Kiddushin, Chapter 2, 5783, Lesson 9

[Rabbi Michael Abraham] In the previous lecture we talked about the framework of the sugya, when we look for a source for the law of agency, and later the Talmud brings several sources: terumah, divorce, consecrated matters, and so on. What exactly is the assumption underlying this search? Meaning, what would have happened if we had not found a source? Is the concept of agency a novelty of the Torah, such that without finding a source it would not exist? Or would the concept of agency still have existed, just perhaps in a different form—we wouldn’t know its details exactly, or something like that? And from there you can also ask another question: what happens after we find the sources? We found several sources for the law of agency, and the Talmud makes a tzrikhuta between them. Now the question is, after we’ve found the sources and made the tzrikhuta, is the concept of agency one concept? Right, was there some kind of ladder here that we used to climb the tree and then threw away the ladder once we no longer needed it? Or not—do we have several distinct laws of agency, and each kind of agency—say, in terumah, in kiddushin, in divorce, in slaughtering the Passover offering, and so on—each such concept is its own concept of agency. Whatever can be learned from this one, we learn from this one; whatever can be learned from that one, we learn from that one; but it never turns into one single overarching concept. And I brought an example of this, from the four primary categories of damages in tractate Bava Kamma, where they discuss the question whether the common denominator—right, at the end of the Mishnah in Bava Kamma—ultimately turns all those primary categories into examples from which I derive one inclusive rule, that anything that is your property and whose guarding is your responsibility makes you liable to pay. Or is it really the case that there are four primary categories, and every derivative must be examined to see whether it can be learned from one of the primaries, or from two primaries, and then you make it liable, and if not, then not. The practical difference would be, what they call there, with respect to their liabilities. Meaning, let’s say I have a derivative that resembles none of the primaries, but it is still my property and its safeguarding is my responsibility—in other words, it meets that general criterion. If this is a general law, then now I’m liable for that as well; I don’t need it to resemble one of the primaries specifically. And vice versa: since it doesn’t resemble any of the primaries, it could be that I cannot learn from the primaries any of the exemptions with respect to it—for example, tooth and foot are exempt in the public domain, a pit is exempt regarding vessels, an ox’s goring is liable only for half damages in the case of an innocuous ox. So none of those things would apply to this derivative, because if it doesn’t resemble any of the primaries, you can’t learn any of the exemptions with respect to it.

[Speaker C] How do I know the basic liability itself?

[Rabbi Michael Abraham] Because this is a general rule, I don’t need to derive it from the primary categories. I know that after I used all the primary categories, made the necessary distinctions, and then set them aside, I already know that anything that involves my property and whose safeguarding is my responsibility, I’m obligated to pay for. So similarly, I want to ask here too. In the simple understanding, it seems both in the Talmud and in the commentators, among the medieval authorities (Rishonim) and the later authorities (Acharonim), that the concept of agency is one concept. Meaning, people don’t make distinctions here between different concepts of agency, and in every context that comes up we don’t suddenly stop and ask, wait, is there agency here? Is this like terumah? Like divorce? Like betrothal? Nowhere do they do that. Meaning, once we say that there is agency, we have all the primary categories, we made the necessary distinctions, we set the primary categories aside, that’s it, we forgot them. Now we basically have the general concept of agency, and it applies everywhere. There are certain specific things that we do learn from the primary categories. For example, in terumah: “so shall you also set apart,” including your agent—just as you are members of the covenant, so too your agent must be a member of the covenant, and so on. So for example, that rule is accepted as applying to all forms of agency, even though it is learned only from terumah. Meaning, what we learn from the primary categories does matter, but the significance is not only for the derivative that comes from that particular primary category. Otherwise I would say it is relevant only to those forms of agency learned from terumah. But if there is a form of agency in divorce or betrothal, would a non-Jew perhaps also be able to serve there? And I said that this does not seem to be the implication, although later we’ll see that it may be that in betrothal and divorce, what disqualifies a non-Jew is only that he is not subject to the relevant legal framework, but there is no intrinsic disqualification. And then it really could turn out that this is divided into different types of agency: forms of agency learned from terumah, where a non-Jew is disqualified in an essential way even if in principle he does belong to the area in question—because we’ll still see that regarding terumah too, a non-Jew does belong in one way or another to terumah; the Talmud discusses this later on. Whereas in bills of divorce and betrothal there is no intrinsic disqualification of a non-Jew, only that he is not subject to that legal framework, so he can’t be an agent for divorce and betrothal. So that’s what we saw last time. Now before we continue, I started dealing a bit with the first source brought by the Talmud, which is a source for agency in divorce. We discussed whether the Talmud is looking for a source for agency in betrothal or a source for agency in general; this is connected to the previous discussion. In any case, the first source brought is a source for agency in divorce. Now there, yes, maybe one more sentence from what I said last time, because it’ll be important here too. The question is: even if sources are brought for the law of agency, there is still room to discuss what their relation is to reason. I said that sources—certainly sources of this kind, not things the Sadducees admit to, namely things explicitly written in the Torah—sources of this kind, it’s pretty clear that there is reasoning alongside them. Meaning, if there were no reasoning, we wouldn’t derive from “and he sent” and “and she sent” and so on that we are talking here about appointing an agent. That is also not the plain meaning of the verse, as I already noted. So clearly there is reasoning in the background. Then why do you need a source, an exposition? So I said there are situations where, despite the fact that there is reasoning, you still need an exposition. There are several scenarios one can present here. Either the reasoning is not unequivocal—you can say yes and you can say no—so I need the verse to tell me that this is reliable reasoning, that I can rely on it and use it. Sometimes there is a situation where even though the reasoning may exist, some halakhic principle prevents using it unless I have a source that permits it. I mentioned “a son and not a daughter,” regarding the stubborn and rebellious son and others, and one witness in place of two witnesses in all areas where two witnesses are required. Therefore the fact that a source is needed does not mean there is no reasoning, and the fact that there is reasoning does not mean a source is unnecessary. Usually the idea is, again, that if the source is something the Sadducees admit to or something explicitly written in the Torah, then at least no interpreter’s reasoning is involved in formulating the law. It says, “You shall not plow with an ox and a donkey together”—you don’t need reasoning there because the Torah says so. That doesn’t mean there is no reasoning; I’m saying you don’t need reasoning. Reasoning was not involved in formulating the law—it is written in the Torah. But in every place where the law comes out of exposition, where it is not written explicitly in the Torah, or through some type of interpretation or exposition or something like that, clearly there is some kind of reasoning in the background. Sometimes the reasoning is of the “least implausible” kind, as I gave the example of “You shall fear the Lord your God”—to include Torah scholars—but never mind, there is some dimension of reasoning there.

Now if I go back to our source in divorce, it’s the same thing. In our source in divorce, we derive the law of agency from verses, from exposition, and it is an exposition, so clearly in the background there is also reasoning—that what a person does, apparently he can also do through an agent. There is some combination here of reasoning and the expositions. Now if there really is such reasoning, and if there really is a source that agency can also be effective in divorce, we can ask ourselves why three different expositions are needed: for an agent of receipt, an agent of delivery, and an agent appointing another agent. Right? There are three derivations brought in our Talmudic passage: “and he sent” teaches that he appoints an agent, “and he sent” and “and she sent” teaches that she appoints an agent, and “and he sent” and “she sent” teaches that an agent appoints another agent. Okay, why do you need three sources for these three kinds of agents? And here at least I assume—I’m not sure it’s correct, but I assume—that these really are sources, and not just some mnemonic support or I don’t know what kind of wordplay, but actual sources. By the way, I’m not entirely sure that’s right. In these sources where there’s an extra letter or conjunction or something, I’m not entirely sure that we should really treat that as a true source, meaning that without it I would not say that law. But at least from some of the medieval authorities (Rishonim), one can get the impression that yes, that is the situation. And then the question arises: why is this needed? Why do we need three different expositions for an agent of delivery, an agent of receipt, and an agent who appoints another agent? What? Aren’t they similar? Why aren’t they similar? Between the last one and the… an agent appointing an agent, versus the last one—you didn’t understand the first one, now understand it—what do you mean you didn’t understand? From the standpoint of reasoning. Meaning, there is a division between an agent appointing an agent and the husband or wife appointing an agent. Say the husband and wife, and the husband and wife. Okay, he betroths, she is betrothed. Okay, so you’re basically saying there are certain considerations that could distinguish them, or primary objections—you’re basically making a necessary distinction among these three sources. Now the Talmud does not make such a necessary distinction. The Talmud makes a necessary distinction between the sources brought for agency in general—divorce, terumah, betrothal, all these things. The Talmud does not make a necessary distinction among these three expositions regarding agency in divorce. And that is an interesting question—why? Perhaps as far as the Talmud is concerned—that is, it is unlikely the Talmud thought no distinction was needed, because if no distinction is needed then you don’t need three sources. On the contrary, the idea is that these three things are so different that there is no need at all to make a necessary distinction. Meaning, no one would have thought that one of them would suffice to derive the other two. Unlike terumah, betrothal, agency, and so on, where the Talmud really does need to find the necessary distinction, because otherwise it would not be clear why three sources are needed. What is the difference? There is a fundamental difference between the different sources in the Talmud among which the Talmud makes a necessary distinction, and the three sources in divorce. In the contexts among which the Talmud makes that distinction, the type of agency is, in the simple understanding, the same type of agency—just in a different place, a different context, a different legal area. This is terumah, this is betrothal, this is divorce, this is slaughtering the Passover offering, and the like. So it is the same mechanism, the same type of agency. No matter how we understand agency—you can argue about that: power of attorney, extended hand, act, power, whatever—it is the same kind of agency in different halakhic contexts. There it is really clear that without the necessary distinctions, the question would arise why different sources are needed. Meaning, fine, once I learned the concept of agency—that someone can act for me to do something—it is reasonable to assume that this is also true in other halakhic contexts, so the Talmud makes the necessary distinctions. In the case of divorce, we are talking about the same halakhic context—it is agency for divorce—but apparently the types of agency are different. And when the types of agency are different, the Talmud does not make the necessary distinction at all. Why? Because when the types of agency are different, it is obvious that one cannot be derived from the other. Therefore there, it is not because they are similar. On the contrary: because they are so different, there is no need to make the necessary distinction. Everyone understands on his own what the differences are between them and why one cannot be learned from the other. Therefore I think that here the Talmud does not make a necessary distinction among these three types.

