introduction to the enumeration of the commandments, and likewise in Negative Commandment 318 in Sefer HaMitzvot, raises a difficulty about the midrash cited above, and writes as follows:

“We say that one does not derive a warning from inference only when we seek, by way of analogy, to prohibit something for which no special prohibition has been made explicit. But when the Torah explicitly states the punishment for one who performs that act, we necessarily know that it is a forbidden act against which one has been warned. We then extract the warning by inference only in order to reinforce the principle that Scripture does not punish unless it has first warned…”

At the beginning of his remarks, Maimonides notes that our midrash apparently contradicts the rule that one does not derive a warning from inference, since it learns the warning for cursing one’s father and mother from the common denominator, which is one of the thirteen hermeneutical principles.² It should be noted that this difficulty does not arise at all according to most medieval authorities, for whom din means only qal va-homer. Here, however, we are dealing with the common denominator, and with hermeneutical principles other than qal va-homer one may derive warnings from inference. Maimonides, as noted, assumes that din means every form of exposition among the thirteen principles, and therefore it is impossible to derive a warning even through the common denominator, just as through all the other principles. Still, we have seen that even regarding the common denominator, later authorities disagreed, and therefore it seems that some commentators would adopt Maimonides’ answer, though not for his own reason.

Maimonides resolves the difficulty by saying that here a warning derived from inference is acceptable, because the punishment for cursing one’s father is written explicitly in the Torah: “shall surely be put to death”—see the first part of the exposition quoted above.

The Rationale of These Two Rules

The commentators do not seem to explain the rationale of the rule that one does not derive warnings from inference. There is, however, a source for it in Babylonian Talmud, Yevamot 22b: Scripture needed to warn regarding intercourse with one’s sister who is the daughter of both one’s father and one’s mother, even though a warning had already been stated regarding one’s sister who is the daughter of one’s father but not of one’s mother; and the former warning could seemingly have been learned by a qal va-homer from the latter.³

By contrast, with respect to the rule that one does not punish on the basis of inference, later authorities offer three well-known explanations; see Encyclopedia Talmudit, s.v. “One Does Not Punish on the Basis of Inference,” and Beit HaOtzar, section 1, principle 56, among others:

1. Since qal va-homer is a rational analogy that a person formulates on his own, there may be a refutation of the a fortiori argument; see Halikhot Olam and Middot Aharon 2:13.
2. Since the derived case is more severe than the source case, perhaps the lighter punishment is not severe enough to atone for it; see Maharsha on Babylonian Talmud, Sanhedrin 64b, and elsewhere.
3. It is simply a scriptural decree, learned from the verse about “his sister”; see Rabbi Elijah Mizrachi on Parashat Shemini.⁴

In light of these reasons, one may also think about the rationale for the rule that one does not derive warnings from inference. At first glance, the second direction seems inapplicable, since we are not dealing here with punishment at all. However, as noted above, a warning from inference certainly does suffice to create a prohibition. The problem of “warning from inference” arises only when we are dealing with the question whether there is a warning as a condition for liability to punishment. If so, the rule that one does not derive warnings from inference also concerns punishment, but at a different level. According to Maimonides, who applies this rule to the prohibition itself, one indeed cannot follow the second direction.

If so, all the explanatory possibilities raised with respect to punishment from inference are also relevant to the rule that one does not derive warnings from inference. More than that: as we saw, the Gemara in the passage in Yevamot states that even one who holds that punishment may be derived from inference agrees that warnings may not be so derived. That is, the rule that one does not derive warnings from inference is even more obvious and more broadly accepted than the rule that one does not punish from inference. But this itself requires explanation according to all views: why is the rule concerning warning more self-evident than the rule concerning punishment?

The obvious explanation lies in the distinction between the situations. A case in which punishment is learned from inference is always a case in which there is an explicit warning in the Torah, but no punishment is mentioned. If so, the very existence of the warning supports the inference made regarding the punishment. The fact that there is a prohibition grounds the qal va-homer regarding the punishment. By contrast, in a case of warning from inference, we are not dealing with a punishment that is explicitly written, and therefore it is harder to infer the existence of a warning. More than that: according to Maimonides, when the punishment is written explicitly, then warnings may indeed be derived from inference. Thus, in the ordinary case of warning from inference, we have no additional support for the inference we made, and we therefore worry about the possible invalidity of its conclusion.

