Re’eh (5765)

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

---

With God’s help — Midah Tovah — Eve of the holy Sabbath, Parashat Re’eh, 5766

Questions

1. Does science deal with concrete objects?
2. Does halakha (Jewish law) deal with concrete situations?
3. Between theoretical analysis and application.
4. On the relation between rules and particulars in halakha and science.
5. The degeneration of concepts and the problem of generalization.
6. What is the philosophical basis for our trust in the tools of generalization, and in reasoning in general?

The Hermeneutical Principles

• A general clause that requires a specific clause
• A specific clause that requires a general clause
• General and specific
• Specific and general

“Sanctify to Me every firstborn, the first issue of every womb among the children of Israel, among man and among beast; it is Mine.”

— Exodus 13:2

“You shall set apart to the Lord every first issue of the womb, and every firstling that comes from the offspring of an animal that you have; the males belong to the Lord.”

— Exodus 13:12

“Every firstborn male that is born in your herd and in your flock you shall sanctify to the Lord your God; you shall not work with the firstborn of your ox, nor shear the firstborn of your flock.”

— Deuteronomy 15:19

“As an example of a general clause that requires a specific clause, and a specific clause that requires a general clause: ‘Sanctify to Me every firstborn.’ One might think that even a female is included; therefore Scripture says, ‘male.’ If it says ‘male,’ one might think this includes even one preceded by a female birth; therefore Scripture says, ‘the first issue of the womb.’ If it says ‘the first issue of the womb,’ one might think this includes even one born after a Caesarean birth; therefore Scripture says, ‘firstborn.’ This is a general clause that requires a specific clause, and a specific clause that requires a general clause.”

— Baraita of examples at the beginning of Sifra; Babylonian Talmud, Bekhorot 19a

“‘Sanctify to Me … every firstborn.’ This is a general clause that requires a specific clause, for I might have understood every firstborn, whether male or female; therefore Scripture says, ‘the males belong to the Lord.’ And similarly it says, ‘Every firstborn male that is born in your herd and in your flock’ (Deuteronomy 15:19). The specific clause came and taught regarding the general clause: male and not female. And the specific clause requires the general clause, lest you say: every male, even one that is not the first issue of the womb; therefore Scripture says: ‘every firstborn, the first issue of the womb.'”

— Pesikta Zutrata (Lekach Tov), chapter 13, under “Sanctify to Me”

A. Summary of Last Year’s Article

The principle of “a general clause that requires a specific clause, and a specific clause that requires a general clause” appears only twice: in the talmudic discussions in Babylonian Talmud, Bekhorot 19a and Babylonian Talmud, Hullin 88b. In last year’s article we dealt with the canonical example of this principle, cited above. We argued there that these principles function as limitations on the ordinary principles of general and specific. Where the specific clause requires the general clause, we do not interpret the text by the rule of “specific and general,” and where the general clause requires the specific clause, we do not interpret it by the rule of “general and specific.” What, then, are we supposed to do in such cases? It is not even entirely clear whether these are two different principles or only one principle. In our article last year, we noted that the medieval commentators may have disagreed about this point: Rashi and Raavad.

We also discussed whether the principle of “a general clause that requires a specific clause” always limits the principle of “general and specific,” and vice versa, or whether both limitations operate only upon “general and specific.” The very small number of occurrences of this principle makes it very difficult to define it, and therefore our conclusions should be taken with considerable caution.

Rashi, in Bekhorot, discusses the principle of “a general clause that requires a specific clause” and distinguishes it from the ordinary principle of “general and specific.” He explains that the principle of “general and specific” deals with a general clause that is already well defined, and its function is to narrow that clause to a smaller group: “only what is in the specific clause.” By contrast, the principle of “a general clause that requires a specific clause” deals with a general clause whose definition is not clear. This principle does not operate by narrowing sets; rather, it chooses between two possible interpretations of the general clause. The specific clause hints to us which of the two possible definitions of the general clause we should adopt. We demonstrated this on the basis of the two occurrences of this principle in the Babylonian Talmud.

