Lesson 12: Vayetze

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

---

With God’s help

Concepts

• Paradoxes and loops in halakhic (Jewish-legal) reasoning.
• Logical and arithmetic frameworks for analyzing halakhic relations.

Abstract

This article begins with the ordinance of Usha that a person should not spend more than one-fifth of his assets on charity and on positive commandments. By contrast, in order to avoid violating a negative commandment, a person is obligated to spend all of his wealth. At first glance, the negative commandment is therefore more severe than the positive one. On the other hand, the halakhic rule states that a positive commandment overrides a negative commandment, and from that it would seem specifically that the positive commandment is more important.

Ramban, in his commentary on the Ten Commandments, gives us a hint toward solving this problem, and we use his remarks in order to examine the issue within a logical framework and within an arithmetic framework.

At the final stage we present another paradox, also a logical loop, involving three halakhic principles, two of them connected to the previous discussion. There too it appears that this problem cannot really be arithmetized at all.

At the end, we hint at possible directions for a solution by searching more carefully for hidden assumptions in the arithmetization we proposed, though we do not actually present that solution here.

The Rules and Principles That Emerge from the Article

This is a corrected version, unlike the printed one. It is also corrected in this book.

Concerning the Rule: One Who Spends Should Not Spend More Than One-Fifth

Logical and Arithmetic Models for Examining Relations Between Halakhic Principles

A. The Ordinance of Usha

Introduction: The Rule of One-Fifth

Jacob awakens from his dream and immediately vows a vow to God, as it says in Genesis 28:22:

“And this stone, which I have set up as a pillar, shall be a house of God; and of all that You give me, I will surely tithe to You.”

And in the midrash (rabbinic interpretive text) Bereshit Rabbati, Vayetze, p. 139, we find:

“‘I will surely tithe to You.’ From here the Sages said: one who spends should not spend more than one-fifth. We find with Jacob that he vowed only one-fifth, as it is said, ‘I will surely tithe to You.’ And from where do you learn that this is one-fifth? Because it does not say, ‘I will tithe one tenth to You.’ Rather, the latter tenth is equal to the former tenth, and that makes it one-fifth.”

At first glance, this appears to be a biblical law. Yet in the Babylonian Talmud, Ketubot 50a, we find:

“Rabbi Ila’i said: In Usha they enacted that one who spends should not spend more than one-fifth. A baraita likewise taught: One who spends should not spend more than one-fifth, lest he come to depend on others. There was an incident involving a certain person who wished to spend more than one-fifth, and his fellow did not allow him. And who was he? Rabbi Yeshavav. And some say: It was Rabbi Yeshavav who wished to spend, and his fellow, Rabbi Akiva, did not allow him. Rav Nahman, and some say Rav Aha bar Yaakov, said: What is the verse? ‘And of all that You give me, I will surely tithe to You.’ But the second tithe is not like the first tithe. Rav Ashi said: ‘I will surely tithe it’ means: I will make the latter like the former.”

At first glance, a person could at any given moment spend one-fifth of his wealth, but if he did so repeatedly he would very quickly lose all his property. Therefore the medieval authorities here, including Tosafot and Ritva, cite the Jerusalem Talmud to the effect that the setting aside is like tithing: one-fifth of the principal the first time, and from then onward one-fifth of whatever was added, that is, the profit. In other words, every sum of money that comes to a person must be “tithed,” and the rate is up to one-fifth, that is, up to twenty cents of every dollar. From the Gemara itself it appears that not only is there no obligation to spend more than one-fifth, but there is actually a prohibition against spending beyond one-fifth.

The Status of the Rule

On its face, this seems to be a rabbinic rule, and the midrash is only a scriptural support. However, the Vilna Gaon, in Shenot Eliyahu on Mishnah Pe’ah 1:1, infers from the Jerusalem Talmud that this is a biblical law, a law given to Moses at Sinai. The same view appears in Maharam Shik, Yoreh De’ah 230, and see also Gilyon HaShas to Ketubot there.

In Responsa Hatam Sofer, Yoreh De’ah 229, he writes that by Torah law one must give all money not needed for daily sustenance, and in Usha they enacted only a prohibition on giving more than one-fifth. If so, the obligation to give one-fifth is biblical. This also appears to be the implication of Maimonides’ language in his Commentary on the Mishnah there on Pe’ah.