Actually, in Tosafot HaRosh here, Tosafot HaRosh really notes this and says: “and she sent” teaches that she appoints an agent. Yes, you have to get used to this—often you can learn quite a lot from what the Talmud does not do, not only from what it does do. Why doesn’t the Talmud make a necessary distinction here, when it does make one among the other areas? That already raises an important point, and I think you can also see it in Tosafot HaRosh, at least by hint. He says: “and she sent” teaches that she appoints an agent. “And if you say, why are two verses needed, one for the husband’s agency and one for the woman’s agency? Why are two sources needed? And one may say that I would have thought that agency applies to the husband, who divorces only by his own will, but a woman, who is divorced even against her will, her act is not so strong that she could appoint an agent in her place.” So Tosafot HaRosh says that if I had a source teaching about an agent on the husband’s side, I would not learn from that the agent of the woman. Why? Because the woman can be divorced against her will. So what? So she is passive, and therefore what? Once she has no control over the act—we discussed, if you remember, we discussed Tosafot in Ketubot, where it talks about an agent for halitzah, and Tosafot says there that whatever cannot be made conditional cannot be done through an agent. So Rabbi Shimon Shkop explains there: what is the idea? What is the connection between a condition and agency? The question is: what degree of control do you have over the act? Meaning, if you have absolute control over the act, you can make conditions in it. That means it is entirely in your hands—you can do it, you can refrain from doing it, you can qualify the doing of it—so you can also appoint an agent. Meaning, if you have control over the act, it is in your hands, then you can appoint an agent. So that is probably what underlies the Rosh’s point here. The Rosh says that the husband is the one in control—here I’m talking about divorce, note that. The husband is in control of the act; he can divorce her against her will, he is the sole master of the matter. So that he can appoint an agent, that I understand. If I had a source showing that the husband can appoint an agent of delivery, I would know it about the husband, but I would not derive it from there for the woman. Because the woman is divorced against her will, she has no control whatsoever over the act. Who says she can appoint an agent for this act?

But there are two comments here. One comment—the obvious one, and for some reason Tosafot HaRosh ignores it—is: reverse it. Learn that a woman appoints an agent, and from that derive all the more so that the husband appoints an agent. An agent of receipt, and all the more so an agent of delivery. Why not? And I think this fits with another comment here. That same distinction can also go in the opposite direction. The fact that the woman is not really a participant in the act, that she does not take an active part in it, could also lead us to say that it is easier to do it through an agent, not harder. Tosafot HaRosh says: because she has no control, I would think she cannot appoint an agent; only the husband can appoint an agent. But on the other hand, the fact that she has no control basically means she has no meaningful role in the process. Someone with no meaningful role—let her send an agent. Why should I care? The person who really is the essence of the act—that is the one about whom you need a novelty to say he too can do it through someone else, can send an agent to act for him. But the woman, who in any case functions here in a passive role, can do this through someone else. Why should I care? The main thing is that someone represents her and receives the bill of divorce on her behalf. Therefore the very same distinction—and by the way, this happens quite a lot—the very same distinction Tosafot HaRosh makes can work in both directions. He says that if there were a source for the husband’s agency, I would not derive from it the wife’s agency; but also, if there were a source for the wife’s agency, I would not derive from it the husband’s agency. And therefore this resolves the first comment. So it is not difficult here to say: fine, then write it only regarding the woman, and we’ll derive the agent of delivery by an a fortiori argument. Because this distinction Tosafot HaRosh makes between an agent of delivery and an agent of receipt works in both directions. On one hand it gives an advantage to the husband’s agent—it is easier for the husband, easier for us to say that the husband can appoint an agent than that the woman can. As if the husband is the lenient case and the woman the stringent one. But by the same token, the woman is the lenient case and the husband the stringent one. Because the woman has no meaningful role, so why should I care—let it be done through a representative. But the husband—it might have been said that he must divorce the woman with his own hands, and so on. Who says it can be done through an agent at all? Therefore we need a source that an agent of delivery is also effective. And as I said, in these necessary distinctions, it does not bother me that the same reasoning itself works in both roles, as long as there really is—I don’t offset leniencies and stringencies against each other, that is a known rule. There are very rare places where the Talmud does make such offsets. For example, collecting principal without an oath is preferable to collecting double payment with an oath, in Bava Metzia. There they really do make offsets, and it is a major question how to understand the Talmud there. The Talmud speaks there about the liabilities of a bailee—who is more stringent, one who falsely claims theft, never mind, a paid bailee, an unpaid bailee. But usually we do not make offsets. So the fact that this reasoning has a lenient side in favor of the husband and a lenient side in favor of the wife does not mean, okay, let’s see in the end which side is more significant and decide which of them is more stringent. No. If there is one side of stringency here and one side of stringency there, that is a two-sided objection. Meaning, we need two sources. Because it is enough that there is one aspect in which this one is more stringent and another aspect in which that one is more stringent for it to be a full necessary distinction. Thus one argument can function as a full necessary distinction in both directions; one distinction or one line of reasoning can function as a full necessary distinction in both directions.

What is harder for me here is how the woman can at all—even in the final conclusion—how do we learn this? After all, apparently this is not connected to the laws of agency. Is this even an agent? Who says this is agency? I mean, a woman is a flowerpot. What? Why do you need anything here at all? We’ll talk about that in another moment. No, but it’s really connected, very much so, and I’ll get to it in a moment. She can receive—not exactly a flowerpot; suddenly she says no. What? She won’t receive? Yes, what do you mean? A flowerpot can receive—put it on the flowerpot. What is that called? A courtyard. You put it in the courtyard. A courtyard is… There really was room here to say that there are two sides. There was a possibility to say that… no, no, fine, I just disconnect it from time to time because when we’re talking I don’t want the words to stay in front of your eyes over there afterward—it splits attention. Two comments regarding this matter of the woman’s agent of receipt. One comment—and again, it’s two sides, lenient and stringent. One comment is that since this thing is some kind of monkey-act, then who says agency is even needed here at all? All you need… yes, it’s a bit similar to what the Talmud in Bava Metzia says about a courtyard. A courtyard serves as the owner’s agent for certain matters. Now there is room to say that this is not really an agent in the sense of the laws of agency—“so shall you also set apart,” and so on—but okay, the courtyard can function as my monkey. Meaning, there are things that do not really require some sort of, I don’t know, legal competence, decisions, judgment, I don’t know exactly what. So fine, a courtyard can also do that. So this raises, again, two opposite directions. On one hand, why do we need any novelty here at all regarding an agent of receipt? Why does the agent of receipt have to be an agent? Fine, so he receives on behalf of the woman, that’s it. But he needs to receive from the hand of the woman. Yes, fine, yes, because that counts as though she received it. It is a monkey-act, okay? Why a monkey-act? Why? Because the bill of divorce needs to be in her hand. The novelty in… so you say: no, this is still a monkey-act, but you need a monkey. Meaning—and this brings me to something I’ll define more clearly in a moment. There is a big difference between the woman knowing that she is divorced the moment the agent receives the bill of divorce, and her only being divorced when it reaches her and he… No, no, he received it on her behalf, like a courtyard, say. Then the moment it is in his hand, she is divorced. Otherwise he is what is called an agent of delivery and not an agent of receipt. This is a concept that the medieval authorities (Rishonim) introduce here: an agent of delivery, where the woman sends an agent but she is divorced only when the bill reaches her, even though he is her agent, not the husband’s agent. But let’s leave that aside for the moment.