From this emerges the criterion that determines which of these principles we apply in a given biblical case: when the general clause is well defined, it does not need the specific clause. There the appearance of the specific clause is meant to narrow it. But when the general clause can be interpreted in more than one way, it needs the specific clause in order to be defined. There the specific clause shows us the correct possibility for interpreting the general clause, and is not meant to restrict it.

We then discussed the logic on which interpretations based on these principles rest. In several past articles we argued that an interpretation by the principle of “general and specific” is based on a difficulty in the text: redundancy, similar to the principle of “an item singled out from a general category,” or a shift in wording from plural to singular. That difficulty leads us to conclude that the Torah repeated itself in order to restrict the general clause to which the law applies. Interpretations based on “general and specific” deal with defining domains of application.

By contrast, when the general clause requires the specific clause, the issue of redundancy does not arise, because in that case the specific clause is not superfluous. Therefore there is no room there to interpret by the principle of “general and specific”; instead, we read the verse according to its plain sense. According to this, interpretation by these principles is not really an interpretive derivation at all, but a plain reading. The explanation is what we mentioned above in the name of Rashi in the passage in Bekhorot: the specific clause selects between two possibilities that already exist for defining the general clause. Clearly, these principles do indeed deal with defining domains of application, but not through narrowing and expanding groups, as in the usual principles of general and specific.

According to this, it is clear that these two principles are really only one. The basis of this principle is the limitation it places on interpretations of the “general and specific” or “specific and general” type. The result is a plain reading of the text, such that the specific clause defines the general clause, or chooses among several possible ways of defining it. In such a case, the order in which the general and the specific appear in the biblical text is unimportant.

At the end of our discussion, we noted that the term “general clause” may bear a different meaning in different hermeneutical contexts. In interpretations based on “general and specific,” a general clause is a broad expression. By contrast, in the context of the principle of “a general clause that requires a specific clause,” we encounter “general clauses” that do not look general at all. We suggested that in this context the term “general clause” does not point to the range of the law under discussion, but rather to the multiplicity of possible interpretations of that range.

With regard to the interpretive passage cited above, we discussed whether the verse on which it is based is the verse in our portion, or, as seems more likely, a parallel verse in the portion of Bo—although the specific term “male” does not appear at all in the verse in Bo. We also discussed the question raised by Raavad: how can we interpret a “general and specific” construction when the two are far apart from one another?

B. Generalization and Specification in Halakha and Science

Introduction

Several of the medieval commentators discuss the relation between the principle of “general and specific” and the principle of “a general clause that requires a specific clause.” We cited Rashi in the passage in Bekhorot, who explains that we do not interpret by the principle of “general and specific” when the general clause is not clear in itself, but requires clarification, or selection from among several interpretive possibilities. In such a case, we interpret by the principle of “a general clause that requires a specific clause.” See also the book Middot Aharon and the other commentators on the hermeneutical principles in their discussions of these rules.

Raavad, in his commentary on the baraita listing the hermeneutical principles, as well as the Tosafists and, followingיהם, several other medieval commentators, challenge Rashi’s position and argue that this is actually the case in every instance of “general and specific.” After all, even when we interpret by “general and specific” and restrict the general clause to what is contained in the specific clause, that too is a clarification of the general clause. Had the specific clause not been written, we would not have known the meaning of the general clause; therefore the writing of the specific clause is necessary.

We saw that this objection poses no difficulty at all for Rashi. Rashi explains that in a biblical occurrence where the general clause is unclear—meaning that there are several equally plausible ways to interpret it—the specific clause is not superfluous. Therefore it is not interpreted by the principle of “general and specific”; rather, it serves to select the correct interpretation of the general clause, and not necessarily to narrow it. By contrast, in an ordinary case of “general and specific,” the general clause is completely clear, and were it not for the specific clause there would be no difficulty at all. Yet the Torah’s intent is that we should not apply the law under discussion to the entire group included in the general clause, and therefore it added the specific clause.

The assumption throughout this discussion is that the principles of general and specific deal with general categories, and the goal of these interpretive readings is to clarify to which category the halakha under discussion applies. The principles of “general and specific” clarify this by operations on domains of application—restrictions and generalizations. By contrast, the principles of “a general clause that requires a specific clause” and “a specific clause that requires a general clause” do so by selecting among equally plausible definitions and characterizations of the categories.