However, according to most opinions, the obligation to give one-fifth is rabbinic, and in practical law, according to most decisors, there is in fact no obligation to give one-fifth at all, only a prohibition against giving more than one-fifth.

For those opinions that regard this as a rabbinic obligation, one may ask: what is the Torah law? Is one biblically obligated to give more, perhaps even all of one’s property that is not needed for one’s livelihood? The simple reading seems to be that by Torah law one may give more, while the ordinance of Usha prohibited giving above one-fifth.¹

For What Purpose Was the Rule of One-Fifth Stated?

Rashi writes there:

“One who spends — for the poor.”

From Rashi it appears that the subject of the midrash is the proper measure for charity, and the midrash learns from Jacob that the maximum measure is one-fifth of a person’s assets. So too in the Shulchan Arukh and Rema, Yoreh De’ah 249:1:

“The measure of giving is as follows: if his means suffice, he should give according to the needs of the poor. If his means do not suffice to that extent, he should give up to one-fifth of his assets; this is the choicest fulfillment of the commandment. One-tenth is an average measure; less than this is a grudging eye. And this one-fifth of which they spoke is, in the first year, from the principal, and from then onward one-fifth of what he earns each year.”

“Gloss: A person should not spend more than one-fifth, lest he come to depend on others. And this applies specifically during his lifetime, but at the time of death a person may give whatever charity he wishes. One should not use his tithe money for another commandment, such as candles for the synagogue or any other commandment-related item; rather, he should give it to the poor.”

However, in the decisors we also find an expansion of this rule to expenditures on commandments in general, and this expansion too was ruled as practical law by the Rema. In Orach Chayim 656:1, in the course of a discussion of how much money one must spend on beautifying a commandment, the Rema writes:

“One who does not have an etrog, or some other time-sensitive commandment, need not spend a great fortune for it, as they said: one who spends should not spend more than one-fifth, even for a passing commandment. And this applies specifically to a positive commandment, but for a negative commandment he must give all his money before violating it.”

Indeed, the wording of the Jerusalem Talmud in Pe’ah is: “In Usha they resolved that a person should set aside one-fifth of his assets for a commandment.” See also Mahar”י Korkos on Maimonides, Hilkhot Mattenot Aniyim 7:5. This implies that there is an obligation to spend one-fifth on commandments in general.

This can also be argued on conceptual grounds. Many have explained that the ordinance of Usha was intended to prevent the giver himself from becoming dependent on others; after all, it makes no sense to give charity and thereby become needy yourself. But the Gemara says that a person never becomes poor through charity. Nor can this be dismissed as a merely aggadic statement, since it is brought in the Shulchan Arukh, Yoreh De’ah 247:2; the Rema there, in 247:4, goes so far as to say that one may test God in this matter; and Maimonides states it in chapter 10 of Hilkhot Mattenot Aniyim. The decisors even drew practical halakhic conclusions from it. The Gemara in Babylonian Talmud, Gittin 7b, inferred from this that even a poor person is obligated to give charity.

But if we say that the ordinance of Usha concerns other commandments and not charity alone, then the matter is entirely coherent. Still, see Rashi to Ketubot there, who explicitly writes that it concerns charity specifically. For a commandment, perhaps the measure is different; see Babylonian Talmud, Bava Kamma 9b and the commentators there.

Maimonides, at the end of Hilkhot Arakhin, writes that one should not give more than one-fifth, whereas in Hilkhot Mattenot Aniyim he does not write this. See Sefer HaMafte’ach in the Frankel edition, Mattenot Aniyim 7:5. In fact, there Maimonides is dealing with other commandments and not charity. See also Kli Chemdah on Vayetze, sec. 4.

Between Charity and Other Commandments

Up to this point we have seen that the ordinance of Usha has been understood in two directions:

1. As a measure for the obligation to give charity.
2. As an obligation to spend on a commandment.

In practical law there is a difference between these two contexts. With respect to charity, there is permission to give up to one-fifth, but not an absolute obligation. By contrast, with respect to other commandments, there is a full obligation to spend one-fifth of one’s assets in order to avoid neglecting a positive commandment.² With respect to charity, the one-fifth is given like a tithe, that is, as a set-aside from the sums the person earns, namely income. With respect to a commandment, however, it is commonly understood that the obligation is to spend up to one-fifth of the total money in his possession at that moment.