The second comment—and I’ll come back to it in a moment—is the opposite comment. If what is really required is that the woman receive the bill of divorce, then it may be that an agent does not help here. Not that the agent of receipt has to be an agent—because, as with putting on tefillin, as we saw, things where the person himself is required to act with his own body, this is not some legal act in which you create a legal effect, make some decision, or something like that. Rather, you simply need to be the flowerpot. You are the flowerpot in this case. If you are the flowerpot in this case, then you need to be there. Your agent cannot be in your place, because this is not a legal act but a physical act. And in a physical act, you specifically are required and not someone else. So again, the same consideration can work in both directions, either in favor of agency of receipt or against agency of receipt. And in this matter, really in the context of… maybe before I go on, in the context of betrothal. In the context of betrothal, the woman does have some part in the action; the woman has to consent. You cannot betroth a woman against her will. On the other hand, yes, the Avnei Miluim speaks about this in a number of places, on the other hand it is clear that the woman does not take part in the act. She has to consent. And there too one can ask what happens regarding an agent of receipt of a woman in betrothal, not in divorce. Again, this can go in both directions, as you can see that almost every argument here can go in both directions. On one hand, it is clearer that agency belongs here because the woman does take some part in the action, unlike in divorce where the woman is a flowerpot. On the other hand, it is not certain that agency is needed at all, because in the context of betrothal, in the simple understanding, it would seem that the woman’s whole role is just to consent—to express her will. She does not need to take part in the action; she needs to agree that the action be done. Fine, obviously she has to agree even if she does it by an agent; she still has to agree. Okay? Once she agreed, why should I care whom she sends to receive the ring or the money, whatever? Basically it’s an act… unless we say that in betrothal too there is some law of “and he shall place it in her hand,” as in divorce, by way of the analogy between “she goes out” and “she becomes.” That the betrothal, at least in betrothal by money or document—for intercourse it works a bit differently—must be placed in the woman’s hand. Some sort of analogy between “she goes out” and “she becomes,” meaning some law of placing it in her hand. Otherwise it is not clear why an agent is needed for betrothal at all. Let her express consent; that is, she has to agree. Once she agreed, what difference does it make now? So what difference does it make? Clearly something else is needed. What, she just announced that she agrees? And the husband performed the action. No, and that is not the action. Why is that not the action? Because what—he gave money to someone? So that brings us back to the first side: then she is a flowerpot. You keep going back to that side. So we went back to the flowerpot. No, I think there are two sides, both of them exist, but both exist. It keeps functioning as two opposite sides. I only showed one side. No, I think it is two sides, and they come together. I agree—just don’t offset them against each other. They both exist. I think the point is this: I think an agent of receipt, in the terminology of the later authorities we saw, is an agent for the act. And an agent of delivery is an agent for power. Let me remind you again: an agent for power is a case where I transfer to someone the ability to bring about a legal result in my name. Instead of my doing it, someone else does it. Therefore it is natural to connect this to the conception of power of attorney and not extended hand, although I made distinctions between them. Agency for an act is agency that says that the one considered as performing the act is me. Say in agency for a transgression—a murder agent according to Shammai the Elder, yes, that was one of the examples I gave. There, according to the view that there is agency for a transgression, if I send an agent to murder, I am considered the one who murdered. Now clearly that is not agency for power, because the murderer does not produce any legal result. He is not creating a legal effect; this is not ownership, not terumah; he did not do anything of that kind. He simply performed a physical action—he killed someone, took someone’s life. Everything the concept of agency is relevant for there is only to say that I, the sender, am considered as the one who did the action. I did not transfer to him some power that was in my hands and not in his and delegate it to him so now he can do it; he could have murdered even without me, he doesn’t need my permission to do that. Therefore this is not some power that exists exclusively in me and that I delegate to someone else. Rather it is only a technical question: if he did it as my agent, is that considered that I did it? The practical difference is whether I am considered a murderer. Okay? So there it is plainly agency for an act and not agency for power.

In the ordinary forms of agency—for acquisition, betrothal, and the like—it is agency for power. I authorize someone to perform an act that in principle he could not have done. For example, acquire something on my behalf: he can acquire something for himself, he cannot acquire something for me; only I can do that. Unless I appointed him as an agent—maybe by the rule of benefiting a person in his absence, never mind. But in principle there needs to be a mechanism of agency that transfers to him the power to produce the practical result—to produce in the sense of legal effect, yes, a legal effect is something that comes about. Okay? So that is agency for power. Now in the simple understanding, the difference between an agent of receipt and an agent of delivery—and I think this is true both in betrothal and in divorce. It is true in both. Even though there is a difference in the woman’s role between betrothal and divorce, that difference is not important for the laws of agency. It is not important because in betrothal all the woman needs is to consent. The agent does not consent on her behalf. She does that in any case. So why is an agent needed? We need an agent so that he will be her flowerpot, that he will be the flowerpot on her behalf, just as in divorce. In that sense it is like divorce. I don’t care that consent is needed, because the consent is given by the woman in any case. Everything beyond that is completely similar to divorce. Even that won’t help in divorce, because agency for an act won’t help either. Why? Because she does nothing, and what about… No, once you say it is agency for an act, then obviously a person on whom an act is done can be considered as though the act was done on the woman, on the sender. What difference does it make? If you say you view him as her. If I need to circumcise my son, then he can do it and I… No, no, there you are moving into another distinction I made in the previous lecture. That is the distinction between a physical act and a legal act. Even in the legal world there is a difference between the question of who did the act legally—who did the act—and the question of who brought about the legal effect. That is agency for power versus agency for an act. In a physical action, the question there is who physically did the action; agency does not help for that. So I am saying that even in the legal world there is a distinction between the act and the result—act in the legal sense. What? Is receiving not an action? You have to do a physical action, but on the legal level I ask myself who it is that on the legal level performed the action, and the law of agency can answer that. In the context of commandments, what is required from me is the performance of the physical action itself, so that is not legal; therefore the law of agency does not help there. So even when I speak about the act, there is a distinction between the legal sphere and the physical sphere. I understand, but I still don’t understand how it would work with a bill of divorce. So I am saying: with a bill of divorce, the question is not who physically received the bill of divorce. The one who physically receives the bill of divorce is effective because legally he is considered as the one who performed, or on whom was performed, the act of giving the bill of divorce. And if on the legal level that is what is required—and not actually putting on tefillin as in commandments, where I am physically required to put on tefillin, so agency won’t help there. But here a physical action is required—why? Because I want the legal action to be attributed, on the legal plane, to the woman. Not the result—the act, in legal perspective, is considered the woman’s act. And that is the concept of agency. No, but one could look at it differently and say no, it’s the same thing. Obviously one could. That’s why a source is needed. That’s what I’m talking about. Ah, if so, then the source is not a source teaching me something about the laws of agency; it teaches me something about the laws of divorce. Either way, you can say it this way or that way. After all, the question after there is an initial assumption is what was renewed. Was what was renewed that agency applies to this too, or was what was renewed that divorce is the sort of thing to which agency applies, an agency I already knew beforehand? Learn it whichever way you want; I don’t care. What practical difference does it make? No practical difference, but fine, I have no way to decide it. So you can learn it this way or that way; I don’t know.

The claim, in the end, is that there really is a different kind of agency here. Wait, but with a bill of divorce for example, to me it’s the same type of agency; you just discovered for me that with a bill of divorce he doesn’t need… No, if you reach the conclusion that there is agency for an act, you are right. But how do you know there is agency for an act? Ah, that is learned only from a bill of divorce? There is no other source. And that is exactly the point. Rabbi Shimon Shkop himself notes this. He says: we find in the Sages that there is agency for power and agency for an act; that we do see. But where do they get that from? After all, the verses brought about agency do not make such a distinction. I’m offering a suggestion now: here is where it comes from. It comes from here, yes. This is the necessary distinction; this is really the depth of the necessary distinction in Tosafot HaRosh. Why do we need a source for agency of receipt and a source for agency of delivery? This is the source Rabbi Shimon Shkop was looking for. That necessary distinction, which the Talmud doesn’t even make because it’s so obvious, as I said before—that it is self-evident that if I had one, I would not derive the other from it and vice versa. Because these are really different kinds of agency, not agency in different halakhic contexts. And now we can give them names. These are different kinds of agency: one is agency for power and one is agency for an act, and therefore a necessary distinction is needed. So if so, we learn two things from this source. One, that agency for an act is effective; and two, that the woman’s role in divorce is indeed a role. Because I could have said she has the role of a flowerpot. Yes, yes, right. It is not the role of a flowerpot, because if it were the role of a flowerpot she could not appoint an agent for it. No, no, no—a legal flowerpot, a legal flowerpot. Not… yes, obviously. That’s heaven and earth apart. No, no, heaven and earth, right. If we say there is agency of receipt in divorce, then necessarily two things were renewed here: that there is agency for an act, and that divorce is the kind of thing to which agency for an act is relevant. Now it’s clear. Now the question is: if I had another source for agency for an act—if it had emerged from some other source, I don’t know exactly, if Rabbi Shimon Shkop had found some source—then I would say, okay, I would answer your earlier question. Then perhaps what was renewed here is the definition of an agent of receipt in divorce, not a definition of agency, not a novelty in the laws of agency. But since Rabbi Shimon Shkop himself says, I found no source for this, then I say, fine, this source really says both things. Because without both things there would be no agency of receipt. So it is not that there is a problem of how you derive two novelties from one source. The law this source states requires both novelties in order to be stated, and therefore it is clear that it renews both of them. Fine? There isn’t the problem here of how you derive two novelties from one source. So yes.