What does it mean to say that a given halakha deals with general categories? What follows from that? What is the relation between the categories studied in theoretical halakha and the particulars with which practical halakha deals? How can we rely on our reasoning, which defines those categories?

These questions arise in various contexts, and we have already needed them indirectly in several earlier articles. They were also put to us, in different forms and in connection with different articles, by several of our readers. We shall therefore take this opportunity to discuss them here directly and explicitly.

The Relation Between Rules and Particulars

The question of the relation between a general rule and the particulars included within it has occupied us before as well. For example, in an article on Parashat Ha’azinu, 5765, we discussed whether general rules are real entities or merely our own fictions that describe various collections of particulars. We saw there that halakha does indeed deal with particulars, but the aim of theoretical study is specifically the general rules. At the same time, we noted that thinking in terms of rules can lead to errors, precisely because in its essence halakha deals with particulars rather than with rules.

For example, the law of sanctifying the firstborn was stated regarding each individual firstborn. There is an obligation to sanctify Reuven and Shimon, and so on. In order to shorten and streamline our thinking, we formulate this as a general rule: every entity that is a firstborn must be sanctified. The subject of this instruction is a certain totality, not an individual case. But the law itself deals with the particulars included within that totality.

In the article on Parashat Korach, 5765, and also in the article on Parashat Vayikra, 5765, we discussed two ways of defining domains of application: definition by extension and definition by intension. Even if we assume that halakha deals with general rules rather than particulars, it is quite clear that it deals with the intensional definition of the domain of application, not its extensional definition. In our example, the Torah determines that all male beings that open the womb and possess firstborn status are holy. From this one can spell out the result for each of the concrete particulars included in that group, and conclude that each of them is holy to the Lord. But this is a claim about whoever possesses the characteristic of firstborn status—the intensional definition of the group. Definition by extension is only a collective expression whose intention is the totality of the particulars.

The principles of general and specific are schemas whose function is to prevent, as far as possible, exactly those errors that arise from the relation between theoretical halakhic study and halakha itself. These schemas guide us in how to generalize correctly, and how not to be misled by definitions of classes and thereby miss the particulars, which are the fundamental purpose of halakha.

However, as we have seen more than once, the hermeneutical principles do not actually carry us all the way to the halakhic conclusion. They hint to us that we should broaden or narrow the domain of application, but they do not tell us how to narrow it, or what the final definition of that domain is. For example, an interpretation by the principle of “general and specific” tells us to narrow the group denoted by the general clause to a smaller group. But the question what exactly that smaller group is, and how we should define it, is left to the reasoning of the interpreter.

Can we really rely on our reasoning here? After all, there are infinitely many possible ways to define a narrower group from a broader one. How do we know where to stop? What is included in that group and what is not? This question has been put to us over the last two years by quite a number of readers, and it also arises and is discussed extensively in the philosophy of science. We shall therefore try to learn from that discussion for our own case.

The Gap Between Rules and Particulars in Science

Parallel theoretical fields, where apparently we have no explicit rules to help us avoid such errors, are the various sciences. The scientific enterprise’s engagement with classes and general principles assumes that such principles really exist. The assumption is that phenomena are organized into domains of application, as opposed to a situation in which each phenomenon stands on its own and has no explanation in terms of a general principle or law. The success of science indicates that despite the lack of explicit rules of this sort, scientific generalizations are in most cases not so dramatically mistaken. Evidently, human beings have such rules built into them intuitively. Those rules ought, in some way, to parallel the various principles of general and specific.

As we mentioned in the article on Parashat Bereshit, 5765, Rabbi David Cohen, known as the Nazir, argued that the system of hermeneutical principles constitutes a map, or a toolbox, of non-deductive forms of inference. If so, the system of hermeneutical principles and the system of scientific tools of generalization and inference are similar systems, and each can teach us something about the other—within certain limits, of course.

In the scientific context, there are of course no difficulties of textual redundancy. There is nothing there like the biblical case of “an item singled out from a general category,” where a general law and a specific case both appear and we need to explain why there is such duplication. Even so, the logic we described above—according to which the specific case reveals to us the nature of the general rule—certainly appears there as well.

