Gate Two: Hermeneutical Principles — Their Source and Meaning

From the book The Spirit of Law by Rabbi Michael Avraham. Translated from Hebrew using gpt-5.4 (reasoning_effort=high, batch API).

---

Hermeneutical Principles, Their Source and Meaning

The hermeneutical principles of derash are an enigma for the contemporary student. On the other hand, the Vilna Gaon already wrote in his commentary to Proverbs 8:9 that the verse, “They are all straightforward to one who understands,” was said about scriptural expositions, and it teaches us that we are charged with the task of deriving one matter from another within the verses of Scripture. One of the foundational tools for deriving one matter from another in the verses of the Torah is the system of hermeneutical principles. In this gate we will discuss the hermeneutical principles in a general way, though not the character and essence of each individual principle; that subject is treated in greater detail in my book Midah Tovah (2005–2006).

The present gate is only an introduction to the second part of the book. We will begin our inquiry in the second root with a brief survey of the different systems of hermeneutical principles, together with a short discussion of their source, nature, and significance. We cannot, of course, encompass all the issues connected to this broad topic, and we will present here only what is necessary for our discussion.¹

Chapter One: The Systems of the Principles and Their Source

Introduction

In this chapter we will describe the different systems of hermeneutical principles, and examine the relationships among them and their development. This background is necessary for what follows in this part of the book.

Different Systems of Hermeneutical Principles

The earliest historical point at which hermeneutical principles appear is apparently the discussion in the Babylonian Talmud, at the beginning of the sixth chapter of Tractate Pesachim, which deals with the laws of the Passover offering when the eve of Passover falls on the Sabbath. The Gemara there—and see also the Tosefta, Pesachim 4:11, and the Jerusalem Talmud there—relates that this law was forgotten by the Benei Beteira, and Hillel the Elder, who had come up from Babylonia, reconstructed it by using several hermeneutical principles: kal va-homer (an a fortiori inference) and gezerah shavah (verbal analogy).

It is known from several sources that Hillel the Elder had seven principles, which he expounded before the Benei Beteira; see the Tosefta, Sanhedrin 7:5, and the conclusion of the baraita of the thirteen principles at the opening of Sifra. There are variations among the different textual versions regarding which principles were included in Hillel’s list; see the remarks of Rabbi David Cohen, known as the Nazir, cited in Dov Schwartz’s article in Sefer Higayon. This is apparently the earliest historical appearance of a formal system of hermeneutical principles.

The best-known system of interpretation is the thirteen principles of the school of Rabbi Ishmael, hereafter: R. Ishmael. Their source is Rabbi Nehunya ben HaKanah, who is described as R. Ishmael’s teacher; see Babylonian Talmud, Shevuot 26a. They are presented in the baraita of the principles at the beginning of Sifra. In several Talmudic sources—see Babylonian Talmud, Shevuot 26a and the parallel passages—there are hints of a parallel system of principles that was used in the school of Rabbi Akiva, whose source is Rabbi Nahum of Gimzo, Rabbi Akiva’s teacher. That school expounded by means of “inclusions and exclusions,” rather than “general rules and particulars.”

Another system of hermeneutical principles appears in Sefer Keritut, at the beginning of part 3, which is called Netivot Olam. Rabbi Samson of Chinon brings there the baraita of Rabbi Eliezer, son of Rabbi Yose the Galilean, which enumerates thirty-two principles. See the remarks of the Nazir, cited by Schwartz there, note 14, that there is some ambiguity regarding their precise number. We have no primary source for this baraita apart from the version cited in Sefer Keritut.² Some also associate this list with the school of Rabbi Akiva, hereafter: R. Akiva.

As noted, the best-known list of principles is the list of R. Ishmael’s thirteen principles, which appears in the baraita that opens Sifra:

“By thirteen principles is the Torah expounded: by kal va-homer and by gezerah shavah…”

The relation between this list and the other lists is unclear.³

It is not really clear what the disagreement is between the school of R. Ishmael and the school of R. Akiva. Did each of these schools use a completely different list? It seems that there was agreement between them on some of the principles, but it is not entirely clear on what they disagreed and on what they agreed. Beyond that, there are several other interpretive methods that do not appear in R. Ishmael’s list, and yet it is difficult to assume that he disputed them; there are places in the Talmud where R. Ishmael himself, or his students, make explicit use of such principles. The same is true of the list of thirty-two principles of Rabbi Eliezer son of Rabbi Yose the Galilean, which lacks several of R. Ishmael’s principles, while R. Ishmael’s list omits several of the principles enumerated there.

The Relationship Between the Different Systems

The author of Sefer Keritut—see the general introduction to the section Netivot Olam—addresses the question why R. Ishmael did not enumerate all thirty-two principles of Rabbi Eliezer son of Rabbi Yose the Galilean, and offers several answers. In the course of his detailed commentary on each principle, he also explains, in each case, which of those answers is relevant. The best-known and most widely accepted explanation is that the thirty-two principles are principles of aggadic midrash (non-legal exposition), whereas the framework of R. Ishmael’s thirteen principles includes only the principles of halakhic midrash (legal exposition). However, as noted, Rabbi Samson offers additional lines of explanation, since there are certainly other hermeneutical principles that operate in halakha (Jewish law) and are not counted by R. Ishmael, and yet it appears—either as a matter of logic, or from explicit evidence in Talmudic discussions—that R. Ishmael accepts them as well.

Rabbi Samson presents the following possibilities there:[^84]

1. R. Ishmael’s principles are those on which everyone agrees.
2. Some of the principles are regarded as explicitly written in Scripture, and therefore were not included in R. Ishmael’s list.
3. Some are meant only to resolve the plain sense of the verses—such as “a great matter is attached to a small one”—and therefore were not included.
4. Others are principles of aggadic exposition, or exegetical principles meant to settle the plain sense of Scripture,⁴ but no halakhic conclusions can be drawn from them, whether lenient or stringent.

Let us add just one further important point: the meaning of explanation 2 above, according to which some of the principles are considered explicitly written in the verses themselves. This principle will accompany us later as well, and so we will pause over it briefly here. As emerges from Rabbi Samson’s remarks in the course of his detailed explanation of that baraita, his intention seems to be that some of these principles hinge on surplus words—“free” terms—that are unnecessary from the standpoint of peshat (the plain sense of the text). The conclusion drawn from the existence of such a surplus word is that an exposition must be made here.

Expositions of this kind are regarded by him as explicitly written in the verse, because they constitute the only possible interpretation of those free words. According to this approach, an interpretation counts as derash only when it comes as a “second story,” beyond the plain-sense interpretation of the verse. But when a given word has no plain-sense interpretation at all, the derash itself becomes its plain-sense interpretation, and therefore it does not function as derash.⁵ In the end, the derash is the only interpretation of that word in Scripture, and therefore the product of such an exposition is regarded as explicitly written in the verse, in that word. That is what the verse says; it is not something derived from it. By contrast, a derash that does not explain a “free” word, but proposes a parallel explanatory layer beyond the plain sense of some biblical words—only that counts as derash. What is written explicitly in the verse counts as plain-sense interpretation, whether we arrived at it through classical interpretive tools or through the tools of derash. Later we will see several important implications of this distinction.⁶

It may perhaps be possible to add another line of explanation for why there are halakhic principles that do not appear in the baraita of the thirteen principles, beyond the four directions proposed by Rabbi Samson:

1. The baraita of the principles deals with the interpretive methods of the school of R. Ishmael; therefore it is reasonable that it does not enumerate principles that are unique to the school of R. Akiva. For example, as we noted above, some associate the thirty-two principles of Rabbi Eliezer son of Rabbi Yose the Galilean with the school of R. Akiva.⁷

