Ki Tisa (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, Mida Tova — Sabbath eve of Parashat Ki Tisa, 5766

Questions

1. Why do we not expound a general followed by a particular when they are far apart?
2. Two kinds of order in the Torah: historical-chronological and textual.
3. Why is the basis of the rule of general-and-particular specifically chronological?
4. Is there a difference between uncertainty about the interpretation of the Torah and ordinary legal uncertainty?
5. What is the law in a case of doubt whether, and how, to make an interpretive derivation?
6. How does all this relate to the rule: when an analogy can support both a lenient and a stringent reading, we follow the stringent reading?
7. What does all this have to do with the dispute between Rashi and Nahmanides concerning the rule that there is no earlier and later in the Torah?

The hermeneutical principles

• General followed by particular
• Particular followed by general
• There is no earlier and later in the Torah
• When an analogy can support both a lenient and a stringent reading, we follow the stringent reading

A. Summary of last year’s article

Every first issue of the womb is Mine, and every male among your livestock, the firstling of ox and sheep. But the firstling of a donkey you shall redeem with a sheep, and if you do not redeem it, you shall break its neck. Every firstborn of your sons you shall redeem, and they shall not appear before Me empty-handed.

— Exodus 34:19–20

Every first issue of the womb of all flesh that they offer to the Lord, whether man or beast, shall be yours; but you must surely redeem the firstborn of man, and the firstborn of an unclean beast you shall redeem. And those redeemed from a month old you shall redeem, according to your valuation, five shekels of silver, by the sanctuary shekel, which is twenty gerahs. But the firstborn of an ox, or the firstborn of a sheep, or the firstborn of a goat, you shall not redeem; they are holy. You shall dash their blood upon the altar and burn their fat as an offering by fire, a pleasing aroma to the Lord.

— Numbers 18:15–17

… And from where does our Tanna derive the exclusion of horses and camels? Rav Pappa said: “Every male among your livestock” is a general term; “ox and sheep and donkey” is a particular term. In a general followed by a particular, the general includes only what is specified in the particular. Ox, sheep, and donkey, yes; anything else, no. And Rabbi Yose the Galilean? The word “first issue” interrupts the subject. And the Rabbis? The conjunctive vav returns and joins the verse together. And Rabbi Yose the Galilean? Then Scripture should have written neither the vav nor the word “first issue.” And the Rabbis? Since this one involves monetary sanctity and that one involves inherent sanctity, Scripture separated them and then joined them again.

— Babylonian Talmud, Bekhorot 5b–6a

In last year’s article we dealt with the general-and-particular derivation from which the law is learned that among impure animals only the firstborn donkey is redeemed, and not a horse or camel. We noted that from the course of the Gemara it appears that the derivations cited here are somewhat forced. The derivation itself is intricate and difficult: it is not clear where exactly the structure of general and particular is located, and the Gemara states that there is an interruption in the middle. Here I will deal only with the general insights that emerged from that article, and not with the interpretation of the derivation itself.

At the beginning of our discussion we examined the rule that there is no earlier and later in the Torah. This rule appears last in the list of hermeneutical principles of Rabbi Eliezer son of Rabbi Yose the Galilean, and at first glance it seems to belong to the realm of aggadic, non-legal exposition. Later we saw that the Gemara in Babylonian Talmud, Pesahim 6b uses this rule in legal contexts as well. It derives from it the principle that one does not expound a general-and-particular pattern when the two are far apart, since, because of the problematic order of verses in Scripture, it is unclear whether the case before us is really general-and-particular or rather particular-and-general.

This application of the rule that there is no earlier and later in the Torah raises difficulties. The main difficulty is that the derivations of general and particular would seem to be based on textual order, not on historical-chronological order. If so, why is there any connection at all between the rule that there is no earlier and later in the Torah, which deals with chronology, and our issue? We saw from the language of the Gemara and the early authorities that the “correct” order with respect to commandments is the order in which they were given by the Holy One, blessed be He.

From the Gemara’s line of reasoning it emerges that derivations of general and particular are based on the order in which the commands were given, and not on textual order. These conclusions raise the question: how was the order of the Torah’s writing determined? It seems that there was a separate command from God to Moses regarding the order of writing.