It is not clear how the ordinance of Usha can be interpreted in these two different ways in terms of legal definition. As we have seen, even the concept of “one-fifth” is understood differently in the two cases. It is possible that there are actually two separate enactments here: the one cited in the Babylonian Talmud, which concerns charity, and the one in the Jerusalem Talmud, which concerns other commandments. In any event, for practical law it is clear that both interpretations are simultaneously valid.³

B. The Relation Between a Positive Commandment and a Negative Commandment

The Rule Regarding a Negative Commandment

In the previous section we saw that one is obligated to spend up to one-fifth of one’s money in order to fulfill a commandment. But what about negative commandments? Here the halakha rules that a person must spend all of his money in order not to violate a prohibition. There is no explicit source for this in rabbinic literature, but in the Rema cited above we saw that he brings this in the name of Ra’avad and Rashba; see Rashba’s novellae to Bava Kamma 9b, s.v. “Ve-ha.” Precisely because there is no explicit source, it is fairly clear that this is a biblical rule. Had it been rabbinic, Ra’avad and Rashba would have had to find an explicit source that such an enactment existed. It seems from their words that they understood by simple reasoning that this is the Torah law.

A Difficulty: Between the Positive and the Negative

From what we have seen so far, it emerges that a negative commandment is more severe than a positive one. In order to avoid violating a prohibition, one must spend all of one’s money, whereas in order not to neglect a positive commandment it is enough to spend one-fifth of one’s assets. We also know that violating a prohibition is more severe than neglecting a positive commandment in other respects: prohibitions carry punishments, whereas for a positive commandment we do not find a formal punishment in halakha.

On the other hand, we know the rule that a positive commandment overrides a negative commandment, and from this the opposite picture emerges: the positive commandment seems more important than the negative one.

Ramban’s Words in Parashat Yitro

Ramban, in his commentary on the Ten Commandments, briefly discusses the relation between positive and negative commandments. He writes, on Exodus 20:7:

“It is also true that the quality of ‘remember’ alludes to positive commandments, and it stems from the quality of love and belongs to the attribute of mercy; for one who performs his master’s commandments is beloved to him, and his master has mercy upon him. The quality of ‘observe’ belongs to negative commandments, and it belongs to the attribute of judgment and stems from the quality of fear; for one who guards himself from doing what is evil in his master’s eyes fears him. Therefore a positive commandment is greater than a negative commandment, just as love is greater than fear, for one who fulfills and performs his master’s will with his body and his money is greater than one who merely guards himself from doing what is evil in his eyes. Therefore they said that a positive commandment comes and overrides a negative commandment. And for this reason the punishment for negative commandments is greater, and legal sanctions are imposed for them, such as lashes and death, whereas for positive commandments no such sanction is imposed at all, except on those who rebel, as when one says: I will not perform lulav, I will not perform tzitzit, I will not perform sukkah. In such cases the Sanhedrin would strike him until he accepted upon himself to perform it, or until he died.”

Ramban’s point of departure is the duplication between the first tablets and the second tablets. He explains that “remember” hints at positive commandments, and “observe” hints at negative ones. It should be remembered that in the context of the Sabbath there is both a negative and a positive commandment, and both have overlapping content: not to do labor.

He goes on to explain that the positive commandment is greater than the negative one, because it expresses service of God out of love, as opposed to the negative commandment, which expresses fear. Therefore the reward of one who fulfills a positive commandment is greater than that of one who merely refrains from violating a prohibition. In this way Ramban also explains the halakhic rule that a positive commandment overrides a negative one, since the positive is greater.

But the continuation of his remarks seems problematic, because he then cites the opposite kind of example, in which the negative commandment appears more severe than the positive: violating a negative commandment carries punishment, whereas failing to perform a positive commandment does not, except at most coercion to perform it in the case of one who rebels and refuses. He does not even bother to explain why, in this example, the negative is more severe than the positive, and it appears that in his view this too is a direct implication of what he has said up to that point. To this we may add what we have already seen: avoiding a negative commandment obligates a person to spend all of his money, which is not true of a positive commandment.