The Source of the Hermeneutical Principles

As for the source of the hermeneutical principles, there is full agreement among all the commentators who addressed the matter that they are included within the category of halakha le-Moshe mi-Sinai (a law transmitted to Moses at Sinai). So wrote Maimonides in his introductions—see, for example, the beginning of his introduction to the Commentary on the Mishnah—and Nahmanides in his objections on this root,⁸ and Rabbi Abraham ben David in his commentary on the baraita of the principles in Sifra, and Rashi on Babylonian Talmud, Pesachim 24b, s.v. “And if it is not germane,” and Sefer Keritut, part 1, section A, no. 9, cites this view from the responsa of the Geonim; many later authorities wrote likewise.⁹

Some, however, were uncertain whether the aggadic hermeneutical principles were also transmitted from Sinai. The rulings of Maimonides, Rabbi Abraham ben David, and the commentators on the principles—that “the principles” were transmitted to us as halakha le-Moshe mi-Sinai—refer only to R. Ishmael’s thirteen principles. Rashi, by contrast, in the passage cited above in Pesachim 24, says this with regard to all the principles. Several later authorities followed these two approaches as well. See, for example, Responsa Radbaz, no. 232, and Mevo ha-Talmud by Maharatz Chajes, chapter 19, who hold that all the principles are halakha le-Moshe mi-Sinai. By contrast, the Shelah, at the end of the section on the Oral Law, raises the possibility that only the principles of halakhic derash were transmitted from Sinai.

The common denominator among all of them, medieval and later authorities alike, is that the principles of halakhic derash, and certainly R. Ishmael’s thirteen principles, are halakha le-Moshe mi-Sinai. This determination itself belongs to the halakhic sphere: the force of the principles is that of halakha le-Moshe mi-Sinai, which according to most opinions has the status of biblical law. Maimonides’ exceptional view on this issue will be discussed below. But it is important to note that the background of this claim lies in the historical sphere: the principles were not created over the generations but were received by Moses at Sinai.

And yet, despite all this, almost all scholars have agreed, and still agree, that the principles are a late development. Most of them maintain that they began to crystallize in the days of Hillel the Elder, and continued to develop through the time of R. Ishmael and afterward. The main evidence for these claims is that within the rabbinic sources themselves there is hardly any source stating that the principles are halakha le-Moshe mi-Sinai. Another piece of evidence is the process of crystallization and elaboration of the principles that we described above.

Still, one can also find rabbinic sources for the claim that the principles are halakha le-Moshe mi-Sinai. For example, in Babylonian Talmud, Sanhedrin 99a, we find:

“Even if one says: The entire Torah is from Heaven, except for this precise inference, this kal va-homer, or this gezerah shavah—this is what Scripture means by ‘For he has despised the word of the Lord.’”

To be sure, one may distinguish between something being “from Heaven” and its being a halakha le-Moshe mi-Sinai.

And in the discussion in Babylonian Talmud, Temurah 16a—this is also quoted at the end of Maimonides’ remarks on this root—we find:

“One thousand seven hundred kal va-homer inferences, gezerah shavah inferences, and scribal precisions were forgotten during the days of mourning for Moses…”

From here it appears quite clearly that the principles as well were transmitted to Moses at Sinai, and had already been forgotten during the mourning for him. At least from the historical perspective, there is here a conception that the principles are extremely ancient. The same is true in several aggadic midrashim, such as the following midrash, from Midrash ha-Gadol, from which David Zvi Hoffmann excerpted passages under the title Mekhilta de-Rashbi on Mishpatim, Exodus 21:1. See also the midrash on the thirteen principles in Midrash ha-Gadol, Leviticus, cited by David Zvi Hoffmann in Birkat Avraham in honor of Rabbi Avraham Berliner’s jubilee.¹⁰

“Rabbi Ishmael says: These are the thirteen principles by which the Torah is expounded, which were transmitted to Moses at Sinai.”

Likewise in Bereishit Rabbah (Theodor-Albeck edition), section 46, s.v. “And I shall give My covenant,” and in the parallel in Vayikra Rabbah 25, s.v. “Rabbi Huna”:[^92]

“Rabbi Huna said in the name of Bar Kappara: Abraham sat and reasoned by gezerah shavah… Rabbi Hanina said to him: Were gezerot shavot already given to Abraham? That is astonishing…”

This proves that the principles were regarded as something that was “given” at some stage after Abraham our forefather. For if the principles had indeed developed naturally, it is not clear what would prevent us from assuming that Abraham too could have arrived at them. In other words, there is here a twofold claim: first, the principles did not develop naturally and spontaneously, but were “given” to us from above; second, there is also a claim here about the timing of that “giving”: after Abraham our forefather, and presumably at the revelation at Mount Sinai.

On the other hand, it is hard to ignore the fact that these are incidental references, and most of them are aggadic. Indeed, in practice the principles are not mentioned before the period of Hillel, and even in his case only seven principles appear. Beyond that, from Hillel onward there is a visible process of crystallization, which also includes the emergence of disputes about them, and in particular the development of two different schools concerning the modes of exposition—R. Akiva and R. Ishmael, mentioned above. We should recall Maimonides’ statement that no dispute ever arose concerning something transmitted from Sinai, in the introduction to the Commentary on the Mishnah and at the beginning of our root.¹¹ If so, we see that disputes did arise over the system of principles over the generations, and perhaps they were even entirely forgotten by the sages of the Land of Israel in the period of the Benei Beteira. It would therefore seem, at first glance, that they are not halakha le-Moshe mi-Sinai.

To understand the matter, let us take as an example the principle of kal va-homer, one of the principles by which the Torah is expounded. We know that even in the Talmudic period there was a dispute concerning the rule of dayyo—“it is enough,” meaning that the conclusion of an a fortiori inference may not be more stringent than its premise—which is part of the principle of kal va-homer; see, for example, Babylonian Talmud, Bava Kamma 24b–25a. More than that, the Gemara there brings sources for kal va-homer and for dayyo from the Torah itself. If so, according to Maimonides’ premises it follows necessarily that kal va-homer—and likewise all the other principles—was not really transmitted at Sinai.¹² Yet Maimonides himself states that the principles were transmitted to Moses at Sinai, and that their halakhic status is that of halakha le-Moshe mi-Sinai.

The conclusion that follows from this is that the hermeneutical principles are apparently not concrete Sinaitic laws, like all the other laws of that category. Rather, they are forms of contemplating the Torah, which were transmitted together with it from Sinai in a primordial and abstract form, and only afterward, over the course of history, underwent a process of crystallization and formalization. Let us now try to sharpen this claim.¹³

Conceptualization and Formalization in Historical Perspective

Our main claim is that the two pictures described so far—the academic-research picture, which sees the principles as the product of historical development, and the traditional picture, which sees the principles as a complete system given at Sinai—are not necessarily contradictory.¹⁴ It is highly plausible that the principles were not transmitted to Moses at Sinai in a crystallized form. Rather, they were modes of understanding embedded within the natural reading of the Torah, as it was transmitted to Moses from the mouth of the Holy One, blessed be He. The Holy One, as it were, studied the Torah with Moses our master at Mount Sinai, and His explanations were given to Moses in this “language.” He read a verse and explained to him the derash that arose from it. But this was a kind of language, not a crystallized formal system with explicitly formulated rules. For example: “You shall fear the Lord your God”—to include Torah scholars. The Holy One did not explain to Moses that there is a general law that the particle et is always expounded to include something additional. Rather, He read the verse with him and accompanied it with the halakhic result that is learned—or, more precisely, expounded—from it. Over time, forgetfulness affected the hermeneutical principles, because it is very difficult to transmit from teacher to disciple intuitive ways of seeing that are neither formulated nor defined. As a result of this forgetting, various methods of reconstruction were used in order to restore the use of the tools of derash, and the halakhic results produced by that use. The matter resembles language acquisition: an infant learns language from his parents and surroundings without needing formulated and defined rules. He uses them intuitively. But a new immigrant studying in an ulpan must enter into the use of the language through explicit, formulated rules. In the process of the forgetting of Torah, the rules become increasingly refined, and more and more intuitions undergo formalization, formulation, and conceptualization, becoming distinct rules.