A hint to this conclusion may be found in various verses in the Torah that command Moses directly with respect to writing, and not with respect to any particular content.¹ Another hint is found in Ra’avan, sec. 34, cited in the glosses of Rabbi Betzalel Rensburg to Babylonian Talmud, Yevamot 4a. He writes that in Deuteronomy there is earlier and later, because Moses wrote it himself and arranged it section by section. Seemingly, this implies that in the rest of the Torah, where there is no earlier and later, God Himself arranged the material. And if God Himself arranged the material not according to the correct order, that proves that there were two levels of command, as stated above.²

In the third chapter of that article we dealt with the rule concerning a general and a particular that are far apart, namely, that they are not expounded. In the sugya in Pesahim it is explained that the reason is the concern that they were not said in order, that is, the principle that there is no earlier and later in the Torah. We saw that there are commentators, such as the author of Yefeh Mar’eh, from whose words it emerges that this is not the reason; rather, the requirement that the general and the particular be adjacent is a purely formal requirement. In the final part of the article we returned and explained the difficulties in the derivations of general and particular cited above, but I will not revisit that discussion here.

B. Legal, interpretive, and hermeneutical doubts

Two kinds of order and two commands

The Torah is arranged according to two distinct systems of order:

1. A historical-chronological order, according to the times at which the commands were given to Moses.
2. An order of writing.

Both kinds of order were given to Moses by God at Sinai, and the order that determines derivations of general and particular is the historical-chronological order.

In last year’s article we noted that, accordingly, the principles of general and particular are not wholly textual. They are indeed built on the forms of expression used by the Torah, whether in general language or in particular language, but the directionality by which one chooses the relevant hermeneutical rule—general-and-particular or particular-and-general—is not determined on the textual plane, but on the historical-chronological plane.

Broadening the picture

These remarks help us understand two claims that appear in two Tosafot comments beginning with the word “But” in the sugya in Pesahim, cited in the final part of last year’s article. One qualifies the rule concerning distant verses, and the other presents an exception concerning a general and a particular that occur within the same verse:

1. With regard to “Remember” and “Observe,” Tosafot state that since both were said in a single utterance, that is, together, then even though they are written far apart in the Torah, in Parashat Yitro and in Parashat Va’ethanan, they can nevertheless be expounded through general and particular.
2. Tosafot further write that if, for some reason, it is clear that the order of words within a verse is not the genuine order but the reverse—in other words, in those cases where we can know what the order of the command actually was—then we must expound according to the “correct” order, and not according to the order of writing.

Both of these directives fit the conclusion that the order which determines derivations of general and particular is the order of command, not the order of writing. Tosafot state that ordinarily the two kinds of order coincide within a single verse, but in a case where we know that the order of command differed from the order of writing, then even though the general and the particular appear in the same verse, we expound according to the order of command. With regard to verses that are far apart, where we are usually uncertain about the order, we therefore do not expound them through general-and-particular. But if the order of command is known, they may nevertheless be expounded. Again, the decisive order is the order of command.

A difficulty raised by Tosafot

From Tosafot it emerges that the reason we do not expound a general and a particular when they are far apart is uncertainty about the order of command. Note that according to the author of Yefeh Mar’eh, whom we cited last year, this is not so. On his view there is a formal principle that one does not expound general-and-particular across distant verses, independently of the principle that there is no earlier and later in the Torah.

But according to Tosafot a difficulty arises. In a case of a general and a particular that are far apart, because the order is unknown, there is doubt between two possibilities: either we should expound it as general-and-particular, in which case the general includes only what is specified in the particular; or we should expound it as particular-and-general, in which case we judge according to the character of the particular. In both cases we end up with a domain of application that is some extension of the particular. If so, we should have applied the ordinary rules of doubt, according to which in biblical law one goes stringently. In other words, we should at least have applied the law to the minimal group—that is, only to what is specified in the particular. So why in fact do we not expound a far-apart general-and-particular in this way? Seemingly, in such a case we should have expounded it like general-and-particular, namely, only what is specified in the particular. Of course, according to the view of Yefeh Mar’eh this difficulty does not arise at all. In fact, there is here evidence for his position.

How should we relate to interpretive doubt?

A possible resolution of our difficulty with Tosafot may be suggested once we take into account that what we have here is an interpretive doubt. If there are two possible ways to expound a set of verses, then the Torah itself has left the matter ambiguous. In such a situation one may argue that we should not expound anything at all, for by leaving ambiguity in the text the Torah itself indicates that we should not expound it. In other words: if the Torah had wanted us to expound these verses through general-and-particular, it should have written them in a single verse, or at least in a way that made the relevant order clear to us, as in Tosafot’s extension above, rather than leaving us in doubt about the true order.