But is agency of receipt also not different from an agent for betrothal and divorce? Is there no difference between them? Again? Is there a difference in an agent of receipt between betrothal and divorce? At first glance, no. At first glance, no, because as I said earlier, the difference between betrothal and divorce is that in betrothal the woman has to consent, but in the act of betrothal she contributes no part at all. That is how the Avnei Miluim defines it: she nullifies herself and allows the husband to take her, basically. She does not… she does not do anything, she does not contribute any part to the act of betrothal. The act of betrothal is performed only by the husband. Therefore it says, “when a man takes a woman,” and not when a woman is taken to a man, okay? The Talmud derives from that that the action has to be performed from the side of the husband. And that is the whole problem seen today with exchanging rings, where the woman gives the husband a ring—today many couples want to do this, symmetrical betrothal. That is problematic from the standpoint of Jewish law, and one has to make sure they do it afterward or in a somewhat different way, so there won’t be problems, because otherwise there is a problem with the very conception of betrothal itself. In any case, the difference is basically the question whether consent is required. The woman can refuse betrothal. But in terms of the required action—after all, the woman gives her consent in betrothal, and even if she does it by an agent, obviously she has to agree. Okay, so that is not something the agent does on her behalf. So what does he do on her behalf? He physically receives the betrothal on her behalf. In that sense it is like divorce. What difference is there? Once there is already consent, there is no difference. What one could perhaps say is that if there were a source for agency of receipt in betrothal, perhaps I still could not derive from that agency of receipt in divorce. Because in betrothal, since the woman has more control over the act, I might say that that is why she can also appoint an agent, like the line of reasoning we saw before. In divorce, where the woman is entirely without control—it can be done against her will—I might say: fine, so who says that there she can appoint an agent at all? And so it is no accident that the Torah chooses to teach us the law of agency in general, and agency of receipt in particular, in divorce and not in betrothal. Because if it had taught it in betrothal, then at least regarding an agent of receipt, it is not certain that it could have been derived for divorce. Okay?

Good. So for our purposes, we basically understood in these three sources—the agent of receipt, the agent of delivery, and an agent appointing another agent—why the first two are needed. An agent of receipt and an agent of delivery are basically an agent for an act and an agent for power. Okay, and this also solves the problem of the source Rabbi Shimon Shkop was bothered by. What about an agent appointing another agent? Yes, that is the third source. Now here, I think this does not even require overly sophisticated definitions to understand that this is really something else. Meaning, the agent appointed by me is not really the owner of the act. True, legally maybe power of attorney, I don’t know exactly what, in some sense he is considered the owner of the act. But surely there was room to say that he cannot appoint an agent in his own name—he is not the master of the matter. Meaning, for what he was appointed he was appointed, and for what he wasn’t, he wasn’t. That is, he cannot appoint an agent to do it in his place, because it is not he who does it; even when he does the act, it is not really he. Unless we say he is an extended hand. Exactly. There really was room here to distinguish between the two things, and I would actually divide it the other way around. I would say that specifically the conception of power of attorney allows an agent appointing another agent more than the conception of extended hand does. Because the conception of extended hand basically says: look, this agent is just my long stick. He has no independent standing. Now when he decides on his own initiative—I’m not now addressing the question what happens if I appointed him to appoint an agent; we’ll get to that soon—but if he suddenly decides on his own initiative to appoint an agent, who is he? He’s a broom. Earlier we talked about a flowerpot, now it’s a broom—we’ve brought the whole inanimate world into play here. Meaning, if it’s an extended hand, then the true actor is the sender. He is just being used as an extension. Specifically there I would say there is less room to say the agent can appoint an agent. I say, after the Torah renews it then fine, but you need a novelty. According to the conception of power of attorney, there would be more room to say: if the agent can separate terumah on my behalf, then he is basically the master of the matter. And once he is the master of the matter, why shouldn’t he also be able to appoint an agent to do it? If he is the one divorcing, the one betrothing, the one acquiring, I don’t know exactly what, then let him appoint an agent just like anyone else who is the owner of the act. I actually think power of attorney more. I’m saying, this doesn’t mean that according to the conception of power of attorney one would not need a verse at all to renew this. You do need a verse to renew this. And not only that—I’d even say that after the verse renewed it, the proponent of the power-of-attorney conception, such as Maimonides, might say: perhaps from this very verse I learned that agency is power of attorney, because the proof is that the agent can appoint an agent. If agency were just an extended hand, then maybe indeed an agent could not appoint an agent. Maybe. I mean, this is not an objection to the Tur, who says it is an extended hand, but I’m saying it could be.

And what is the law if I make a condition with the agent that he not appoint an agent? Is it still effective if he does? Of course not. Why on earth would it be? Then he was not sent for that. What do you mean? If that’s the situation, then you infer from it that everything he does in appointing an agent is by my power. Correct. So really the whole question is… no, not by my power; only the question is whether I object. We’ll get to that in a moment. Maybe it is enough that I don’t object; perhaps I don’t need to appoint him for that, I just need not to care. We’ll get to that, that is in a minute. No, but even without that, I say by reason, after all I have the power to appoint an agent. No, but even without that, I say by reason, after all I have the power to appoint an agent—that is obvious. But who says that this person whom I appointed as an agent has that power? The only way to justify that is to say that I transferred to him my power to appoint an agent. Not necessarily. Even though… because if he is now the one divorcing, then one who divorces can appoint an agent. No, he received from me… the fact that I can appoint an agent was a novelty of the Torah. Right. Now he received from me only the right to divorce, not the right to appoint an agent. The question is whether I gave him that too. No—if he is now standing in your place and now he is the one divorcing, let’s say in a full power-of-attorney sense, if we take that all the way, then he is now the one divorcing. If he is the one divorcing, the Torah renewed that one who divorces can appoint an agent. So according to that, a condition would not help—that is what I’m saying. What do you mean? According to that, it would not help for me to say to him: I appoint you on condition that you cannot appoint an agent. Of course it would help. I did not appoint you to divorce on that basis—what do you mean? The meaning is that your appointment is qualified; my appointment determines in what respect you stand in my place. It is not that now, on your own initiative, you decide. True—for divorce you stand in my place. But if I say, no, for divorce yes, but without an agent—because if you do it by an agent, then simply no, then I am not appointing you to stand in my place. I made that conditional. Yes, fine. It is like “I sent you to improve matters, not to distort them”; it’s the same kind of condition. Yes, an implied condition—you don’t need to state it.

Good. So if you see it that way, then the whole question is basically this: when someone appoints an agent in an unspecified way, what is implicit here? Okay. No, that is why I told you earlier that I am not sure. And we’ll get back to this later, because it may be enough that the person does not object. If I appoint him to be the one divorcing in my place, then in principle he can appoint an agent. If I do not object, I did not give him the power to appoint an agent, but I do not object, no problem. If I object, then there is an implied condition and he cannot do it. That is one possibility. A second possibility is no—that by appointing him to be the one divorcing, I thereby also gave him the possibility of appointing an agent, something I also had. These are two different definitions. For example, regarding an agent appointing an agent, there is already a dispute among the medieval authorities (Rishonim), though the later authorities define it more, but it already starts among the medieval authorities, as to whose agent the second agent is. Is he the agent of the first agent, or the agent of the original sender? A practical difference would be if the first agent dies. If the first agent dies, does the second agent’s agency lapse, because he is really the agent of the first agent? Or not—once he has been appointed, I have left the picture and now he is the agent of the sender, so what difference does it make if I die on the way? I only appointed him. So maybe we’ll touch on this later too, okay? If there is the person, and the first agent, and the second agent—can the first agent add a condition to the second agent? Why not? Anything he can do, he can make conditional, yes. No, that is not the practical difference. The first agent can condition the appointment of the second in any case according to both conceptions, because in the end, even if the second agent is the sender’s agent, the one who appointed him was the first agent. So if he appointed him, he can also appoint him with a condition.

Okay, so this perspective explains why a source is needed—we spoke before about the source for an agent of delivery and a source for an agent of receipt. Now I’m asking: why is a source needed for an agent appointing an agent? Well, this already seems self-evident, right? Meaning, that an agent can appoint an agent is not trivial at all, and therefore a source is needed saying that yes, he can. This enters the topic of verbal matters, which we’ll soon touch on a bit—that verbal matters are not entrusted to an agent—but there is a problem when an agent appoints an agent. It may be that even after the novelty that an agent can appoint an agent, it is still not simple, and not everywhere can an agent really appoint an agent. There are places where even according to the final law he cannot appoint an agent.