In the course of the continuing reconstruction of what was given to Moses at Sinai, explicit and defined rules of interpretation developed, like rules of grammar that are created only after a language already exists and functions intuitively. The reconstruction of such a phenomenon is always accompanied by conceptualization and formalization. As we described in several essays in Midah Tovah,¹⁵ the dialectical reconstruction by Othniel ben Kenaz after the mourning for Moses, which restored the use of the hermeneutical principles, was probably such a process of reconstruction.

It is reasonable to assume that, as a result, the process of reconstruction—which is fundamentally a synthetic process of abstraction and generalization, and therefore not a one-to-one process—generated several different ways of looking at these principles. Thus different approaches, or different schools, arose, each using the hermeneutical principles somewhat differently. The reconstruction did not yield a single, unequivocal result, because what is at issue is a kind of “language,” not a system of rules of a mathematical character. As the generations passed, more external intuition entered into the use of the methods of derash, and differences emerged due to differences in outlook among different sages and among those schools. For this reason there arose an increasing need to formalize and crystallize the system of principles into a rigid and well-defined body of rules, and thus various systems of hermeneutical principles were gradually cast into shape. Such systems are easier to transmit reliably from generation to generation.¹⁶

In the course of this crystallization, the principles were divided into different rules and classified in different ways. But by this point in the historical process, differences had already arisen in the conceptions of the various schools regarding the modes of exposition. The process of formalization created—or at least greatly sharpened—the distinction between two well-defined and separate schools, each characterized by a different general and principled conception:
1. The method of exposition by “general rules and particulars,” founded by Rabbi Nehunya ben HaKanah, and transmitted through Rabbi Ishmael and his school.
2. The method of exposition by “inclusions and exclusions,” founded by Rabbi Nahum of Gimzo, and transmitted through Rabbi Akiva and his school.¹⁷

A study of the Talmudic discussions shows that the same abstract rule given to Moses at Sinai, dealing with biblical formulations of the pattern “general-particular-general,” was expounded in one way by R. Akiva—“inclusion-exclusion-inclusion”—and in another way by R. Ishmael—“general-particular-general.” Some have traced the root of the dispute to a more fundamental meta-halakhic point.¹⁸ By the third generation of the tannaim, the generation of R. Akiva and R. Ishmael, the formalization was probably complete—hence these methods are named after them, rather than after their founders—and from that point on the sages had in their hands two different formulations of the system of principles.¹⁹

And what is the situation today? The system of R. Ishmael is in our possession, at least as a list of rules, but the skill needed to use it has been lost over the years. By contrast, the system of R. Akiva is not explicitly preserved for us. We know of its existence, and of many of its features, through various expositions in rabbinic literature that can be attributed to that school. On the other hand, there are quite a few disputes in the Talmud concerning the modes of application of the different interpretive systems. The ambiguity here is great, and it is difficult to derive a clear picture from rabbinic literature.

Moreover, it is very difficult to find a consistent method in practical halakhic decision-making with respect to the systems of principles. If we take Maimonides as an example, and examine all his rulings in discussions where disputes appear between these two schools—the school of R. Ishmael and the school of R. Akiva—it seems on the face of it that he does not maintain consistency. Some rulings follow the method of R. Ishmael, and others follow that of R. Akiva.²⁰

It may be that the reason for this is that, over the years, a certain blending took place between the two schools. Expositions from both schools came to be used in mixed fashion by the sages, and in fact, after the original schools disappeared, hermeneutical principles drawn from both of them continued to be used together. Over time, systems of exposition developed, each of which tried to synthesize those two original systems—and perhaps there were more than two—and fuse them into a single unified system.²¹ It may be that Maimonides and the other halakhic decisors relied on this, and therefore did not take into account, in their rulings, the specific character of the principles used by the various interpreters in their expositions, nor did they attempt to build a consistent structure in terms of the midrashic infrastructure underlying their rulings.

In the years that followed, in the amoraic period, the use of the hermeneutical principles steadily diminished and became fairly rare. In that period, midrash became primarily a tool that supports existing laws—that is, anchors them in Scripture—but does not generate new laws. The number of expositions, and especially those that create new laws, found among the amoraim is relatively small compared with what is found among the tannaim. With the closing of the Talmud, the situation became even more acute, and a broad cessation of the use of the hermeneutical principles came about, apparently because the skill had been lost. There is no clear record of a formal decision of that kind, but the reality is that the use of the principles—and of derash in general—almost completely ceased. One can still find more intuitive forms of exposition, and especially supportive expositions, among some medieval and later authorities, but this is a very rare phenomenon, and today it is almost entirely absent.²²

It is important to note that, from a halakhic point of view, there is no impediment to any sage’s using the interpretive system. There is no source that requires the interpreter to possess formal ordination, or any other condition, in order to use the tools of derash. Moreover, Maimonides’ language in several places implies that the use of the interpretive system was not unique to the creation of new laws in a legislative process before the Great Court, but that these were tools used by judges in the regular work of legal decision-making. One of the tools for interpreting the Torah is the system of derash, and it is used in exactly the same way as plain-sense and other interpretive considerations. For example, we will later see the view of the author of Kinat Soferim on Maimonides, on our root, that the rabbinic dictum that a person may not derive a gezerah shavah on his own does not mean that this mode of exposition is not available to human beings at all—that is, that everything was transmitted from Sinai, as Rashi and most of the medieval authorities hold. According to Maimonides, the Gemara means that a gezerah shavah may be formulated only by the Great Court.²³ By implication, the other methods of derash are entrusted to every sage.²⁴

To summarize: the picture proposed here resolves the tension between the traditional approach and the scholarly approach.²⁵ The principles were given at Sinai in a primordial form, and their force is that of halakha le-Moshe mi-Sinai. Yet they underwent crystallization, and their present shape was cast only in the later generations, from the time of Hillel the Elder onward.

We may now also be able to solve the problem of the rise of disputes concerning the hermeneutical principles according to Maimonides. As noted, Maimonides’ view is that no dispute arose regarding matters that are halakha le-Moshe mi-Sinai. Yet at the same time he states that the hermeneutical principles were given to Moses at Sinai, even though they clearly gave rise to various disputes. According to our approach, the principles are a special kind of halakha le-Moshe mi-Sinai, because they were given at Sinai only in a primordial form, not in an exact and sharply defined one. What was transmitted through the generations was not a crystallized formulation of the methods and rules of derash, but rather general descriptions and the halakhic results themselves. This enables us to understand that the reason for the disputes lies in the unique manner in which these laws were transmitted: in the process of reconstruction and crystallization, different positions arose concerning them. On the other hand, since we are speaking of tools that were transmitted to Moses at Sinai, their halakhic status is that of halakha le-Moshe mi-Sinai. The dispute concerns only the content of those tools, like any interpretive dispute, which can certainly involve biblical laws. The arguments here concern what was transmitted to Moses at Sinai, but each side in the dispute claims that what it says is what was transmitted to Moses, and therefore the laws in question are biblical.

Chapter Two: Productive and Supportive Derashot

Introduction

A common distinction is made in the world of rabbinic exegesis between “supportive midrash,” or “confirmatory midrash,” and “productive midrash.” Supportive derashot are expositions that find a scriptural source for a known law received by tradition. Productive derashot are expositions by whose power we generate a new law that was not previously known.