Accordingly, what Maimonides wrote is quite plausible: when the punishment is written explicitly, there is no problem of warning from inference. The reason is simple: in such a case, the problem of warning from inference is completely parallel to the problem of punishment from inference, since the written punishment supports the inference regarding the warning.

Maimonides’ View on the Matter

All this applies to the other medieval authorities. Maimonides, however, understands the rule “one does not derive warnings or punishments from inference” as referring to all forms of exposition, not specifically to qal va-homer. If so, the question of severity is not relevant to understanding the rule. Beyond that, the question of refutation also seems puzzling if one applies it to all the hermeneutical principles, and not only to qal va-homer, which any person can formulate on his own.

As we have seen, Maimonides also understands this rule on the plane of the prohibition itself, not merely as the existence of a condition required for imposing punishment. According to Maimonides in Root 2, even a prohibition learned from one of the hermeneutical methods is not itself a Torah prohibition. Consequently, there is no warning here, and punishment cannot be imposed. For Maimonides, the rule that one does not derive warnings from inference does not concern a condition attached to punishment, but the prohibition itself.

The deeper reason for this lies in an entirely different conceptual framework. According to Maimonides, the concepts de-oraita (of Torah origin) and de-rabbanan (rabbinic) do not express halakhic force; rather, they classify the sources of halakha. In many places it is clear that for Maimonides a de-oraita halakha is a law written in the Torah, whereas a de-rabbanan halakha is a law not written explicitly in the Torah, and not necessarily only enactments and decrees, as that term is commonly understood. For Maimonides it includes laws given to Moses at Sinai, and also laws that arise from derashot; see, for example, his Commentary on the Mishnah, Kelim 17:12, and elsewhere.⁵

We saw in last week’s sheet that according to Maimonides, derashot are tools for extending the written text, not interpretive tools that uncover its intentions. If so, it now becomes clear that a halakha that emerges from an exposition is not latent in the text from the outset; rather, it is an extension created by the Sages who expound it.

For precisely this reason, according to Maimonides, one does not derive warnings from inference. For Maimonides, a “warning” is not a condition for punishment, but the creation of a prohibition, in the original sense of the term “warning.” When there is no explicit source in the Torah, this is not a de-oraita law, and therefore there is no warning. The law that arises from the exposition is not found in the Torah, and therefore we are not commanded regarding it directly by the Creator, but only through the obligation to obey the Sages, under the prohibition “Do not stray.” Consequently, the Torah does not assign a punishment to it. According to Maimonides, the problem here is not a lack of prior warning; rather, there is simply no prohibition with full biblical force. Therefore one does not punish for it.

By contrast, if the warning is written explicitly, the prohibition certainly exists, and the problem concerns only the punishment. Here the difficulty is smaller, for a Torah prohibition was certainly violated, and the fact that we learned the punishment by exposition should not prevent us from imposing it. Therefore the rule that one does not punish from inference is more novel than the rule that one does not derive warnings from inference, which is virtually necessary on Maimonides’ view. Nevertheless, in the end, one still does not punish from inference, because the punishment too must be written explicitly in the Torah, and an exposition is not considered as though it were written in the Torah.⁶

By contrast, when the punishment is written explicitly in the Torah, it is clear that there must be a warning, even if hidden, since the Torah explicitly establishes the punishment; see the quotation from Sefer HaChinukh above. In such a case, if we have a warning from midrashic inference, Maimonides understands that the exposition reveals the hidden warning for which the written punishment is the sanction. Therefore Maimonides determines that in such a situation warning from inference is valid as well.

B. Returning to the Midrash: The Structure of Two Complex Binyan Av Inferences, One Built on Top of the Other

Description of the Midrashic Argument

Let us now return to the midrash above. The midrash is built as a complex argument: first, it learns from the common denominator of judge and prince; then that common denominator is refuted; then the prohibition against cursing a deaf person is added. In the final accounting, the prohibition against cursing one’s father is learned from all three together. Let us try to examine the structure of this complex exposition.

The midrash begins by attempting to derive the warning for one who curses his father from one who curses a judge or a prince. But this attempt is puzzling at first glance, for it is clear that even if his father is a judge or a prince, the prohibition against cursing him would be because he is a judge or a prince, not because he is his father. If so, how can one extract from here a prohibition against cursing one’s father? One might say that from judge and prince we learn that it is forbidden to curse a person who holds authority, a kind of office, and therefore there is also a prohibition regarding a father. But this too is implausible for two reasons:

1. The common denominator mentioned in the midrash is not that they hold authority, but that they are “among your people.”
2. Immediately afterward, later in the midrash, a refutation is raised against the attempt to learn from judge and prince, namely, that both are “prominent among your people.” But that is precisely the reason for the derivation, for one’s father too is “prominent,” relative to him. Why, then, does the midrash treat this claim as a refutation?