Now perhaps I’ll finish with a doubt that Rav Shmuel raises here. You need to note there—what? Yes. You need to note there. Oh, I forgot to switch it, it’s just here all the time, not good that I leave it up all the time. You could project a landscape picture there. That would split attention even more. Right, nice, but that’s exactly why they put a landscape picture there. He put a black picture here in parallel. That’s the interesting philosophical question—what does a blind person see? Does a blind person see black? Or does he not see? That is not the same thing. To see black is to see something. There is actually an answer to that question, because there are people who became blind during their lives and people who gained sight during their lives, so you can ask them. Yes, I’m not saying there is no answer; I don’t know, but it is an interesting question. You’re saying it’s not a philosophical question, it’s a scientific one. Okay. Scientific more or less. I mean, if you rely on subjective reports of people, then yes, it is a scientific question. Psychology is usually like that. The research is basically done by a person’s report. The philosophical question is whether we see black at all. Huh? The philosophical question is whether we see black at all. Why? Because maybe black is what I don’t see, what I don’t perceive. No, black is really seeing black, it seems obvious to me, like seeing the black here. No, maybe it means there is a section there that you don’t see, don’t perceive, and you interpret that as seeing black. Basically we are blind in those places. Everyone is a bit blind wherever he sees black. I’m not convinced. It seems to me that’s not it. I think I see black, I don’t know. It’s like the point—we once had some argument with Rabbi Shilat, he wrote some critique about something, I already don’t remember what. What is the length of a point? So he said that a point has zero length and infinitely many points still produce something that has length—yes, that’s measure theory. It has some specific length. And I said that’s not right. A point’s length is not zero; a point has no length. It’s not zero length. Zero length is an infinitesimal—yes, a segment as small as you like, it has zero length. A point has no length, not that it has zero length. Length belongs to one dimension. In the real world, a point… no, no, in the real dimension there is no point. Right, but in the non-real dimension, points really don’t create a line. Only in our world they do. And all this is an abstraction intended to deal with real problems. Not at all, what do you mean? Once you define a line mathematically, it exists in the Platonic world just as much as it exists here. Of course it’s defined. It is well defined by means of an axiom. In fact, to define what a point is, one way is by means of an axiom. One of the definitions of a point is to take something and let its width approach zero. No, no. That’s Russell. No, that doesn’t define a point, what are you talking about. So what definitions of a point are there? Mathematics usually defines it as the intersection of two lines, but the intersection of two lines assumes that the line is defined. So the question is what you choose as a basic definition and what as a derived definition. A basic definition is a nice way of saying you accept it as an axiom. No, obviously, there is nothing without axioms, what do you mean? Russell tried to provide this, and his idea really was to give that definition—to take some object that has volume and let it approach zero, that has length and width and let their size approach zero. And then that thing creates a line. I don’t know it, but it sounds very hard for me to believe. Hard for me to believe. You’re talking about infinitesimals. Infinitesimals are not Russell; that’s Newton and Leibniz. But the idea was to define a point by means of something—you’d have to see; I don’t know it, but it’s very strange. A point is simply something else; the abstraction is wrong. Because a point has no length. And a line has length even if the length is zero—it’s still a line. One dimension. The dimension of a point is zero. The dimension of a line is one, even if its length is zero. These are creatures of completely different types; they live in different worlds; you can’t build one from the other. Right, exactly. And therefore a line is not built from a point. Fine, it is built from points with the property of density, never mind that now, but it is built from points with an added requirement. Good, that’s enough.

Okay, so for our purposes, Rav Shmuel says as follows: “It teaches that an agent appoints an agent. The later authorities were uncertain whether only in agency over the power of action in divorce does an agent appoint an agent, or whether this law also applies in agency of act. And some say that in agency of act an agent does not appoint an agent.” What does that mean for us? It means regarding the woman’s agent of receipt, yes, in divorce. Can the agent appoint an agent? What is written in our source is: he appoints an agent, it teaches that she appoints an agent, and that an agent appoints an agent. But this “an agent appoints an agent”—to what does it apply? Is it the agent of delivery, the agent of receipt, both? And all this is in divorce—what about betrothal? So that is not a simple question. There is a dispute among the medieval authorities (Rishonim), which we will see. But Rav Shmuel says: the later authorities were uncertain. They were uncertain whether this applies only in agency over the power of action, as in divorce, because the simple assumption is that the derivation from the verse here speaks first and foremost about the agent of delivery, right? Whenever it says he appoints an agent, that is the basic agency. Then there is an inclusion that she too appoints an agent, and when it says an agent appoints an agent, we have returned to the normal, regular agent—that is the agent of delivery. The regular agent is the agent of delivery. So when we say an agent appoints an agent, the meaning is that an agent of delivery appoints an agent. But an agent of receipt—who says that he appoints an agent? And this continues what I said before: that an agent of receipt is agency for an act and not agency over power. Does the agent of delivery maybe have both an element of act and an element of power? Why? He also physically handles the bill of divorce, and maybe… No, the opposite. What you are talking about is an agent for power. In betrothal? No, it is an agent for power. I’m talking about an agent of receipt. In betrothal not? Both an aspect of power and an aspect of… Fine, he also does an act, but that is not the interesting part. Once there is power, obviously agency is there. Maybe he cannot transfer the act? He cannot transfer the act? If he has the power, if he has the power, then that is what he needs. He performs the act because he has the power to perform it. The point is the opposite. In agency for an act, it is only agency for an act and not agency over power—that is agency of receipt. There the question is whether an agent appoints an agent. In agency of delivery, even if you define it somehow as containing an element of agency for an act and not only agency for power, it is obvious that an agent appoints an agent. Because once he has the power, he can transfer the power onward. But why can’t he appoint that the act be attributed to him? In agency for power, you don’t appoint regarding the act. There is no need for the act to be done on my behalf. What is needed is that you have the power to bring it about. You are not doing the act for me. You have the power to do it—go do it. This is not agency over the act. You are not functioning here as someone who… there is no need for me to be considered the one who did the act. What is needed is that the result apply to me—that I acquire, that I divorce, that I betroth, and so on. That is enough. In agency of receipt, the whole problem is that there is no dimension of power there, because the woman brings about nothing. So all that is required is only that she be considered the one who received the bill of divorce. Or in murder, that he be considered the murderer. The question is whether the concept of agency really considers the sender as the one who actually performed the act. So that is the novelty of agency for an act. But there, who says the agent can appoint another agent? This, I think, very much resembles what I said earlier to Doron—the difference between power of attorney and extended hand. It is not accidental that agency for power and agency for an act keep getting mixed together with power of attorney and extended hand. Because once you are empowered by power of attorney, there is no problem at all—everything is in your hands; you can appoint another agent to be your agent. Everything is fine. Even if in the end, halakhically, he is considered the agent of the original sender, still you can appoint him. That is obvious, since all the power is in your hands. But if you are only an agent for an act, then you do not really have the ability to make decisions here. You are not the operating factor. It is like an extended hand; really the sender is the one acting. So who are you to appoint an agent? Why on earth can you appoint an agent? We need to posit some assumption that in agency for an act an agent appoints an agent. And there are medieval authorities (Rishonim) who say that in agency for an act an agent does appoint an agent. They claim—again, I am translating this into agency for an act—they say that in agency of receipt too, an agent appoints an agent. In my terms, that means that in agency for an act they apparently argue that although the fellow is my broom, this broom is a human being. The person functions here as a broom, and once he is a broom with legal competence, meaning that he has the ability to make decisions, then he can also appoint another broom that extends this broom. That is basically what was renewed here in “an agent appoints an agent.” Then the claim is that “an agent appoints an agent” is really a dispute among the medieval authorities (Rishonim) whether it was said only about agency for power, in which case it applies to agency of delivery, or also about agency for an act, in which case it would also be renewed regarding agency of receipt. And according to the medieval authorities who say it also applies to agency of receipt, that means that “an agent appoints an agent” is an inclusion on top of the first two inclusions. It is not three inclusions all on the same plane, but one built on the other: from the agent of delivery, on top of that the agent of receipt, and on top of both of them the rule that an agent appoints an agent in both cases—both an agent of receipt and an agent of delivery. And according to the views that an agent of receipt does not appoint…