Rashba on liability for damage caused by tooth and foot

The Gemara at the beginning of Bava Kamma, on pages 2b–3a, discusses the source for liability in the various primary categories of damages: horn, tooth, foot, and others. With respect to tooth and foot, the Gemara there, on 3a, brings a verse that can be interpreted either as a source for tooth or as a source for foot, and it is uncertain which of the two should be learned from it. The Gemara states that since neither has priority over the other—”which one of them would you exclude?”—we should derive both of these categories together from that verse.

Rashba, in his novellae there, asks why we should not rule the other way. We should exempt both tooth and foot, for this is a monetary doubt, and in monetary doubt one rules leniently for the defendant. Therefore, so long as we do not have clear proof that the verse obligates payment for tooth or for foot, we should exempt the damager in both cases. Rashba answers that if we learned neither of them, the verse would turn out to be superfluous. If so, it is clear that the verse comes to teach something, and since both derivations are possible and there is no way to decide between them, apparently we must derive both from it. Let us add that the liability for tooth and foot would obviously not remain merely doubtful, but would become certain. This initially doubtful consideration reveals to us the meaning of the verse, and from that point onward the law is certain.

At first glance, this is the same mode of thought that we suggested above regarding a general and a particular that are far apart. When there is no problem of leaving a verse superfluous, we should give up both derivations and leave the situation as though there were no derivations at all. But we should note that Rashba says this only because it is a monetary doubt. In another kind of doubt, if there is no consideration of leaving the verse superfluous, we should follow the ordinary laws of doubt, and in a case of prohibition we should decide stringently.

Moreover, according to our approach above, if the Torah wrote a verse in an ambiguous way, such that it can be interpreted either about tooth or about foot, then apparently it intended neither of them specifically.

Rashba on liability for damage caused by horn

At the beginning of that sugya, on 2b, the Gemara discusses the source for liability for horn. In damage by horn, the rule is that for the first three gorings one pays only half-damages, while from the fourth goring onward one becomes liable for full damages. In the course of the discussion the Gemara raises a doubt about an attached horn, that is, the horn of an ox: does the distinction between the first three gorings and a habitual gorer apply there, or not? The Gemara states that if this distinction exists only for a detached horn—an iron ram, of the sort used by Zedekiah son of Chenaanah; see 1 Kings 22—then with an attached horn it would be deemed fully dangerous and would pay full damages from the outset.

Rashba asks there why we should not say that with an attached horn, if indeed we have no source for distinguishing between the first three gorings and later ones, one should always pay only half-damages, as in the less severe case. This is, after all, a monetary doubt, and in such a case one rules leniently for the defendant. Therefore the damager, who is the defendant, should pay according to the minimal possibility. Rashba explains that in a doubt in the laws of damages, the rule is like a doubt in prohibition, where one goes stringently.

These remarks are very difficult, and later authorities have already pointed this out. First, even if this were a doubt in prohibition and one had to be stringent, stringency for the defendant is leniency for the plaintiff. Beyond that, we find many doubtful cases in the laws of damages where one rules leniently for the defendant, so why specifically here should one go stringently? And moreover, in the previous passage we saw that Rashba himself treats a doubt in the laws of damages as a monetary doubt, in which we decide leniently for the current holder.³

Rabbi Baruch Dov Leibowitz, in his Birkat Shmuel on Bava Kamma, end of sec. 2, cites his teacher Rabbi Hayyim Soloveitchik of Brisk as distinguishing between doubt about the interpretation of a verse and ordinary monetary doubt. He explains that there is a principle in the Torah that when there are two possible ways to learn a verse, one always follows the more stringent direction. In this context he cites the rule: when an analogy can support both a lenient and a stringent reading, we follow the stringent reading, as will be discussed below.

This approach directly contradicts what we proposed above. There we wanted to argue that when there are two possible ways to expound the Torah, we should forgo both of them. In a situation where otherwise the verse would become superfluous, as in the case of tooth and foot above, that is impossible. But here the verse is not superfluous, and therefore we should have forgone both derivations.⁴

Doubt in plain meaning and doubt in hermeneutical derivation

It seems, however, that all of this is not relevant to our present issue. The cases we have seen thus far involve doubts in the plain-sense interpretation of verses. The question whether a given verse speaks about tooth or about foot is a question about the meaning of the verse. There it is obvious that we must find the optimal interpretation, and certainly not leave a verse superfluous. But in our case the doubt is whether to make a derivation at all, or which derivation to make—general-and-particular or particular-and-general. This is a situation in which the verses have a complete plain-sense meaning, and no word in them is superfluous. The question is whether to add an additional layer of interpretive derivation beyond the plain meaning, or not. Here our claim is that if the Torah formulates itself ambiguously, in a way that does not allow us to know what to derive or how to derive it, then we should not derive from those verses.