The Explanation of the Author of Sdei Chemed

The author of Sdei Chemed, in the entry “A Positive Commandment Overrides a Negative Commandment,” cites these words of Ramban and explains that Ramban meant to say that fulfilling a positive commandment is greater than refraining from violating a negative one; but on the other hand, the relation reverses in the opposite direction: violating a negative commandment is far worse than neglecting a positive one.

The explanation is that, although fulfilling a positive commandment is a higher level of service of God than refraining from a prohibition, precisely because of that, the failure to perform the positive commandment is a lighter transgression. At most, the person is not at a high level. The claim against him is therefore weaker than in the case of violating a prohibition, which shows contempt for something more basic, and is therefore more severe.

This can be summarized by saying that the positive commandment represents a higher and more important form of divine service, whereas refraining from a prohibition is more basic. This is a relation between the basic and the elevated. Violating the basic requirement is more severe specifically because it is a basic demand; but the fulfillment of the higher demand is more important.

A Note on Positive and Negative Commandments in the Labors of the Sabbath

Ramban explained that fulfilling the positive commandment is more important than refraining from violating the negative one, while neglecting the positive is less severe than violating the negative. The reason is that the act of violating a prohibition is a direct confrontation with the divine will, and is therefore more severe than neglecting a positive commandment, which is merely non-fulfillment of the divine will. By contrast, refraining from violating the prohibition is a passive fulfillment of the divine will, unlike the active performance of a positive commandment.

One should notice, however, that the context Ramban is discussing is the labors of the Sabbath. In that case, both refraining from the prohibition and fulfilling the positive commandment have the same practical meaning: not doing labor. If so, why should the positive commandment in that case be greater than the negative one? In both instances the person refrains from violating the divine will; that is, in both instances this is a passive fulfillment of the Torah’s will.

We can sharpen the question still more, in light of these definitions: what is the distinction between a positive and a negative commandment in the first place? We see that both with positive and with negative commandments there are situations in which fulfillment consists in inaction and violation consists in action, and vice versa. If so, what actually defines the difference? We will address this point in one of the following articles.

Application in Different Halakhic Contexts

Let us now return to Ramban’s remarks and examine the application of the principles he describes in the various halakhic contexts. As to the reward of one who fulfills a positive commandment, it is clearly greater than the reward of one who merely refrains from violating a negative one; and the reverse is true with respect to the punishment imposed for violating each.

But what about the other halakhic aspects that arose here? Avoiding a prohibition requires spending all of one’s money, because the alternative being weighed is violating the prohibition, and that is very severe. By contrast, spending money in order to fulfill a positive commandment means spending money in order to attain a higher level, and the demand to do so is weaker. True, the “commodity” is worth more, but precisely for that reason the demand on us is less absolute. We are not required to spend more than one-fifth, and perhaps it is even forbidden to do so.

If so, the rule governing how much money one must spend in order to avoid a transgression is set by comparison with the alternative. This indeed seems to have been the reasoning of the aforementioned Ra’avad and Rashba. But that very criterion leads us into a dead end when we try to examine the rule that a positive commandment overrides a negative one.

A Positive Commandment Overrides a Negative One: The Logical Problem

At first glance, the picture emerging from Ramban naturally leads to the rule that a positive commandment overrides a negative one. The positive is more important, and therefore it overrides the negative. But this is a superficial and incorrect conclusion. After all, the opposite consideration is also present here: when we violate the negative commandment in order to fulfill the positive one, we are in a more severe state, since violating a prohibition is more serious than failing to fulfill a positive commandment. Hence Ramban’s criterion does not actually provide a decision in a case of collision between a positive and a negative commandment.

According to the method we defined above, when we seek to determine which of the two overrides the other, we must weigh and compare the two alternatives. We are in a situation in which there is a collision between a positive and a negative commandment. The two possibilities are:

• either to fulfill the positive commandment and in doing so violate the negative one,
• or to guard against violating the negative one and in doing so neglect the positive commandment.

The special complexity here arises from the fact that we are not comparing one action against the Torah with another action against the Torah, namely neglecting a positive commandment versus violating a negative one, in which case it is clear that neglecting the positive is preferable. Nor are we comparing two actions that accord with the Torah, namely fulfilling a positive commandment and avoiding a negative one, in which case it is clear that it is preferable to avoid the prohibition. The comparison here combines fulfillment and neglect of both positive and negative commandments in a single equation. The question is whether fulfilling the positive commandment together with violating the negative one is preferable to neglecting the positive commandment together with avoiding the negative one, or not.