With that I finish the necessary distinctions among the three sources for the law of agency in divorce. Now I really want to enter—this is part of the continuation of the same topic—the matter of an agent appointing an agent. In the kinds of agents we have seen so far, we basically have four agents: an agent of delivery and an agent of receipt in divorce and in betrothal. Those are four types of agents. Why is that important? Because regarding each one we need to discuss whether an agent appoints an agent or not. Each of these four types of agents. So the baraita itself speaks about an agent in divorce. I mentioned a dispute among the medieval authorities whether “an agent appoints an agent” also refers to an agent of receipt in divorce. That same dispute also exists regarding betrothal, and some distinguish between betrothal and divorce—saying that in divorce yes and in betrothal no. So the whole thing starts from the issue of verbal matters. The Talmud in Gittin 66 says: Rabbi Yirmiyah bar Abba said, “They sent from the study hall of Rav to Shmuel: Let our master teach us. If one said to two people, ‘Write and give a bill of divorce to my wife,’ and they said to a scribe, and he wrote it, and they themselves signed it—what is the law?” A man says to two people: write and give a bill of divorce to my wife. Then they did not write it themselves; they told a scribe to write it, and afterward they signed it. The assumption is that “write and give” means they also need to sign as the signing witnesses; never mind right now whether that follows Rabbi Eliezer or Rabbi Meir. What is the law? Apparently they appointed the scribe as an agent to write, whereas the sender wanted them to write. It has to be by the husband’s instruction and it must be for her sake. And that’s what signing does. No—writing. Writing must be for her sake and by the husband’s instruction. Yes, that is the Mishnah in Gittin 24, where the Mishnah says that one who writes standard forms of divorce documents—it is invalid, because you can’t do that. You have to write the bill of divorce—I mean the essential text, not the standard template. You have to write the bill of divorce for the sake of the woman and by the husband’s instruction. Not a scribe writing a generic form—that is invalid if he writes a generic form. I mean, if he writes only the standard wording, the template, and inside that they fill in the woman’s name and details specifically for her sake, that is fine. That is called the standard text and the essential text. I am speaking about the essential text. So the question is: what is the law? Shmuel sent back: “She must leave, and the matter requires study.” Meaning, the woman already got divorced this way, married someone else, and now the question comes up: wait, is she really divorced or not divorced? So he says: she must leave her second husband, and the matter requires study. Out of doubt she must leave, but I am not sure—that is Shmuel’s answer.

What does “the matter requires study” mean? What was Shmuel’s doubt? If you say it is because these are verbal matters, and he was uncertain whether verbal matters can be entrusted to an agent or cannot be entrusted to an agent—didn’t Shmuel say in Rabbi’s name that the law follows Rabbi Yosei, who says that verbal matters cannot be entrusted to an agent? Meaning, the first possibility the Talmud raises is: what was Shmuel uncertain about? The question is whether verbal matters can be entrusted to an agent or not. What does that mean? We will see different explanations among the medieval authorities, but in the literal sense, verbal matters means words. The question is whether one can appoint an agent for something that is only verbal. Appoint an agent to tell the scribe to write. Agency to say something. Agency to say something is nothing; it is something without substance, mere words, and mere words are not entrusted to an agent. Why is it without substance? Words! Words are nothing. Words are a thing in which there is no act when you move from the… then there is no prohibition of slander? There is no prohibition of slander because it is words and that is nothing? The prohibition of slander is not on producing the words as an object. The prohibition of slander is exactly on the effect. The words, as he says, may ultimately cause someone to kill, so people say nowadays that words kill. I don’t like that hysteria, but obviously words can have consequences. But the prohibition of slander is a prohibition on what the words do, not on the production of the words themselves. And isn’t the point that it’s words? If it were something written, would that not be verbal matters? Wouldn’t that be… No, if it were something written, then yes, it would not be verbal matters. We’ll see in a moment. There are different definitions among the medieval authorities—we’ll see. But in the literal sense, verbal matters means words. And the literal meaning of verbal matters is that I am doing no act on which we can speak about legal effect or something like that; it is just a trigger, I cause someone else to do something. No, that is not the accepted definition—we’ll get to that. No, but it’s true even if it is not accepted—what is going on here? Why would that help? I’m asking: suppose there were no problem of verbal matters. In the plain laws of agency, would I understand how agency is effective here? After all, if I send… if there is a law that I need to tell the scribe to write, then I send you to say it in my name. There is no such law. There is only a law that I need to write, and I can send the scribe—that’s all. No, I didn’t understand. It’s very simple. There is no law that I have to tell the scribe. There is a law—and for that, agency is needed. No, it doesn’t work like that. It works that there is a law that I need to write it myself, “and he shall write for her.” Obviously. But now that I do it through a scribe, I can appoint the scribe as an agent. Okay? By the way, it’s not certain he is an agent, but never mind, we’ll see. And he appointed me to write. Right? No, he appointed them to tell the scribe to write, right? What is the problem? I also had no law requiring me to tell the scribe. Yes you did, because without that the scribe cannot write. No, fine, but if I do it through a scribe, then I need to tell him; that comes from the requirement of for-her-sake. He needs to be told to write, otherwise it is not writing for her sake if I did not tell him. But the fact that I need to tell him is not because that is my role in divorce. No—it is the role such that if I want… No—it is the role. Fine, so maybe we’ll see this later. Yes, it is the role. It turns the writing into writing for her sake. No, if someone else writes, not because of agency—that is not appointing an agent, that is exactly the point. Saying it to the scribe is not simply appointing an agent. According to some views, it is not appointing an agent at all. Some authorities say that a scribe is not my agent at all. The whole point of saying it to the scribe is only to turn his writing into writing for her sake. Because writing for her sake must be done, as you said earlier, by the husband. Because “for her sake” is my role. I can do it for the sake of my wife; nobody else can do “for her sake” in my place. So when I tell the scribe, I turn his writing into writing for her sake. So the agency is over the “for her sake.” Correct. No, so again, same issue: there is no law that I have to say it; there is only a law that there be… fine, and that is called verbal matters. And that is called verbal matters because there is nothing tangible here.

Good. And it is known that even without that it would be verbal matters, because he did not send them to him—even if there were something… No, no, whether he did or did not send them to him, that will come in a moment; here in this Talmudic passage it will come up as “say to them.” So the first possibility the Talmud raises to explain Shmuel’s doubt is the question whether verbal matters can be entrusted to an agent or not. Is there such a thing as agency where the content is basically speech, words? Okay? For now I am translating it literally; later we will see more fully what exactly verbal matters means. So that is the first possibility, and that is Shmuel’s doubt. Now, that cannot be, says the Talmud, because Shmuel said in Rabbi’s name that the law follows Rabbi Yosei, who says that verbal matters cannot be entrusted to an agent. Shmuel is not in doubt about that; he already established the law that verbal matters cannot be entrusted to an agent. Therefore there cannot be a doubt here. So what is the problem? If verbal matters cannot be entrusted to an agent, then she must leave—and that is certain, she is not divorced. “And if we hold that ‘write’ means in your own handwriting, then the writing of the bill of divorce is valid, and Shmuel said in Rabbi’s name that the law follows Rabbi Yosei, who says verbal matters cannot be entrusted to an agent.” What does this mean? The question is whether when he said to the agents “write,” he meant they should write in their own hand, or whether he meant they should appoint an agent. So the Talmud says: if he told them to write it in their own hand, and that is the problem when they handed it over to an agent, then that implies that if they had written it themselves, that would have been fine. And to that the Talmud says: but why? Didn’t we rule like Rabbi Yosei that verbal matters cannot be entrusted to an agent? If they had written it themselves, the agency they received would still be agency to write—agency involving verbal matters. So that should be verbal matters not entrusted to an agent, and it should not work. So why, if they had written it themselves… The answer: if we hold that “write” means “our handwriting,” then writing the bill of divorce is considered as though he said “say [to others],” and Rabbi Yosei agrees in a case where he said “say [to others].” Meaning, if a person says to them, “tell someone to write a bill of divorce,” that is valid. Meaning, if I appoint them to write a bill of divorce and then they go and tell someone else to write it, that is a problem of verbal matters according to Rabbi Yosei. Yes, there is a tannaitic dispute whether verbal matters can or cannot be entrusted to an agent. The law follows Rabbi Yosei that verbal matters cannot be entrusted to an agent. But if I said to the agent, “tell a second agent to write,” in that case Rabbi Yosei also agrees that it is valid. Only if the agent does it on his own initiative, passing the task to someone else, then it is not valid. But if I appointed him for that role, then it is valid—even Rabbi Yosei agrees. What does that basically mean? There is an assumption here, apparently, that it is considered as if he said “say [to others].” Yes, even if he did not say it—but if it is clear to us that he did say “say [to others],” then it is valid, yes. The picture that emerges from here, at least simply—and we will see later that it is not so simple—is basically that the problem of verbal matters not being entrusted to an agent arises only when the first agent tries to transfer it to another agent. So this is connected to our topic of an agent appointing an agent. When I myself would say to the scribe “write,” that too is verbal matter. So why is that effective? It is effective, right? The whole problem is when I appointed agents and told them to write, and they went and transferred it to a scribe—or even if I told them to tell a scribe, depending on whether “say [to others]” works or not. That is a dispute among amoraim in the Talmud there. Okay? The problem always arises when the first agent tries to transfer these verbal matters to a second agent. But the very entrusting of verbal matters to the first agent—that is fine. And also with the second agent, if I said “say [to others],” meaning the whole problem is when I gave them the task and they transferred it on their own initiative to a second agent. But if I initially sent them to pass it on to someone else, that is fine according to this view. I said there is a dispute among amoraim there about that too—not everyone agrees that Rabbi Yosei concedes in a case of “say [to others]”—but that is the view appearing here. So that is what emerges from the Talmud here; explanations later. In the Shulchan Arukh itself there are three views regarding “say [to others].” These are disputes among the medieval authorities. There is a view that “say [to others]” works. Meaning, as practical law, verbal matters cannot be entrusted to an agent—that is ruled like Rabbi Yosei. But in a case of “say [to others],” some views hold that even there it does not work, some hold that it does work, and Maimonides apparently is uncertain. Okay? Maimonides is uncertain whether it works in a case of “say [to others]” or not, and apparently that is basically the conclusion of the Talmud here, because the Talmud here is trying to identify what Shmuel was uncertain about. When it brings here that Rabbi Yosei concedes in a case of “say [to others],” well then, if he concedes, what is the problem? What is Shmuel uncertain about? That Shmuel was uncertain whether Rabbi Yosei concedes in “say [to others].” Meaning, they explain here the side of Shmuel’s uncertainty that was not clear. It still remains a doubt. That is apparently Maimonides. Therefore, according to Maimonides, it remains doubtful whether “say [to others]” works or not. In any case, that is for our purposes.