It follows that the learning from judge and prince is to be taken literally: both are “among your people,” meaning part of the people in the sense of one who lives as a member of your people by observing the basic commandments, and therefore one owes them the ordinary duties between one person and another. The derivation is rejected because these two are objectively prominent, unlike an ordinary father, who is “prominent” only relative to his son. Therefore the prohibition against cursing a deaf person is brought, since he is usually a simple and vulnerable person—”the most downtrodden among people.”⁷ In the end, the derivation is from all three sources together, and their common denominator is that all of them are “among your people.” That is, there is a prohibition against cursing any Israelite.

The Difficulties in the Midrash: Four Possible Structures

There are several difficulties here. First, it would seem that if the conclusion learns from three verses, then even the prohibition against cursing a judge and a prince is not because they are authorities. If so, all these are details of the same foundation: proper conduct, rather than laws of public discipline and governance toward authorities. Yet it seems strange to say that all these verses deal with prohibitions included under one and the same foundation. There are four possibilities for understanding the structure of the exposition, and they are described in the diagram mentioned below.

One possibility is to understand the learning as built from two common-denominator structures composed one upon the other,⁸ each of them based only on two source cases: one learns from the common denominator of judge and prince, and that whole unit becomes one side, which joins the side of the deaf person; together they then teach in another common-denominator structure. If the structure is like this, then one can indeed understand that two foundations are joined here: from judge and prince we learn the attitude toward authorities—a kind of commandment between human beings and God, since prince and judge are also representatives of Heaven—and from the deaf person we learn the demand for respectful treatment of every human being, which is an interpersonal duty. The prohibition toward one’s father is a combination of both these foundations: both interpersonal and God-directed.

Two proofs may be adduced that this is indeed the logical structure of the exposition:

1. The exposition presents the mutual necessity of the three elements not as the relation of each one to the other two, but as the relation between judge and prince, followed by the relation between those two and the deaf person.
2. In the account of why all three are needed, it is not explained why it would not have been enough to write only the prohibition against cursing a deaf person. At first glance, the whole argument indicates that the deaf person is the most innovative case, and therefore one would expect the midrash specifically there to explain why that warning alone would not suffice. According to what we have said about the logical structure of the exposition as a composition of two common-denominator structures, which are ultimately built from two different foundations of prohibition, the answer is simple. Had only the prohibition against cursing a deaf person been written, we would not have known the second foundation, namely the duty of respectful treatment toward authorities. We would have thought that the prohibition of cursing is only the duty of proper treatment toward every other person, at least one included in “your people.” And conversely, if only those two had been written, we would not have known the duty to relate respectfully to the individual person, and we would have understood the prohibition only as a duty of respect toward authorities.

Thus we arrive at the first possibility regarding the structure of the exposition: the triple necessity is not built on a single level of common denominator from three verses. Rather, these are two levels composed one upon the other, and each level is built as a common denominator from two verses.

On the sheet for Parashat Shemot we considered several ways to understand the structure of the common denominator. In the end, all the possibilities seemed to assume that there cannot be more than one reason for a given law, and therefore there is always one common denominator that is the cause of the derived law. Here, however, we see a different situation: we learn the prohibition against cursing a father from the prohibition against cursing an authority, combined with the prohibition against cursing another Israelite. This raises the question: why should we not say that there is here a “refutation from a stricter side”? In other words, why can we not refute this complex derivation as follows: what is unique about a judge and prince? They are authorities. And what is unique about a deaf person? He is another Israelite.⁹

It would therefore seem that along this line we must conclude that there is here an entirely new structure. We learn that with regard to a father there are two different laws: he is both an authority and another ordinary Israelite. Both prohibitions apply to him together. If so, we arrive at a second possibility regarding the structure of the exposition: this is not a common-denominator argument at all, but two independent derivations—what yeshiva language would call “two laws,” or “two aspects.”

A third possibility for the structure of this exposition is that one can build a more basic common denominator out of these two foundations—for example, the duty not to damage elements of the social order, whether toward the downtrodden or toward authorities, or in some other formulation.¹⁰ In any case, that does not seem to be the intent of the language of the midrash, for it presents a necessity among these three components themselves. The midrash also summarizes everything by saying that all of them are “among your people,” implying that this is the common denominator from which we also learn about a father. If so, this indeed appears to be a one-level derivation whose result is a prohibition against cursing someone from Israel.