In accordance with this distinction, there is a widespread dispute among commentators, and also within the academic world, concerning the nature of derashot in general: are all derashot supportive in character, or are there also productive derashot among them?[^^108] We will address this matter briefly here, though without offering a comprehensive survey of the various positions. The discussion will deal with it mainly from the essential perspective—the underlying reasoning and considerations on each side—because those are what we need for what follows. As we shall see, this issue also arises at the beginning of Maimonides’ discussion of this root, and forms a foundational element in his view.²⁶

The Prevailing Approaches

Already among the medieval authorities we find thinkers who support the view that derashot only ever support preexisting laws.²⁷ The most prominent among them is Ralbag, who writes the following in the introduction to his Torah commentary:

“Their purpose in attaching these true matters, which they had received concerning the commandments of the Torah, to those verses, was that the verses should serve as a kind of hint and support for those matters. They did not mean that these laws are actually derived from those passages. For a person could overturn all the laws of the Torah by means of such inferences, to the point that one could even declare a creeping creature pure, as the sages mentioned. But I shall attach them [that is, my own interpretations] to the plain sense of the verses from which these laws can plausibly emerge, because the mind is thereby more satisfied.

“There is no departure here from the words of the sages, for they did not intend, as we have said, that those laws must necessarily be derived from the passages to which they attached them. Rather, for them these laws were received from one person to another all the way back to Moses our master, and they sought for them a hint in Scripture, as Maimonides wrote in Sefer ha-Mitzvot and in the Commentary on the Mishnah.”²⁸

From Ralbag’s words we learn that he understands derashot as supporting preexisting laws, not as creating new ones. It is important to notice that he also explains the reasoning underlying the matter. In his view, these interpretive methods cannot be trusted, since by using them one can overturn all the laws of the Torah, and even declare a creeping creature pure. He does not trouble himself to bring any evidence for this claim. It is an a priori argument, based on a line of reasoning that casts doubt on the reliability and univocality of these interpretive methods. This seems to be the main reason among almost all those who adopt such views, and what leads them to claim that all derashot are merely supportive.

We find the same in the Kuzari, at the end of Essay III, where Judah Halevi writes:

“One of two things: either they had hidden secrets, unknown to us, regarding the method of interpreting the Torah, which they possessed by tradition, such as the practice of the thirteen principles; or else their citation of verses was by way of asmakhta (scriptural support), using them as a sign for their received tradition.”

So too Rabbi Hasdai Crescas wrote in the introduction to his book Or Hashem:

“He left the totality of the commandments and beliefs in writing and orally, together with rules and signs—namely, the thirteen principles by which the Torah is expounded, which are the foundations of the Oral Law, for this is one of the mnemonic devices: to make signs for matters so that they not be forgotten.”

Similar approaches can be found among the Geonim,²⁹ and many have connected this to the polemic against the Karaites.³⁰ All these views assume that derashot function as support, as memory aids, or as signs of the received status of these laws; but it is implied that, in their opinion, derashot cannot generate new laws at all.

At the end of the last quotation, Ralbag, and also Zecharias Frankel in his Darkhei ha-Mishnah—who likewise agreed with this approach—sought support for their position in Maimonides’ introduction to the Commentary on the Mishnah and in his remarks on the present root. But it seems that this is a mistake, though a fairly common one, in understanding Maimonides. In chapter 4 of his introduction to the Commentary on the Mishnah, Maimonides divides the laws of the Torah into five types, and the third type includes laws derived through derashah or through sevara—reasoning—which in Maimonides’ usage also includes derashot, as is clear from that introduction and elsewhere. It is therefore explicit in his words that there are laws that emerged from derashot, meaning that derashot are not only a tool for supporting preexisting laws, but also a tool capable of creating new laws.

The reason for this misunderstanding lies in Maimonides’ explanatory remarks in the introduction to the Mishnah. In the introductory remarks there, he does not refer directly to derashot that generated new laws. This created the impression for many that, in his view, derashot are only supportive. But, as noted, this is a mistaken interpretation of his words, as is clear from his summary of the types of laws in chapter 4. A full quotation appears in the introduction at the beginning of the book. In the next section we will bring further proof that this is indeed Maimonides’ view.

Proofs for Our Claim About Maimonides’ View

In his introduction to the Commentary on the Mishnah, Maimonides writes as follows:

“It follows that all the laws stated in the Mishnah are divided into these five categories. Some are interpretations received from Moses, and they have a hint in Scripture or can be learned by one of the principles. Some are halakha le-Moshe mi-Sinai. Some were learned by one of the principles, and dispute arose concerning them. Some are decrees, and some are ordinances.”

Here Maimonides divides the laws of the Torah—apart from “rabbinic law” in the ordinary sense—into three main categories:
1. Laws transmitted by tradition, which also have a scriptural grounding by means of the principles or other interpretive tools. These are laws grounded in supportive midrash.
2. Laws transmitted by tradition that have no scriptural grounding at all. These are the laws called halakha le-Moshe mi-Sinai.
3. Laws that do have scriptural grounding, but were not transmitted to us by tradition. These are newly generated laws, created by productive midrash.

The third halakhic category consists of laws created by productive derashot, and from this it is proven that, according to Maimonides, the methods of derash also have the power to create new laws. This is also explicit in his remarks on our root, as will be explained below. Likewise in Hilkhot Mamrim 2:1, where he explains that every court in every generation may innovate laws, and may even disagree with an earlier court, provided it is greater in wisdom and number, by means of reasoning or by derivations through the thirteen principles.³¹ It also seems that this follows from all the Talmudic discussions in which we find disputes about some derivation. To see this, let us briefly examine the possible types of appearances of disputes concerning derashot.

For example, in Babylonian Talmud, Berakhot 4b, amoraim dispute whether one must place the blessing of redemption immediately adjacent to the Amidah in the evening prayer. One possible explanation suggested in the Gemara is that the two sides disagree over how to expound the same verse: “when you lie down and when you rise.” Rabbi Yohanan links “lying down,” that is, evening, to “rising,” that is, morning, and learns from the rule that the blessing of redemption is adjacent to prayer in the morning that it must also be adjacent to prayer in the evening. Rabbi Yehoshua ben Levi, by contrast, links the same verse in the opposite direction: just as in the morning the recitation of the Shema is the element closer to the bed—because it is recited before the Amidah, and thus closer to rising from bed—so too in the evening the Shema should be recited closer to the bed, that is, after the Amidah, closer to lying down. Now let us ask ourselves: from where does the law requiring redemption to be adjacent to prayer emerge? If it were a received tradition, then according to Maimonides no dispute could have arisen about it. Note that the dispute here is not only about the derivation that serves as the source of the law under discussion; they also disagree about the law itself. It follows that this law was created from Rabbi Yohanan’s derivation, whereas Rabbi Yehoshua ben Levi disputes it by means of a different interpretation of the verse. If so, this is apparently a productive derash.

Another type that characterizes many disputes is where the sages do not draw opposite conclusions from the same verse, as in the previous example, but merely different conclusions. But the analysis above still stands.

A third type of dispute revolves around the question whether to accept the law that emerges from the derivation under discussion or not. In such disputes, again one must ask: if the law had been received by tradition, then the amora who disputes it ought to have accepted the tradition held by his colleague, rather than disputing him. For Maimonides states explicitly in his introduction to the Mishnah:

“The first category consists of interpretations received from Moses, and they have a hint in Scripture, and it is possible to derive them by way of reasoning—or derash, as we noted above—and in this there is no dispute. But when one says, ‘This is what I received,’ there is nothing to say against him.”