Scientific experimentation is conducted on concrete phenomena, which are particulars. The conclusions of the scientific process are generalizations that take the form of general laws, and these deal with classes of phenomena. The particular case measured in the laboratory reveals to us the character of the general rule. Exactly as in halakha, the aim of concrete science is the concrete phenomena. It is supposed to explain to us and predict what will happen in a concrete situation. Yet scientific theory is more interested in general questions—that is, in questions concerning classes and not the particulars included within them.

And from here the problematic nature of scientific generalization also arises. In one direction: the scientist examines particular cases in the laboratory and tries to generalize the results into general laws. This is the first problem: what guarantees that the generalizations are correct? There are many other possible ways to generalize the particular facts observed in the laboratory. The second problem lies in the opposite direction: once we have discovered the scientific law, we apply it to a concrete situation. But how do we know that concrete situations behave exactly as the general law dictates? The general laws of science deal with classes of phenomena, not with concrete phenomena. Yet the conclusions we draw from them concern a concrete and specific situation, and usually such a situation has additional characteristics that are not addressed by the general law.

The claim that theoretical science deals with a class of phenomena assumes that phenomena are indeed divided into classes and do not behave in wholly idiosyncratic ways. In other words, it assumes that one may use scientific generalizations that rest on relevant characteristics and ignore incidental characteristics that are irrelevant to the phenomenon under discussion.

Let us now try to define this somewhat more precisely.¹

Information and the Characteristics of Objects

Information theory deals with the quantification of information. For example, the claim that a certain person is one of the prime ministers of some democratic country contains relatively little information. If we say that he is one of the prime ministers who served in Israel, the amount of information increases. When we specify when he served as prime minister of Israel, we increase the amount of information still further. Thus, the more focused we become, the greater the amount of information contained in what we say.

If we look at a certain object that has several characteristics, we can say that the more characteristics we enumerate, the more we say about it—that is, the more information we convey. One indication of this is that the more characteristics we list, the fewer objects fit the description. Once we list enough characteristics, we are left with a domain of application containing only a single object. In such a situation, we have fully singled out the object to which we refer. But we have not necessarily defined it thereby. For example, the statement “David Ben-Gurion is the first prime minister of the State of Israel” singles him out clearly. No other person fits that description. But it does not necessarily define him fully, for he has other relevant characteristics, and some of them appear to be no less important than the characteristics we have mentioned.

A full description of any object must contain all its characteristics. An exhaustive definition of that object is the totality of its essential characteristics—those without which it would not be what it is. Its individuation is the totality of the characteristics sufficient to know with certainty who or what is intended—that is, when the extension of the definition contains only one item.²

Degeneration: From Objects to Concepts

The law of gravitation in practice deals with concrete objects. According to that law, every concrete object that has mass is attracted to the earth. On the other hand, one can say that the law deals with a concept: an object that has mass. Or, in another formulation, that the law deals with the class of objects characterized by having mass. But every concrete object also has many other properties, which the law of gravitation ignores. It applies equally to objects with mass that are black or pink, triangular, rectangular, or square in shape, made of wood or metal, and so on.

That is precisely the function of generalization. Generalization filters out from the concrete object those characteristics that are irrelevant to the specific scientific context under discussion. It tells us that the objects we observed to be attracted to the earth all had mass. Therefore it seems that only the property of having mass is important to the discussion of gravitational attraction. All the other properties are “peeled off” from those objects and thrown aside.

Generalization is an expansion of the domain of application: from the individual object to a class of objects with similar relevant properties. On the other hand, we saw that expanding a domain of application means reducing the number of characteristics. Generalization, then, is the filtering out of irrelevant characteristics.

This is a process of degeneration. The object examined in the experiment is stripped of all irrelevant characteristics, and left with very few characteristics; those are the ones that characterize it from the standpoint of the law of gravitation. The result is that the law of gravitation deals with objects that possess only one characteristic: mass.

But in reality there are no such objects. Every concrete object has many additional properties. The necessary conclusion is that the law of gravitation does not deal with concrete objects at all, but with abstract objects—objects possessing only one characteristic. Physicists call them “masses.” Its application, however, is to concrete objects, and the assumption is that the other specific properties of real objects are irrelevant.