Beyond the question of which of these two is preferable, a further question arises: can this be decided solely on the basis of Ramban’s considerations alone? From his language it seems that these considerations by themselves suffice to explain why a positive commandment overrides a negative one.

Arithmetization of Commandments and Transgressions

Above we presented the problem in logical terms, that is, within a framework of principles dealing with comparisons of priority. But, as we saw, that framework does not yield a clear decision. We therefore turn to an arithmetic presentation of the equation, to see whether that will help us reach a conclusion.

Let us assume that the “price” of avoiding a negative commandment is X, and the “price” of fulfilling a positive commandment is Y. From Ramban’s words we learn the following claim: X < Y. Moreover, although Ramban does not mention the relation between positive and negative commandments with respect to how much money one must spend on each, in light of our earlier discussion we can add that there is a more precise measure for the relation between commandments and transgressions in terms of percentages of one’s property: the “price” of a negative commandment is 100% of one’s property, whereas the price of a positive commandment is only 20% of one’s property.

We must now decide what the “price” is of neglecting a positive commandment and of violating a negative one. At a superficial glance, it would seem that the price of violating a negative commandment is -X, and the price of neglecting a positive commandment is -Y. These satisfy the relation -Y < -X, which also seems, at first glance, to be implicit in Ramban’s words.

But this picture is incorrect. If indeed -Y < -X, that does not mean that it is preferable to neglect a positive commandment rather than violate a negative one. On the contrary: it would mean that the cost, in the negative sense, of neglecting the positive commandment is greater than the cost, in that same sense, of violating the negative commandment.

How, then, should we move from the positive price, namely the price of fulfilling a positive commandment or avoiding a negative one, to the negative price, namely the price of neglecting a positive commandment or violating a negative one? The mathematical function here is not simple negation.

Take an example. If the positive price, that is, the reward, for fulfilling a positive commandment is 5 units, and the price of avoiding failure with respect to a negative commandment is 3 units, then the negative price, that is, the punishment, of neglecting the positive commandment ought to be lower than the price of violating the negative one. For example, if the price of neglecting the positive commandment is -3, then the price of violating the negative one is -5.

This is obviously not a simple arithmetic inversion, because the positive and the negative commandment have exchanged roles here. But it is not difficult to find a mathematical function that performs this conversion from commandments to transgressions. For example, if the positive price of the commandment, that is, the reward, is X, then the negative price of the transgression, that is, the punishment, is obtained by subtracting 8 units, which, as expected, is the sum X + Y. Thus the price of fulfilling the commandment, which is 5 units, becomes -3 units for neglecting a positive commandment, while the price of avoiding the negative commandment, which is 3 units, becomes -5 units for violating it.

We may now define the price of each move in the situation of a positive commandment overriding a negative one. Let A be the sum of the two positive prices, so in our example A = 8.

• Fulfilling the positive commandment and violating the negative one: Y + (X - A)
• Neglecting the positive commandment and avoiding the negative one: X + (Y - A)

The problem is that these two prices are equal, and therefore one cannot derive from them any criterion for deciding in favor of one option over the other.

If so, both difficulties raised above remain in place: why, according to Ramban, does a positive commandment really override a negative one, and how does that conclusion follow solely from his assumption, without the addition of any other assumptions?

Put differently: in order to reach the conclusion that a positive commandment overrides a negative one, Ramban’s claim that X < Y is not enough. We must add another principle, namely that the gap between the price of violation and avoidance with respect to the negative commandment, A1, differs from the corresponding gap with respect to positive commandments, A2.⁴ This assumption does not arise naturally from Ramban’s words, but he does seem to assume it implicitly in his conclusion that a positive commandment overrides a negative one.

It should be noted that this is an arithmetic model, not an actual quantification. Clearly each commandment and each transgression has a different “price.” There are severe ones and lighter ones, both among prohibitions and among positive commandments. Nevertheless, the very fact that something is a negative or a positive commandment is weighted in this way. If one wishes to move toward a more realistic quantification, one must take into account a range of severe and light transgressions whose prices are distributed around X, and similarly a range of positive commandments distributed around Y. Even so, the assumption remains that the prices of fulfilling positive commandments will always be higher than the prices of avoiding prohibitions, and that the gap between commandment and transgression in the two cases will also be different.