The conclusion at this point is that there may be a fourth possibility for the structure of the exposition: there are three verses in the exposition, each needed for a different reason, yet all of them together teach one foundation. They all operate on one level; this is the simple understanding from which we began. Here the question arises with even greater force: why did the Torah not suffice with the deaf person alone? After all, from him one can learn the prohibition against cursing any Israelite, including a prince and a judge. One might perhaps say that the deaf person is unique in being especially downtrodden, and therefore perhaps only with respect to him is there a prohibition of cursing, unlike a prince and a judge, who possess power and security and are less harmed by the insult.

In any event, even the fourth understanding, according to which this is a one-level structure, is problematic for the two reasons raised above: the wording of the midrash indicates that there are two common-denominator structures composed one on top of the other. How, then, should we understand the structure of this midrash? Is it one-level or two-level? It seems that, in order to understand this, we must return to the discussion of warning and punishment.

The Solution: Distinguishing Between Warning and Punishment

The exposition in this midrash is unique. As noted, ordinarily the question of warning from inference arises in a situation where no punishment is stated explicitly in the Torah. Here, by contrast, we are dealing with a case in which a warning is needed for a punishment that appears explicitly in Scripture.

In such a case, the essential character of the law may play no role at all. When there is an explicit punishment in Scripture, all we are looking for is a source that establishes the very fact that there is a prohibition, not necessarily a source that serves as the foundation of the prohibition itself, for that may also be taken from the punishment. If so, one may say that there are indeed several different foundations here, such as authority and vulnerability, while the foundation regarding a father is a third foundation, different from both. If we can prove that there is a prohibition against cursing a father, regardless of whether we prove it by the foundation of authority or by the foundation of respect for every upright member of the community, then the punishment written in the Torah can be imposed for that prohibition. In other words, we need only proof that the prohibition exists, even if its essential foundation in the source cases is different.

As an example, commentators dispute whether one may derive a warning from inference when the Torah already contains a positive commandment, or a prohibition implied by a positive commandment. In such a situation, the derivation is only a gilui milta (a clarification rather than a substantive derivation); see Encyclopedia Talmudit, s.v. “One Does Not Derive a Warning from Inference,” chapters 3-4, and Nachmanides’ novellae to Babylonian Talmud, Yevamot 22b. That is enough to count as a warning, even though it is “from inference.” In such a case, the exposition yields a prohibition whose definition differs from that of the source case—and not only its legal force, for the difference between a positive and a negative commandment is not only one of legal force but of the very nature of the prohibition. Even so, the inferential reasoning is effective in showing that there is a prohibition here—a negative prohibition, that is, a warning. The foundation and definition of the prohibition are learned from the explicit commandment that was written.

The same is true in our case. Since there is a prohibition against cursing one’s father either because he is like any other Israelite, as learned from the deaf person, or because he too is a kind of authority relative to his son, as learned from judge and prince, we have proved that there is a prohibition against cursing a father. It is now possible to punish for it, as is written explicitly in the Torah. But now one may return and learn from the explicit punishment in the Torah—”shall surely be put to death,” which differs from the punishment for the other prohibitions of cursing—that the definition and nature of the prohibition against cursing a father is different. This is similar to what we saw above in Maimonides, when he determined that a warning from inference is valid where the punishment is explicit in Scripture.

It should be noted that at the end of mitzvah 69 in Sefer HaChinukh it is written that one who curses a judge incurs two violations punishable by lashes: one because the judge is like any other upright Israelite, and the second because he is a judge. It follows from this that these are two different prohibitions, with different foundations. The same is true of one who curses a father; that too is a distinct foundation. We thus see that the derivation from all the other prohibitions is only a clarification that a prohibition exists, and therefore that punishment can be imposed. The definition of the prohibition is learned according to the specific case, and it is different in each of them.¹¹

Now, in light of all this, we can see that these two structures are equivalent. The derivation from two different foundations—the first structure, that of two common-denominator derivations composed one on top of the other—and the fourth structure, that of a single level learning the father from all three together, are both merely a clarification teaching the fact that a prohibition exists. In the end, it is clear that there are two different foundations even in the cases of judge and prince, and certainly the prohibition regarding the deaf person is different from both of them, while the prohibition of cursing a father defines a fourth foundation, different from them all.

Footnotes