It is important to clarify that our remarks do not concern doubt about a law that was learned from derivations. There one must go stringently, like any doubt in biblical law. Here we are dealing only with doubt whether to derive from a verse in the first place, and the claim is that in such a doubt we should not derive from it, but should leave it with its plain meaning alone and without any added interpretive layer.⁵

“When an analogy can support both a lenient and a stringent reading, we follow the stringent reading”

The question of how we should act in a situation of doubt whether, and how, to derive arises in the well-known rule: when an analogy can support both a lenient and a stringent reading, we follow the stringent reading. According to this rule, if there are two possible analogies, one leading to leniency and one to stringency, we draw the analogy in the stringent direction. At first glance, this rule appears to contradict what we have said thus far. It would seem to show that where there is doubt whether to derive, we should follow the ordinary laws of doubt.

But this conclusion is not necessary at all. Several early and later authorities have already noted that this rule does not belong to the laws of doubt; rather, it is one of the rules of derivation themselves. For example, Tosafot, s.v. “Similar to,” on Bava Kamma 3a cites Rabbenu Tam as saying that even in monetary law we draw the analogy stringently, although according to the ordinary laws of monetary doubt we should have been lenient toward the defendant. It therefore appears that the ordinary laws of doubt do not apply when we are uncertain how to derive.⁶ There are, however, many disagreements concerning the application of this rule to monetary law; see Yavin Shemu’ah, principle 111. It may be that the disagreement itself turns on this very point: whether this rule belongs to the laws of doubt or not.⁷

Even so, this rule, even if it is not based on the laws of doubt, still seems to run counter to what we have said so far, for in the end it determines that in a case of doubt whether to derive one should go stringently. If so, then in the case of a general and a particular that are far apart we should also have gone stringently.

The answer is that one must examine whether this rule applies to all hermeneutical principles, or perhaps only to analogy. In Yavin Shemu’ah, principle 111, it is brought that the early authorities disputed whether this rule was stated also with respect to verbal analogy, or only with respect to analogy. From their words we learn that the rule was certainly not stated with respect to other forms of derivation. We also learn that this rule is not part of the laws of doubt, for if it were merely a formulation of the laws of doubt rather than a special hermeneutical rule, it should apply equally to all forms of derivation without distinction.

If so, what should we do in situations where there is doubt whether, and how, to derive, and the case does not involve analogy or verbal analogy? In other words, what happens in the absence of this special rule? There is good reason to say that in such a case we do not perform a derivation, exactly as we argued above. The Tosafot cited earlier, who say that one does not expound a far-apart general-and-particular, also provide support for this conclusion. Let us note only that refraining from making a derivation is sometimes a stringency and sometimes a leniency. That is, no general rule can be stated for doubtful derivations as such, whether they are stringent or lenient. The rule is simply that when there is doubt whether and how to derive, we do not derive.

C. The meaning of the two kinds of order

Introduction

We have pointed to two kinds of order in Scripture, each of which was separately commanded to Moses: the historical-chronological order and the order of writing. We also saw that derivations of general and particular are determined specifically by the historical-chronological order.

This picture raises two main difficulties:

1. What was the purpose of the second command, the command concerning the order of writing? Why should the order of writing differ from the order of command at all? Clearly its purpose was not to guide the interpretive plane—that is, to tell us whether to expound general-and-particular or particular-and-general—for we saw that these derivations are determined by the chronological order of the commands and not by the order of writing.
2. This picture runs contrary to simple intuition. We would have thought that the substantive commands were given according to the inner connection among their contents, that is, on essential grounds, whereas the arrangement of writing might be intended to produce a text fit for exposition. Here the picture is precisely the reverse: the derivations are based on the historical-chronological order, while the textual order serves other purposes, which we do not yet know at this stage, as noted in difficulty 1. Why, then, are the derivations of general and particular based specifically on the historical-chronological order rather than on the order of writing?

At first glance, it is reasonable to assume that the two questions are connected. Presumably, if we understand the reason for the Torah’s different textual arrangement, we may also understand why the interpretive derivation is not determined by the textual order of editing.