There is, however, another type of dispute, in which there is an agreed law, and the sages disagree only about the reasoning or the derash that serves as its source. Here it seems that we are dealing with supportive derash, because the law is agreed upon and therefore is probably known, and the derivations come only to anchor it in Scripture. So Maimonides writes there in his introduction, and he brings several examples, such as the different derivations brought to prove that the “fruit of a beautiful tree” in the Torah refers to the citron; see Babylonian Talmud, Sukkah 35a, and others.³²

Thus Maimonides’ view stands in a clear and consistent opposition to the opinions cited above: midrash can also create new laws, not only support existing ones. This is also the accepted view among most commentators, and among the great majority of contemporary scholars. As an example, let us cite Nahmanides, who writes in his objections to this root, in explaining why a person may not derive a gezerah shavah on his own:

“For to them all the principles are like something explicit in the Torah, and they expound them by their own understanding. But with gezerah shavah they required explicit tradition, because it is something with which a person could expound all day and overturn all the laws of the Torah…”

Nahmanides writes explicitly that a person can reason independently with all the principles except gezerah shavah; many others wrote the same. See, for example, the Encyclopaedia Talmudit, entry “Gezerah Shavah.” It is forced to explain that he means only supportive derivations. If he meant only those, then what would distinguish them from gezerah shavah? And why the concern that the matter could lead to mistakes? Does he mean that one might make mistakes only in attaching laws to Scripture? What would be the problem with that? Would that not still help memory?

In fact, this can also be shown rather clearly from several places in the Talmud. A prominent example, cited by many, appears in Babylonian Talmud, Shabbat 64b:

“As it was taught: ‘And she shall remain in her menstrual impurity.’ The early elders said: she may not apply eye paint, nor rouge, nor adorn herself with colored garments, until Rabbi Akiva came and taught: If so, you make her repulsive to her husband, and the result will be that her husband divorces her. Rather, what does Scripture mean by ‘And she shall remain in her menstrual impurity’? She remains in her state of impurity until she comes into the water.”

Rabbi Akiva acts against what had been accepted in earlier generations, and he does so on the basis of his own reasoning. He expounds the verse “And she shall remain in her menstrual impurity” differently, in opposition to the law that had previously been accepted.³³ There are many other examples of this in rabbinic literature. The very fact that many discussions in the Talmud engage in extended give-and-take about what each sage learns from every word in the verses, and what he does with each superfluous word, proves quite clearly that these are not merely mnemonic aids. It also makes it fairly clear that we are not dealing merely with supportive derivations.

The Problem: The Ambiguity of the Hermeneutical Methods

If so, the words of the group of medieval authorities who hold that derashot are nothing more than mnemonic supports are puzzling. Why, in fact, do they choose such a problematic position? As we have seen, the basis of their view lies in understanding the methods of derash as lacking logic and consistency, and perhaps also as a means of defense against attacks—whether from Karaites or from modern scholars. But we have seen that this claim does not withstand the factual test. Rabbinic literature shows fairly clearly that there are indeed productive derashot as well.

On the other hand, as we have seen, there is indeed a problem with Maimonides’ view: how can one rely on such ambiguous methods to create new laws? Seemingly, one could thereby alter the entire Torah and turn it upside down. Yet against that background it is important to emphasize that Ralbag’s view, and that of his camp, is not free of problems either: what is the point of producing derivations merely to support laws that are already known to us? Especially when the support is so weak that it is difficult to see what purpose is served by constructing such support. It seems that such tortuous methods do not really aid memory, and in general it is a great stretch to say that all of them were made solely as memory aids and nothing more. Some claim that the purpose of the midrashim is to show the rootedness of the Oral Law in the Torah itself, for apologetic and other reasons, but it is hard to see how such methods could persuade anyone that the accepted law is indeed rooted in the Torah itself. Midrashim of that kind would only intensify the question mark over the connection between the laws and the text.

In Maimonides’ opinion the problem becomes even more severe, because he maintains that there are derivations of both kinds: both productive derivations and supportive derivations. That means that, on his view, both difficulties arise together: regarding supportive derivations, it is unclear why one should perform them at all; and regarding productive derivations, the difficulty is how one can rely on laws produced in ways that are so far from unambiguous.

The root of both difficulties lies in the same point: we assume that there is ambiguity in the use of the hermeneutical principles. But in the passage from the Kuzari quoted above, one can see a hint of a possible solution to both problems. There, the Kuzari raises the possibility that in the past the sages possessed a systematic, clear, and unambiguous way of deriving expositions. He explains that the fact that the methods of derash now appear to us ambiguous and non-univocal is rooted in the process of forgetting that we described above.³⁴ Therefore, when that ability was still in the hands of the sages, they could produce productive derivations and rely on the laws thus generated. In such a situation there is certainly also reason to perform supportive derivations, because the tools are reliable and the result genuinely connects the law under discussion to Scripture—and perhaps also helps memory. True, by the time of the redaction of the Talmud the way of using the tools of derash had already been lost to us, and therefore they now appear amorphous and unreliable. This claim explains both difficulties: both the difficulty of why supportive derivations should be done at all, and the difficulty concerning the reliability of productive derivations. Below we will see further benefits that arise from performing supportive derivations.

Thus, even from the standpoint of pure reasoning, Maimonides’ view and that of most medieval authorities seems firmer, because they apparently understood the methods of derash to be univocal, and therefore it was entirely possible both to support existing laws by means of them and to create new laws. specifically the method of Ralbag and his camp, which assumes that the methods of derash are not univocal—as indeed they appear to us today—remains puzzling: if these methods really are so ambiguous and non-univocal, then there is no point at all in using them, for the reasons just given. So too from the actual evidence of rabbinic literature, which quite clearly shows that derashot can also create new laws. In the final analysis, Maimonides’ view appears to be primary.

The Halakhic Significance of Supportive Derash

When one examines the issue of supportive derivations, several directions immediately emerge that give important halakhic meaning to supportive derash. We will now try to clarify this through four main lines of explanation:

1. A productive significance that supportive derivations have for other halakhic and biblical contexts.
2. The dissolution of the dichotomy between productive derash and supportive derash.
3. The conferral of de-oraita status—that is, the status of biblical law—on the laws being supported. This direction exists only in Maimonides’ view, which regards a law that has a supportive derivation as a biblical law, and a law based on a productive derivation as rabbinic.
4. A non-halakhic significance for supportive derivations: even if there is no direct halakhic consequence, the principles themselves, and their use, teach us how to interpret Scripture, and perhaps even how to think in broader domains.

1. The Productive Significance of Supportive Derash³⁵

Suppose that a given derivation is entirely supportive: the law was received by tradition and is already known to us, and the derivation is meant only to anchor it in Scripture. I would like to argue that even in such a case there remains halakhic significance—and not merely the fulfillment of the commandment of Torah study—in the fact that we have driven an interpretive stake into some biblical word. This is true on several main planes:

A. First, the law may already have been known to us, but its halakhic status may not have been clear. If we find it an anchor by means of the tools of derash, then it has the status of biblical law; if we find no such anchor, it is probably rabbinic law. If so, there is a productive dimension here from the halakhic point of view, though it can still be defined as supportive derash.

B. A second direction that gives supportive derash halakhic significance arises when the derivation confirms a line of reasoning or a principle that can be applied in other contexts as well.³⁶ It can be shown that many times a halakhic or aggadic derivation does not stand alone. There is a rationale that leads to it and supports its conclusions, and that rationale forms an integral part of the derivation itself. In such situations, the derivation confirms the rationale itself, something that has many consequences for applying the same rationale in other contexts. Thus, anchoring a given law in a rationale may validate the rationale itself and give it force, and from there one may derive entirely new halakhic consequences—that is, create new laws.