C. The Paradox of New-Grain Matzah

The Loop Presented by the Author of Pardes Yosef

In Pardes Yosef on Genesis, at the end of chapter 28, a paradox is brought that touches on two of the principles with which we dealt in the previous section. Let us imagine a situation in which a person stands on the eve of Passover and must obtain matzah for the Seder night. Matzah is selling at a very high price, more than 20% of his wealth, a sum he is not required to pay in order to fulfill the commandment of eating matzah.

It turns out, however, that there is another possibility: to eat matzah made from new grain, meaning grain from the new crop, which is forbidden to eat before the day of the waving of the offering, that is, before the first intermediate day of Passover. At first glance, a positive commandment overrides a negative one, and therefore such a situation would justify eating matzah made from new grain despite the prohibition. The positive commandment of eating matzah overrides the prohibition of new grain.

But in order not to violate a negative commandment, a person must spend all of his money. Therefore, if he arrives at the point of eating new grain, he should have to spend all his money and buy regular matzah in order not to eat from the forbidden new grain. Yet in the final analysis this means that the person is spending all of his money in order to fulfill a positive commandment, since he could simply refrain from eating matzah altogether. On the other hand, not eating matzah at all is likewise unjustified, since a positive commandment overrides a negative one, and he can eat matzah made from new grain.

If we summarize the tangled situation that has emerged here, three options stand before the person:

1. To eat matzah made from new grain. In that case he violates a negative commandment, fulfills a positive one, and pays a reasonable price.
2. To eat matzah made from old grain. In that case he avoids violating a negative commandment, fulfills a positive one, but pays an exorbitant price, which one is not required to pay in order to fulfill a positive commandment.
3. Not to eat matzah at all. In that case he neglects a positive commandment, avoids violating a negative one, and pays nothing.

The problem is that if we examine each of these steps on its own, there is another step that is clearly preferable to it, or, put differently, in each case we are doing something forbidden.

When we consider option 1, it follows that we are violating a negative commandment when it is possible to avoid it at a high cost. But halakha requires a person to spend all of his money in order to avoid violating a negative commandment. If so, option 2 is preferable, since although one pays a high price, one avoids the prohibition.

But that too is not correct, because option 3 is preferable to option 2. In option 2 we are paying a very high price in order to fulfill a positive commandment, while there exists the possibility of not paying and not fulfilling it at all.

Yet option 3 is also not optimal, because one can fulfill the positive commandment at a reasonable cost by choosing option 1, namely by buying matzah made from new grain.

Comparing the Prices of the Alternatives

As in the previous case, the logical framework does not advance us toward a decision. In logical terms, which deal only with binary comparison between alternatives, there is here a non-transitive relation among the three options: option 2 is preferable to option 1, option 3 is preferable to option 2, and option 1 is preferable to option 3. This appears to be a loop with no way out. Yet perhaps here too, in order to reach a decision, we must move from the logical framework to a more detailed arithmetic model.

To do so, we must quantify the prices of the three alternatives. In order to do this, we must decide what the “price” is of each of the three principles involved in the loop. Above we already defined the prices of violating and fulfilling positive and negative commandments. What remains is to define the meta-halakhic “price” of paying an exorbitant amount for a positive commandment. Let us denote this price by Z, while for the moment we do not know with certainty what its value is relative to X or Y.

Here it seems that the problem is much deeper than in the previous case. We must remember that Ramban taught us that X < Y, whereas the data here imply that Z is greater than Y and less than X. The problem here is not in the collision between the principles but on a more fundamental plane: the basic prices themselves do not satisfy any coherent order relation. It appears, at first glance, that they cannot be arithmetized at all.

A Possible Solution

In order to solve the problem, we must go deeper and examine the hidden assumptions in the procedure we proposed. We have already seen that the relation between the price of fulfilling a positive commandment or avoiding a negative one and the price of their opposites is not a simple arithmetic inversion. We must search for a price function that satisfies all the correct relations, and from it derive the criterion that will resolve this paradox.

The continuation of the discussion requires a more complex formulation, and therefore we will stop here. The interested reader may find it, from a somewhat different angle, in the ninth gate of the book Shtei Agalot Ve-Kadur Pore’ach and also in note 29 there.

Footnotes