Now I want to ask: what exactly is the problem with verbal matters, and why does “say [to others]” solve the problem? Apparently if “say [to others]” solves the problem, that would seem to say that really this has nothing to do with the question of verbal matters at all. It just means: you did not receive the mandate to appoint another agent. If you did receive it, then fine, everything is fine. And that takes us right back to the previous discussion of an agent appointing an agent, right? The question is whether the first agent got a mandate to appoint, or whether consent is enough, or whether he did not get the mandate. But the language of verbal matters seems to indicate that there is something else here; otherwise this should also arise in forms of agency that are not verbal matters. The point, apparently, is that the reason to think I did not give him the mandate is precisely because this is verbal matters. Meaning, the attempt here is to infer the sender’s intention in an unspecified case. I don’t think—you’re not right, I think, though we’ll see later—but I don’t think you’re right, because the language of verbal matters suggests that there is something else here. I understand, I understand, but the problem is verbal matters. Good, we’ll see later—not sure if later today, I already see we won’t get there today—Rabbi Akiva Eiger on this. If verbal matters are not entrusted to an agent, is that only regarding writing or also regarding delivery? What do you mean? For example, with an agent of delivery of a bill of divorce—is that verbal matter? No, that is not verbal matter. Verbal matter is only in writing. No, words—what exactly counts as verbal matters? That is not a simple question; there are disputes about it, we’ll see later. Okay? I’m careful with definitions, because the definitions here are disputed, so I’m sticking somehow to the literal meaning.

So there is the position of Nachmanides, whose claim is that in “say [to others]”—that is, “say [to others]” is valid, but specifically in a bill of divorce it is invalid, and perhaps that was Shmuel’s doubt—that perhaps even in “say [to others]” it is invalid in a bill of divorce because it has to be written for her sake. Since the husband has to write the bill of divorce for her sake, the command must be heard from the husband’s mouth, and therefore even “say [to others]” does not help. That is also brought by Ketzot. Yes, there are those who validate “say [to others],” and this means that although verbal matters are not entrusted to an agent, that is where one person was appointed as an agent in a matter and he cannot appoint another agent in his place when it is verbal matters. But in “say [to others]” it is valid. And see the Tur, who wrote this in the name of Nachmanides, distinguishing between a bill of divorce and a gift document: in a bill of divorce it is impossible, but in a gift it is possible. Meaning that regarding a bill of divorce, even “say [to others]” is invalid, because it requires for-her-sake. But regarding a gift, “say [to others]” works. However, Nachmanides’ view is that he validates “say [to others]” in a gift, and that is specifically only where he says “say to so-and-so,” because that is like appointing an agent not in his presence. As taught: one who appoints an agent, tell Reuven to write a bill of divorce for my wife—that is, I told him to whom to pass the task. That is called “say [to others].” In a gift that is valid, but in a bill of divorce that too is invalid. But in a gift it would be valid. Why? Because it is simply a case of appointing an agent not in his presence. After all, I can appoint someone as my agent not in his presence; he doesn’t have to be right there. What difference does it make if I conveyed the order for him to be my agent through someone else? It is like calling him on the phone and saying, be my agent. But regarding a bill of divorce, even in such a case it is invalid because it requires for-her-sake, and we specifically need the husband to command the witnesses, and it does not help if it is not in his presence. And regarding a bill of divorce too, in the case of an agent of receipt, in such a case of “say [to others]” to a specific person, it does help. “In such a case” meaning with a specific person, yes. Then regarding a bill of divorce, in the case of an agent of receipt, it is effective. Why? Because it is like appointing an agent not in his presence, and an agent of receipt does not require for-her-sake, only writing and signing do. So Nachmanides basically shifts the whole discussion to the laws of for-her-sake. The whole problem here is essentially the problem of for-her-sake. Therefore, if we say that “say [to others]” works, that still leaves the problem of for-her-sake. Where does the problem of for-her-sake exist? In an agent of delivery, or an agent for writing and not delivery, yes? Why? Because there there is indeed a law of for-her-sake. The writing has to be for her sake, and therefore even in “say [to others]” that does not help. And that is just a law in divorce, not a law in the laws of agency. By the way, this does connect to the limitations on the laws of agency that I spoke about: that there are things the person himself must do physically, and then the law of agency will not help. For example, saying the words in order to make the writing be for her sake, which is apparently a conception like putting on tefillin—that is, you yourself must do a physical action; it is not enough that your agent does it. Even agency for an act would not help here. What about an agent of receipt for a bill of divorce? After all, with the woman there is no issue of for-her-sake; she does not give… But in “say [to others]” at your own discretion—what does that mean? Choose someone and tell him; I did not determine to whom you should tell it. Then that is no longer an appointment not in his presence, right? Here there is no appointment at all; it is the agents’ appointment, not mine. Then it is invalid because verbal matters cannot be entrusted to an agent. There is no distinction between a bill of divorce and a gift. Because the distinction cannot be due to for-her-sake, since in “say [to others]” they stand in the husband’s place to appoint an agent, and they are appointing an agent in his presence—power of attorney. They are basically the husband now. So if they are the husband, then they are appointing an agent in his presence. Rather the problem is verbal matters. Meaning, there is no problem of for-her-sake; there is a problem of verbal matters. And on that there is no distinction between a bill of divorce and a gift. Therefore it would be invalid both in a bill of divorce and in a gift. And in “say [to others]” to a specifically identified second agent, it would be valid in a gift and invalid in a bill of divorce. And from the plain wording of the Tur, who wrote in the name of Nachmanides to distinguish between a bill of divorce and a gift, it seems that in a gift, an unspecified “say [to others]” is valid—even if he does not specify to whom to say it, “say [to others]” to a particular person. However, in Even HaEzer 141 it appears that the Tur’s view is that even in a bill of divorce, “say [to others]” is valid. In short, again, there are three views here. Where is an unspecified “say [to others]” valid? Both in a bill of divorce and in a gift; only in a gift and not in a bill of divorce; or in neither, unless it is “say [to others]” to a specific person. In that case, at least in a gift it would be valid according to Nachmanides; in a bill of divorce not; according to others, even in a bill of divorce yes. There are several different views here. Again, I’m not entering the details of “say [to others]”; I only want you to get the sense of it. What? From here I want to move to an attempt to define the problem of verbal matters, and “say [to others],” and an agent appointing an agent. This “say [to others]” is a general introduction. Now I want to get into a few medieval authorities, because the medieval authorities dispute whether an agent of receipt can appoint an agent in our case. For the moment forget the introduction about verbal matters—I’ll use it, but right now I’m returning to our topic. Okay?