C. A clear and unique example of a process similar to the one described in the previous subsection is the principle of gezerah shavah.³⁷ The Gemara, Babylonian Talmud, Kritot 22b, states that gezerah shavah cannot be applied “by halves.” Let us suppose, then, that we know that a woman can become betrothed by money, and that we find a scriptural anchor for this by means of the gezerah shavah of “taking” and “taking”; see Babylonian Talmud, Kiddushin 2b. This may indeed be a supportive derivation. Nevertheless, we must now take into account that gezerah shavah cannot be applied partially; see Babylonian Talmud, Kritot 22b. That is to say: if two halakhic contexts are compared to one another, we must compare them in every respect. Thus the supportive derivation becomes a basis for deriving additional laws through parallel productive derivations.³⁸

D. Another example of the mechanism by which supportive derivations acquire halakhic significance and importance is the principle that the location of the interpretive anchor in Scripture may significantly determine the contours of the law under discussion. For example, if we learn that a woman can become betrothed by money from the gezerah shavah of “taking” and “taking” from a field, then it may follow that we must understand betrothal as a kind of acquisition, like the purchase of a field. Such a conception has many halakhic and philosophical implications. But if the interpretive source for this law is a different one, then a completely different halakhic and philosophical conception may emerge from it.³⁹

E. There is yet another direction, different in kind, that gives the act of supportive derash halakhic significance, even where there is full certainty about the status and contours of the law under discussion. Let us take as an example, one among thousands, the derivation in Babylonian Talmud, Berakhot 9a. There Rabbi Eleazar ben Azariah and Rabbi Akiva dispute the end-time for eating the Passover offering:

“As it was taught: ‘And they shall eat the flesh on this night.’ Rabbi Eleazar ben Azariah says: It says here, ‘on this night,’ and elsewhere it says, ‘And I will pass through the land of Egypt on this night.’ Just as there it means until midnight, so here it means until midnight. Rabbi Akiva said to him: But does it not already say, ‘in haste’—until the time of haste, that is, dawn? If so, what does Scripture mean by ‘at night’? One might have thought that it may be eaten like consecrated offerings by day; therefore Scripture says, ‘at night’—it is eaten at night and not by day.

“Granted, according to Rabbi Eleazar ben Azariah, who has a gezerah shavah, Scripture had to write ‘this.’ But according to Rabbi Akiva, what does he do with this word ‘this’? It comes to exclude the following night.”

Rabbi Eleazar ben Azariah derives from the words “on this night,” by means of a gezerah shavah, that the Passover offering may be eaten only until midnight. Rabbi Akiva derives from the word “in haste” that it may be eaten until dawn. Now the word “night” remains superfluous. Let us suppose that Rabbi Akiva’s derivation is supportive rather than productive. Even so, since this is his derivation, the word “night” is now free for additional derivations, and new laws can be learned from it by means of productive derash. By contrast, if Rabbi Eleazar ben Azariah offers a different derivation, then the word “night” is not free for another derivation. He will therefore dispute the law that would emerge from the productive derash Rabbi Akiva can make from that word.

We thus have before us a result in the form of a new law, whose creation depends on a supportive derivation. Even if we have performed a supportive derivation, the result is that certain words in Scripture have now been clarified as “occupied,” and other words are clarified as free. Other productive derivations that depend on those words can therefore either arise or be blocked. The conclusion is that our derivation can have a direct result with respect to several new laws, even though the derivation itself is supportive. In light of this, it actually appears very important to attach all laws to Scripture by means of supportive derivations, if possible, if only in order to know what role each word in Scripture plays on the plane of derash, and from which words one can still derive and create new laws.

We should note that this is indeed a typical and very common situation in the Talmud, but in the particular example before us the picture is somewhat different. Rabbi Akiva learns from the word “night” in order to reject an opposing rationale, but he does not generate a new law from it. That word teaches us that the meat of the Passover offering is eaten at night, and only afterward can there be debate whether the time limit is midnight or morning. In other words, the word “night” serves as a condition for the supportive discussion—we are assuming here, for the sake of the example alone, that it is supportive—and not as the source of a new productive derivation. In fact, this pattern continues throughout their discussion there. They are not disputing laws, only sources, and therefore the discussion there could indeed fit a wholly supportive model.⁴⁰

All this was said about a type of Talmudic discussion that deals with clarifying the sources of a given law—for example, when the Gemara asks from where one sage derives the law of another sage. But when the Gemara deals with clarifying the laws that emerge from given sources—for example, when the Gemara asks, within the give-and-take, what sage A does with a word that sage B used for a certain purpose, rather than from where he derives sage B’s law, which is a different type of Talmudic inquiry—then sage A usually answers that he learns from it another, new law. In such contexts one certainly sees a situation in which a disagreement over a supportive derivation can generate a new law, by means of a productive derash on the word that, according to sage A, has become free.

As an example of such a move—again, one among very many—and of its relation to the earlier mechanisms, let us consider the give-and-take in Babylonian Talmud, Bava Metzia 27a, where a tannaitic dispute appears in a baraita, and Abaye and Rava disagree about it:

“The sages taught: ‘That which is lost’—this excludes a lost object worth less than a perutah, the smallest coin. Rabbi Yehudah says: ‘And you found it’—this excludes a lost object worth less than a perutah. What difference is there between them? Abaye said: They differ only about which expression is to be expounded. One derives it from ‘that which is lost,’ and the other from ‘and you found it.’ According to the one who derives it from ‘that which is lost,’ what does he do with ‘and you found it’? He needs it for the teaching of Rabbanai, for Rabbanai said: ‘And you found it’ implies that it has come into his hand. And according to the one who derives it from ‘and you found it,’ what does he do with ‘that which is lost’? He needs it for the teaching of Rabbi Yohanan. For Rabbi Yohanan said in the name of Rabbi Shimon ben Yohai: From where do we know that a lost object swept away by a river is permitted? For it is said: ‘So shall you do with every lost thing of your brother that has been lost from him and you have found it’—something that is lost from him but available to everyone; this excludes something that is lost from him and not available to everyone. And the other, from where does he derive Rabbanai’s teaching? He derives it from ‘and you found it.’ And the other, from where does he derive Rabbi Yohanan’s teaching? He derives it from ‘from him.’ And the other: ‘from him’ does not imply that to him. Rava said: The practical difference between them concerns a perutah that decreased in value. According to the one who derives it from ‘that which is lost,’ there is a practical difference; according to the one who derives it from ‘and you found it,’ there is not. But according to the one who derives it from ‘that which is lost,’ do we not also need ‘and you found it,’ which is not present? Rather, the practical difference concerns a perutah that increased in value. According to the one who derives it from ‘and you found it,’ there is a practical difference; according to the one who derives it from ‘that which is lost,’ there is not. But according to the one who derives it from ‘and you found it,’ do we not also need ‘that which is lost,’ which is not present? Rather, the practical difference concerns a perutah that increased in value, then decreased, and then increased again. According to the one who derives it from ‘that which is lost,’ there is a practical difference; according to the one who derives it from ‘and you found it,’ we require that it have the measure of a found object from the time of loss until the time of finding.”

The Gemara opens with a dispute between Rabbi Yehudah and the sages concerning the source for the exemption from returning a lost object worth less than a perutah. The law itself is agreed upon, and perhaps was even transmitted by Sinaitic tradition. The dispute concerns only its source. That is indeed how Abaye understands it: the dispute concerns only the source—“they expound meaning”—and has no halakhic consequence. This is one of the cases that is clearly supportive derash.⁴¹ And yet, immediately afterward, the Gemara asks what each side does with the superfluous words, and from those words new laws emerge. In other words, there are indirect halakhic results even to supportive derash: laws created by productive derivations operating on the words left over.⁴²

So far we have understood the tannaitic dispute according to Abaye. But Rava disagrees with Abaye and holds that the tannaitic dispute has direct halakhic significance: it concerns a perutah that increased in value, then decreased, and then increased again. From where does this practical difference emerge? From the context of the derivations. That is to say, the precise source of the supportive derivation—though it remains supportive—determines the contours of the law under discussion: must the lost object be worth a perutah only at the time it was lost, or must this continue until the time it is found? This is an illustration of the mechanism we presented above in subsection D. According to both Rava and Abaye, supportive derash has newly emergent halakhic consequences, albeit through different mechanisms.

Summary of the first possibility. Everything said so far shows that supportive derivations possess very great halakhic significance. In fact, it seems from our discussion that there is an obligation to anchor every law in Scripture, where this can be done, for otherwise we may err and derive something from a word that is not really free, because it actually serves as the anchor for a derivation we failed to perform; or we may rely on a rationale that is not correct; or assign mistaken contours to a law transmitted to us by tradition; or lose a correct rationale that we could have extracted and established from the derivation.