The transition from a concrete object to an abstract object, or concept, proceeds in two directions:

1. In the process of generalization: inferring from particular experiments performed on concrete objects to the general law, which deals with a concept or abstract object.
2. In the process of specification: applying the general law, which deals with concepts and abstract objects, to the analysis of a concrete situation involving concrete objects.

This is the process of degeneration of objects and concepts that occurs in scientific generalization. Science deals with concrete objects, but theoretical science draws its conclusions by moving to an abstract plane of objects, the plane with which scientific laws are concerned.

As noted, there is an assumption here that scientific laws, which are valid for concepts and classes—that is, for the collection of objects possessing mass—can be applied to concrete objects. In other words, the assumption is that the other characteristics do not interfere with the validity of the scientific generalization.

The Relation Between Theory and Application

The picture that emerges from what we have described so far is that theoretical science deals with abstract objects and with relations among classes of characteristics. We are concerned with the relation between a property such as “having mass” and a phenomenon such as attraction to other masses, or to the earth. Its application, by contrast, is carried out on concrete objects. The chair standing before me has the property of having mass, but it also has many other properties. We apply the general scientific law to it as well, while ignoring—peeling away, or degenerating—its other properties.

Thus there is a world of shadows, or Ideas in Plato’s language, that lies above, or within, our concrete world. Theoretical science deals with the one world, and applied science with the other. For some reason, the degree of overlap between them is quite impressive. Although theoretical discussion takes place within that world of Ideas, and deals with relations among abstract objects and classes of properties, its application succeeds in fitting what happens in the real world. This is a critical point for our discussion.

The Problem of Generalization and the Problem of Degeneration

The problem of the reliability of scientific generalization can be formulated in terms of the phenomenon of degeneration. Any collection of concrete phenomena can be generalized in infinitely many ways, and therefore also described in infinitely many idealized ways. As we saw, the problem of scientific induction—how do we know that we have arrived at the correct generalization?—is equivalent to the problem of degeneration—how do we know that we have abstracted from the concrete object precisely those properties that are truly irrelevant? In other words: how do we know that we have arrived at the correct abstract object, or the correct Idea?

We mentioned at the outset that this question troubles us both in the scientific context and in the context of the hermeneutical principles. The principles help us understand when and in what direction to generalize, but they do not provide us with a recipe that uniquely determines and defines that generalization. That is done by our reasoning.

Actualism and Informativism

As a result of this whole cluster of questions, two approaches have developed over the years, both with regard to the hermeneutical principles and with regard to science:[^3]

1. The actualist approach—according to which generalizations are fictions, and in fact say nothing about the world or about Scripture, but only about us.
2. The informativist approach—according to which generalizations too are judged in terms of truth and falsehood, and they too make claims about the world or about Scripture. According to this view, generalizations are not merely efficient descriptions of the particular scientific or scriptural facts, but themselves constitute claims about the world.

A detailed account of the actualism-informativism problem appears in M. Avraham’s book That Which Exists and That Which Does Not. A summary of it, which tries to distill the main argument against actualism, is presented in the attached article. If the reader is indeed convinced that we possess tools for sound generalization, then no motivation remains for an actualist approach that casts doubt on our reasoning and on our power to generalize. There is, of course, always the possibility of error in any generalization, scriptural or scientific, but it is clear that generalizations and reasoning are not “shots in the dark,” as actualism claims.

This common basis helps create trust in scientific generalizations, as well as in the interpretive generalizations with which we dealt in last year’s article and in many other articles. This also answers the questions mentioned above, which expect the system of hermeneutical principles to provide tools that lead us all the way to the halakhic conclusion itself. As we have noted more than once, the hermeneutical principles do not even claim to do that. For example, gezerah shavah (verbal analogy) instructs us to compare two contexts, but it does not explain in what respect to make the comparison and in what respects not to compare them. The same is true of the various principles of general and specific that deal with the definition of classes, as mentioned above. The completion and formulation of every interpretive act, just like every scientific law, is accomplished by means of our reasoning. A possible philosophical-empirical basis for our trust in those tools is given in the attached article.

Footnotes