Tosafot Rid on our passage writes as follows: “That the agent appoints an agent—meaning an agent of the husband appoints an agent. An agent of delivery appoints an agent, as we learned in chapter 2 of tractate Gittin: he forms a court and sends it. But an agent of the woman cannot appoint an agent, because these are verbal matters, for the woman transmitted to him only mere words: ‘Receive my bill of divorce for me.’” Ah? What’s this, first time? The woman’s own agency? The woman transmitted to the agent of receipt verbal matters, and therefore this agent of receipt cannot appoint an agent. Note: the agent of receipt himself is no problem—there is such a thing as an agent of receipt, that is the novelty of our Talmudic passage, of the baraita here—but this agent cannot appoint his own agent to receive the thing on the woman’s behalf, because this is verbal matter. Whereas an agent of delivery is not verbal matter, and therefore there the agent can also appoint an agent. Fine? This is even without “say [to others],” without any… without any qualifications at all. But with an agent of receipt, since it is verbal matter… Or—that is a question, and we’ll get to it. But that is what he says. He says: of course the agent of receipt himself exists, as the novelty of the Talmud here, but that agent cannot appoint a second agent because it is verbal matter and verbal matters are not entrusted to an agent. “For the woman transmitted to him only mere words, ‘Receive my bill of divorce for me,’ and we hold like Rabbi Yosei in chapter ‘The one who says receive,’ that verbal matters cannot be entrusted to an agent. And it is not comparable to the husband’s agent, to whom he transmits an object, and who can appoint an agent, as I explained in the booklet of responsa.” It is commonly understood—this is a first definition—it is commonly understood in Tosafot Rid, and I think this is also the common definition among the later authorities, though we will see not all the medieval authorities agree with it—that the question is whether physically you gave him an object. Did you hand the agent something when you appointed him, or did you only say words to him? If you hand him something to do something with, I don’t know what, a tangible object, that is not verbal matter. If you tell him or appoint him to perform some action, some task, that is verbal matter. And the difference is, again, one can be an agent for both this and that. The difference is only in the question what happens when that agent wants to appoint a second agent in his place. There there will be a difference. In verbal matters you cannot, and in regular agency you can. And with “say [to others]” you can too—either specifically or not specifically, as we saw above in Tosafot. By the way, does that also apply in matters involving an agent of delivery? Again? If we define it with an agent of delivery, then perhaps it can be verbal matter on the husband’s side too. Why? Say at the beginning… No, the agent of receipt… I’m speaking now about an agent of delivery in divorce. You hand him the bill of divorce. In betrothal too, in principle, it is like that—you hand him the money or the document of betrothal. But we’ll see, there are medieval authorities who distinguish here. Okay? So basically the claim is that with an agent of receipt he cannot appoint a second agent because of the rule of verbal matters. That is basically the point. More than that—note that here this is not an explanation of why a source is needed for it. Earlier I brought these ideas as an explanation of why a source is needed for the rule that an agent appoints an agent. He says this is the practical law. As practical law, an agent of receipt cannot appoint an agent, even after the novelty that an agent appoints an agent. The novelty that an agent appoints an agent applies only to an agent of delivery, which is not verbal matter. But with an agent involving verbal matters, even after the novelty that an agent appoints an agent, it is impossible. Verbal matters are not entrusted to an agent. Even if she said it explicitly? No, if she said it explicitly, then yes—or at least, again, that is the dispute we saw above, that in “say [to others]” perhaps Rabbi Yosei concedes. In “say [to others],” the reasoning says that Rabbi Yosei concedes, right? Now the question is how we rule. I said that among the medieval authorities, I don’t know how Tosafot Rid rules, but according to the various rulings, it could absolutely be that in “say [to others]” it would help. But if she said to him, be my agent of receipt, and did not say to him to tell someone else to be an agent of receipt, then no. Why does it matter whether he passes an object? No, according to Tosafot Rid’s explanation it is not so clear why it matters. What difference is there? Why should it matter whether you give him an object? I’ll touch on that. It’s more technical. Okay.

The Ritva and the Rashba here—the Rashba, for example, writes here as follows: “It teaches that an agent appoints an agent. And if you say, granted that his agent can appoint an agent, but how can her agent appoint an agent? These are verbal matters, and verbal matters cannot be entrusted to an agent, as we say in ‘One who brings’? And Rabbi Barzeloni explained that here we are dealing with a case where she appointed him with a formal act of acquisition, which has the strength to enable him to appoint an agent.” Maybe it is an appointment by a formal act of acquisition, and that gives him greater authority, so he can also appoint a second agent. “And it seems more reasonable”—a more plausible explanation—“that she gave him permission to appoint an agent. For we only say verbal matters cannot be entrusted to an agent where the agent appoints him on his own initiative, but with the owner’s intention he may appoint an agent, as is proven there, since in a case of ‘say [to others],’ an agent appoints an agent.” A second possibility. A third possibility: “Alternatively, where the agent fell ill or was prevented, the agent appoints an agent because the presumption is that whenever he fell ill, it is the owners’ intention that the agent appoint an agent.” Like “say [to others],” but not exactly. We’ll see in a moment. “Here too, we are speaking in such a case, and they only came here to teach that the owners may appoint an agent and that their agent may appoint an agent according to the law. They did not come to say here that every agent appoints an agent.” There is a problem of verbal matters, as Tosafot Rid says. In places where the law of verbal matters allows it—that is what they are really saying here, that in those places too an agent may appoint an agent. By the way, that implies that even if in principle it were possible, you still need a source for an agent to appoint an agent. Because one might have said that this very thing is what the verse comes to say—that an agent appoints an agent. That it comes to say that in “say [to others],” verbal matters can be entrusted to an agent. But from him it sounds not like that. In “say [to others],” that is obvious, because that solves the problem of verbal matters. Even then, however, a verse is still needed that in general an agent may appoint an agent according to the law.

Okay, but for our purposes, I just want to finish because we need to stop—there is a difference between the last two explanations, and it is important for our purposes. The first possibility is where he says to him, appoint an agent: “say [to others].” The second possibility is where the agent became ill. If the agent became ill, then there is some presumption that this accords with the sender’s will. Now the question is why that presumption is enough—that it accords with the sender’s will. One could have said that maybe in every appointment we assume there is some implied appointment also to appoint another agent if the first agent becomes ill. But perhaps not. Perhaps one can say that the whole problem of verbal matters not being entrusted to an agent is that the original sender does not want this transfer to another agent. Therefore in “say [to others]” it really works. Now if we are dealing with a case where he became ill, then from our perspective he is not objecting. This is not an implied appointment; rather, it is enough that he does not object, since the original appointment includes within it also the possibility of appointing another agent. If the sender objects, then this takes us back to the earlier discussion: if the sender objects, then of course not, because “I sent you to improve matters, not to distort them,” that is, I did not appoint you on that basis. So I cancel it. Now clearly, if he became ill, then we assume he does not mean to cancel the agency if I appoint a second agent, because he wants the task to be carried out. True, he would prefer the first one to carry it out, but if the first one can’t, then he surely wants it to be carried out in one way or another. So his consent is enough; an actual appointment is not needed. His consent is enough, and this is very important. Because the question is whether the first agent appoints the second agent as the original sender’s agent, or whether he appoints the second agent as his own agent, so long as the sender does not object. That is the question of how to understand this ending of the Rashba. If he became ill, then somehow it appears that this is fine. Why is it fine? Simply because I do not object. Not because there was an actual prior appointment on that basis; not because I had thought in advance about him becoming ill. Rather, if he becomes ill, then we estimate his intent and say that he has no problem with it. That is the meaning. And then this means that the sender’s consent is enough; there is no need for the sender to delegate to the agent the authority to appoint the second agent. He didn’t know in advance. No—sometimes people say it is as if he made an implied condition, and then it is basically as if he really made the condition, like in “on condition that I go up to the Land of Israel.” I cannot think of any reason why it should matter to the sender whether it is this agent or that one. Or rather, in places where it is clear to me that he doesn’t care, then it is the same as becoming ill. And for example a woman who receives… So I say, one can discuss it, but at the conceptual level, anywhere my presumption is that he does not care, according to this Rashba that is apparently enough. More than that, I want to ask: in what case would he really care? No, the Talmud itself says—the Talmud itself that we read earlier on 66 says that if he says “write,” maybe he wants them to write and not someone else. I don’t know why—he trusts their “for her sake,” I don’t know exactly what. It is strange. Strange, but the Talmud says there is such a presumption. Now I say again: I don’t need to learn the facts of reality from the Talmud. Meaning, if today I came across a case that was exactly the same, where he said to them “write the bill of divorce” and so on, I would not necessarily take from the Talmud the presumption about what people intend. If it were clear to me that the person does not care, then I would rule that he does not care and everything is fine. That is, this is not… but never mind, for our purposes, if he does care—not important right now how I assess that, how I know—but I reached the conclusion that he may care, then it won’t help. I need an example so that… that there is someone fit to write. What? Listen, maybe he is embarrassed that he cannot write, so he wants others to say to write it, so that more people won’t know he can’t write. I don’t know. Right, something like that in the Talmud. For that, all the other embarrassments are written, and for that it is enough that a scribe who knows the format write the standard text, and the divorcing husband fill in the names. There are things where it is a matter of appearance, as if—he wants the ceremony to be ideal. It doesn’t matter right now. What matters is that this is a question of factual presumption. Here I say: for me it really makes no difference, because I think in the end the judge must decide on his own regarding the case before him. Even if that exact case appears in the Talmud and there it says he does care, that does not really have to be applied automatically in our case. If it is clear to me that he does not care, then in the Talmud’s time they assessed that he did, and today I assess that he doesn’t. It is a factual question, not a legal one. Therefore I do not even care to enter the discussion why the Talmud thought he cared or did not care. It thought so—so what? Okay? Good, we’ll stop here. Next time I’ll continue a bit more with this issue of verbal matters and an agent appointing an agent, and then we’ll go back to our sugya.