Textual editorial considerations in the Torah

The early authorities mention several kinds of considerations that lead to changes in order in the Torah, some interpretive and some legal.⁸ From their words it emerges that the considerations governing arrangement were local, and it is difficult to subsume them under a single heading. The natural order remains in force so long as no local consideration causes it to be altered. From this it follows that the natural and “correct” order is not the textual order. The correct order is the order of command, and the text sometimes deviates from it because of local considerations.

If so, we may now understand why textual order is not the basis for derivations of general and particular. These derivations are based on the essential order of the matters themselves, and not on their textual order, which does not reflect that essence.

Still, with respect to the order of commandments, as opposed to events, the matter is more difficult. What considerations could stand behind a change in the written order of commandments, if not considerations of interpretation of the commandments themselves? Various literary considerations may be relevant to the narrative sections of the Torah, but not to its legal sections.

A puzzling midrash and a solution to the problems

In Midrash Tehillim (Buber), Psalm 3, s.v. “Another interpretation,” we read:

Another interpretation: “A psalm of David when he fled.” This is what Scripture says: “No man knows its value” (Job 28:13). Rabbi Elazar said: The sections of the Torah were not given in order. For had they been given in order, anyone who read them could revive the dead and perform wonders. Therefore the order of the Torah was hidden, and it is revealed before the Holy One, blessed be He, as it is said: “And who is like Me? Let him proclaim it and tell it and set it in order for Me.”

— Midrash Tehillim (Buber), Psalm 3⁹

We find here that the Holy One, blessed be He, deliberately scrambled the order of the Torah’s writing in order to withhold knowledge from human beings. The correct order is known only to God, and possession of it would enable anyone who possessed it to revive the dead. Therefore God kept it to Himself and had the Torah written in a different order.¹⁰

From this we learn three things:

1. There is a general consideration underlying God’s command to Moses to write the Torah out of order, and not merely a collection of local considerations. God commanded Moses and told him the commandments in the correct order, but there was an additional command to mix them and write the Torah in a different order.¹¹
2. The correct order is essential, and not merely chronological. Whoever knows the “correct” order of the Torah knows how to revive the dead. In other words, this is the order appropriate to knowledge of reality itself.
3. The textual order of what is written does not necessarily have significance in itself. It does not contain information or ideas; rather, it is intended specifically to conceal information.

Thus we have understood the significance of the second command, namely, to alter the order of writing. We can now also understand why derivations, which ought to be based on the essential order, are made specifically on the basis of the chronological order. The chronology of the command is itself the essential order. The textual order is partly arbitrary, and it is not intended for the transmission of information but for the concealment of information.

Once again: there is no earlier and later in the Torah

We mentioned in last year’s article that Nahmanides, in his commentary on Numbers 16:1, maintains that the Torah does not alter the order of writing without reason. See also his commentary on Numbers 9:1 and Leviticus 8:1. By contrast, Rashi and Ibn Ezra generally use this rule with much greater freedom, and Nahmanides criticizes them on this point.

At first glance, both sides are difficult. First, with respect to Rashi and Ibn Ezra: why should Scripture be written in an incorrect order without any reason? But in light of the midrash cited above, we learn that this indeed happens without local reasons. In fact, many sources imply that the rule that there is no earlier and later in the Torah is invoked without any justification at all, and no justification is even sought. Usually it is brought as a definitive reason why we should not rely on the order of the verses, and the Sages do not feel obliged to explain why the verse in question departed from the proper order.

Second, Nahmanides is difficult, since he holds that the order changes only when there is a reason. If so, why do we not expound general-and-particular when the two are far apart in those cases where there is no reason to think the order has changed? Seemingly, in such cases we are not in doubt at all, and the textual order is also the correct one. Nahmanides’ statements surely contradict the midrash cited above.

One might have tried to explain that in principle there must be a reason, but we do not always know what it is, and therefore we must always suspect that the order has changed. But Nahmanides’ objections to Rashi and Ibn Ezra show that this is not his view. After all, he argues against them that one should not disturb the order unless a reason has been provided. So why, when no reason to alter the order is apparent, do we not expound general-and-particular when the verses are far apart?

Here we return to the position of the author of Yefeh Mar’eh mentioned above, according to whom the fact that we do not expound a far-apart general-and-particular does not result from the rule that there is no earlier and later in the Torah, but is instead a formal limitation within the rule of general and particular. The matter requires further investigation.

Footnotes