2. Dissolving the Dichotomy Between Productive and Supportive Derash⁴³

Up to this point we have assumed that the distinction between productive derash and supportive derash is sharp and clear. But, as we shall now see, that is not so. This point can clarify further the value of supportive derivations, and can also remove many puzzling obscurities in various discussions. Here I will present it briefly, only insofar as needed for our present topic.

Let us begin with a preface concerning the principle of gezerah shavah.⁴⁴ The Talmud, Babylonian Talmud, Pesachim 66a and parallels, states that although a person may formulate a kal va-homer on his own, a person may not formulate a gezerah shavah on his own. The usual explanation is that a gezerah shavah derivation always comes by tradition from Mount Sinai; see, for example, Rashi there, and on Babylonian Talmud, Sukkah 11b, s.v. “We do not learn,” and Nahmanides in his objections to our root, whose words we cited above. We already noted above, in the name of Kinat Soferim on our root, that according to Maimonides’ position here—namely, that the halakhic status of laws derived by means of the thirteen principles is that of rabbinic law—apparently even gezerah shavah derivations were not given at Sinai. When the sages said that a person may not derive a gezerah shavah on his own, they meant that only the Great Court may derive by gezerah shavah, not necessarily that the derivation must be received by tradition from Moses our master at Mount Sinai.⁴⁵

Many have already pointed out that according to Rashi and those who follow him—which is an overwhelming majority of the medieval authorities—it is very difficult to understand several phenomena found in the Talmud: the fact that there are disputes regarding gezerah shavah; see Babylonian Talmud, Hullin 88a. We also find people being praised for discovering the source of a law by gezerah shavah; see Babylonian Talmud, Shevuot 7a. It has also been noted—see Nahmanides in his objections to our root, among others—that if everything was indeed transmitted from Sinai, it becomes very difficult to understand why gezerah shavah is called a “principle by which the Torah is expounded,” since there would be no room at all for the sages’ own derivations.

For this reason, many medieval authorities wrote—see, for example, Nahmanides in his objections to our root, pp. 60–64—that this cannot mean that the gezerah shavah was given at Sinai in its full concrete form; see also Halikhot Olam, gate 4, and others. Sometimes a certain law was transmitted to Moses, together with the information that it is learned by gezerah shavah, but he was not told in what exact form or from where. Sometimes the source was transmitted, but not the law. Sometimes the hermeneutical principle was transmitted, but not the precise word, and so forth.

There can thus be various distributions between the components transmitted by tradition and the components renewed by the later interpreters. A standard derivation is composed of three elements: the scriptural sources, the hermeneutical principle, and the law derived from it. For each of these we must decide whether it belongs to the part given by tradition, or to the part added by the interpreter.

There is also room here to take into account the identity of the principle under discussion. As we have seen, gezerah shavah is unique in that a person may not derive it on his own. By contrast, kal va-homer a person may derive on his own, although it too can be received by tradition. As for the other principles, the medieval authorities disputed this, as noted above, and therefore one must ask whether in their case as well different relations are possible between the elements received by tradition and those introduced by the interpreter.

For our purposes, it is very important to notice that this picture greatly qualifies the sharp dichotomies involved in the common distinction between productive midrash and supportive midrash. There are very many derivations in which one component is productive, but other components are given to us by tradition. If the law itself is given by tradition, the derivation is supportive. If the law is not given, the derivation is productive—even though at times it produces the new law with the aid of other data that were indeed transmitted, such as the sources and-or the hermeneutical principle. It is even possible that the law itself was transmitted only partially—for example, we were told that there is a comparison between field and woman regarding betrothal, but were not told exactly what the contours of that comparison are.

If so, the productive element in productive derash is not free-floating, but guided by components received through tradition. Some of its elements are “supportive” in essence, while others are creative in essence. Such a derivation is partly productive and partly supportive, and it is difficult to classify it sharply into one pole or the other. There is no doubt that it is very important to carry out such a derivation, since it contains a great deal of new information, especially in light of the range of aspects we presented in the previous section.

It should be noted that this division of derivations into elements received by tradition and elements renewed by the interpreters can also somewhat allay the concerns raised above by the Kuzari and by Ralbag regarding the ambiguity of the methods of derash. Where there is a danger of error in the use of these methods, the Holy One transmits to us certain components through tradition in order to direct us toward the right path and prevent possible mistakes. This is a mechanism that can explain some of the trust that Maimonides and those who follow him place in the methods of derash, and that underlies the possibility of trusting new laws created through productive derivations, as well as the various forms of significance that supportive derivations possess.

3. Conferring the Status of Biblical Law on Supported Halakhot

Another way to see significance in supportive derash emerges in light of what we will later see in Maimonides’ own view. According to Maimonides, laws transmitted by tradition to Moses at Sinai that also have a midrashic support in Scripture receive the force of biblical laws. The logic of this will be explained later. According to this, it is clear that there is direct halakhic significance to the fact that there is a derivation supporting laws that were transmitted to us by tradition.

To be sure, this is a halakhic consequence only if we interpret Maimonides, as some commentators do, to mean by the term “biblical” not merely a description of the source of those laws, but their operative halakhic status. For further detail on this, see below.

4. A Non-Halakhic Significance of Supportive Derash⁴⁶

This is the fourth and final path we will present here for understanding the benefit of performing supportive derivations for known laws. Sometimes the derivation is meant to teach us something that arises from the very method of derash itself, and not necessarily from the direct halakhic results of the derivation. In section 1B, we dealt with a similar principle, according to which the rationale learned from a derivation has halakhic implications in other contexts. Here we will broaden that principle, and show that the methods of derash themselves, in general, can have halakhic implications, similar to what we saw there, and also highly important non-halakhic implications.

By way of preface, one can understand the logical place and significance of the system of principles by which the Torah is expounded in three basic ways:

1. The principles are a kind of axiomatic system, or an interpretive code.
2. The principles are an alternative logic in place of conventional logic.
3. The principles are a foundation of principles with philosophical significance.

Let us explain this briefly.

A. An axiomatic system, an interpretive code. Sometimes a person sends his friend a text encrypted in a certain code. In order to decipher the message, the reader must possess a “key.” A well-known example is the Atbash cipher, which is composed of a system of twenty-two decoding rules telling us how to “translate” each letter in the encrypted text into our language. The rules of the cipher are arbitrary, and therefore have no significance in themselves. The same text could have been encrypted by different means, and there is no importance to the particular method chosen.

If we understand the system of principles as a code that enables us to extract laws hidden in the Torah, then we will necessarily regard this system as arbitrary. The Holy One gave Moses at Sinai a key by which we can “decode” the midrashic layer of the Torah. According to this, there is no point in seeking any significance in the principles themselves. They are simply additional, arbitrary interpretive methods by means of which the Holy One encoded further information beyond the plain sense⁴⁷ in the Torah, and then gave us the key by which we can decode that layer and extract the hidden information contained in it.

Such an approach is suggested—though not necessarily compelled—by Maimonides in Millot ha-Higgayon, at the end of gate 7, where he writes:

“We have other inferences, called juridical inferences, and there is no place to mention them in what we are now pursuing.”

That is, the system of principles does not belong to the domain of formal logic. These are arbitrary rules, like keys or a cipher, relevant only within the study of Torah, and there is no place to discuss them in the context of the laws of logic.

Let us note that according to this conception of the system of hermeneutical principles, it follows clearly that we are dealing with a one-to-one, unambiguous system. Since there is no logic underlying it, it must be defined completely and with mathematical certainty. Without that, one certainly could not rely on the interpreter to derive the correct conclusions from it.

B. Alternative logic. There are those—mainly thinkers of our own time—who understand the matter differently. In their view, the system of principles is not arbitrary at all. Rather, the Torah operates on the basis of a deeper, prophetic logic, which is different from conventional logic. The words of the Nazir, who in several places writes as though the system of principles constitutes a special Hebrew logic—“auditory,” in his terminology—as opposed to Greek “visual” logic, seem at first glance to express such positions and to inspire them.⁴⁸

But this approach, taken literally, is problematic, if not impossible. It is not plausible that we are required—or even able—to abandon the modes of thought of classical logic when we study Torah. Classical logic is imposed upon us, as it is upon all mortals. Imagine that you were commanded to abandon the law of non-contradiction—the law that states that a proposition and its negation cannot both be true at the same time. In that case, when the Torah says that it is forbidden to marry the co-wife of one’s daughter, we could also infer that it is permitted to marry her, because on this approach the law of non-contradiction would not apply within halakhic discourse. This is of course impossible to say. The conclusion is that these statements of the Nazir cannot be understood literally. In practice, we do not operate this way, and cannot operate this way, even in Torah study.

C. A foundation of principles with philosophical significance. In the writings of the Nazir there are apparent contradictions on this issue, and in several places a third, mediating approach emerges; it seems likely that this is his fundamental view, and also the view of the commentators on the principles whom he himself cites.⁴⁹

Greek logic rests on necessary inference, that is, deduction. According to the Nazir, Hebrew logic—the principles by which the Torah is expounded—adds to this the methods of analogy. It does not come to replace scholastic logic, but to supplement it, and in fact to spell out, in certain ways, additional forms of thought: analogy and induction.

These matters require clarification. As noted, it is not plausible that the Nazir means to reject deductive logic, since it is imposed upon us. On the other hand, surely no one would dispute that there are non-deductive methods of thought that one can also use. If so, it is not clear what exactly is meant by the claim about an “alternative logic.”⁵⁰

It seems to me that the difference lies in the question whether this addition is itself treated as “logic.” In the course of our thinking we use many modes of thought, but not all of them are granted the status of “logic.” For example, within the discipline called “logic,” one generally deals mainly with deductive and mathematical modes of thinking. Analogy and induction are not usually regarded as belonging to that field.⁵¹

This difference is expressed in the degree of trust we place in the results of non-deductive inferences. As noted, even the “Greek” thinker uses these modes of inference, but he does not fully trust them; at times the attitude is that they yield only subjective conclusions—some would call this “emotion.”⁵² Hebrew logic grants full trust to these forms of inference, since we punish, and even execute, on the basis of laws derived from the principles by which the Torah is expounded, aside from a few exceptional cases.

In the terminology of the present quartet, we may say that those who hold an analytic position do in fact use non-deductive inferences, but they treat them as subjective and non-binding. At most, they justify them after the fact out of sheer necessity. Those who hold a synthetic position, by contrast, think that analogy and induction are the basic tools for accumulating information, and that information acquired through them can be treated as reliable, even if not certain. We are dealing here not with mere practical compulsion but with an additional logical tool for gathering information.

Accordingly, as the Nazir argues at length in the fourth essay of his book, the principles by which the Torah is expounded spell out and define the rules of analogy, thereby turning these forms of inference into “logic” in the full sense of the term. The essence of his innovation is that the non-mathematical character of the system of principles does not necessarily imply that it is ambiguous or vague.⁵³ In the first volume we pointed out that the Nazir sees in the hermeneutical principles the basic rules of synthetic thinking.

Against this background one can also see the difference between this position and that of Ralbag. As we saw, Ralbag argued that these forms of inference can be used in various ways, and therefore they are not univocal. Ralbag, like most of the medieval thinkers, adopts a “Greek,” or in our terminology “analytic,” approach on this point. He maintains that non-deductive modes of inference cannot have a definite meaning, and therefore he is unwilling to believe that one can create new laws by means of them. The Nazir, by contrast, sees them as another kind of logic, in addition to—and not in place of—deductive logic.⁵⁴ Deductive logic describes the logical tools of analytic inference, whereas auditory logic describes the tools of analogical and inductive inference, that is, synthetic inference. As we saw in the previous three books, human thought requires the use of both types of tools. It follows that the hermeneutical principles are the foundation of synthetic logic.

Against this background we may add that the vagueness that characterizes the principles in the eyes of the modern observer also derives from the influence that “Greek,” that is, analytic, thinking exerts within the study hall. Someone who uses auditory logic—which in the Nazir’s view is a mode of prophecy—is willing to grant high reliability even to analogical and inductive arguments. This of course requires an “auditory” capacity, which, at least in the Torah realm, can develop only through Torah study.⁵⁵

Let us emphasize that according to this third approach, and only according to it, there is room to learn philosophical principles from the principles themselves, because they are not arbitrary rules, like a cipher, but principles that have significance in themselves. As far as I can tell, the Nazir’s book Kol ha-Nevu’ah is devoted entirely to spelling out and clarifying this point: Torah-halakhic logic is indeed “logic,” and it also provides the philosophical basis on which a Torah understanding of the world—that is, Jewish philosophy—must be built. Just as Aristotle placed the Organon, his book of logic, at the head of his philosophy, so we must place at the head of our philosophy the logic of the hermeneutical principles, alongside Greek logic and not in its place.

Let us now return to our present topic, namely, the implications of supportive derash. What we gain from using the hermeneutical principles, even when they support existing laws, is the development of synthetic capacities—analogical and inductive thinking. The use of the methods of derash sharpens those capacities. There is here a kind of school for analogy, or for synthetic thought. One should notice that, unlike the halakhic implications discussed above, these results are achieved precisely when the derash is supportive, not when it is productive. In such situations the law itself, which has been transmitted to us by tradition from Sinai, can provide us with indications, by way of feedback, that tell us whether we have used the methods of derash correctly, and with them the methods of analogical and inductive thought. In productive derash we have no such indication, because we cannot know independently whether our conclusions are correct. The law transmitted to us from Sinai therefore serves as a kind of feedback on our use of synthetic modes of inference. Such feedback builds our capacities in these areas—some might call it our “Torah intuition.”

These capacities have halakhic implications, because the interpretation of Torah is based entirely on such capacities; see the language of Nahmanides in the introduction to Milhamot Hashem: “For the wisdom of our Torah is not like astronomy and mathematics, whose proofs are sharp and cut…” One should notice that the implications can be framed in terms of the law itself, or the rationale, that we extracted in a specific derivation, but that belongs more properly to section 1B above. Here our concern is with the implications that arise from the mere fact that we possess a more developed analogical capacity. The application of that capacity yields halakhic consequences throughout the Torah, and not necessarily only as a direct result of the specific law whose scriptural anchoring we were discussing.

But the development of these capacities also has other implications. A person endowed with more developed analogical and inductive abilities can reach important conclusions in other fields of thought as well, such as science and others. Today we have no explicit, detailed map of the methods of analogy and induction, because, as noted, the discipline of logic barely deals with them. It is a non-mathematical map, because these methods cannot be charted precisely in the way we customarily chart methods of a mathematical character. Hence the apparently blurred and non-univocal form of the methods of derash. But someone who becomes skilled in their use can employ them reliably, and can draw conclusions from them in a fairly determinate way—at least no less determinate than the way of plain-sense interpretation, which is certainly not pure deductive logic either.⁵⁶

To conclude this section, let us simply note that the Nazir sees in auditory logic, as it appears in the principles by which the Torah is expounded, a prophetic logic that will become the focal point of the renewal of prophecy in Israel. In a state in which prophecy dwells among us, we will be able to test inductive and analogical inference with “precise” tools, and we will have certainty in the conclusions of those inferences, just as in the deductive case. At that time the distinction between emotion—which is not part of the intellect—and intuition—which is part of the intellect—will become sharpened, whereas today it appears blurred and elusive.⁵⁷ Then the fog that hovers over the hermeneutical principles will also disperse.

Footnotes