Learning from Experience – Lesson 9

This transcript was generated automatically using artificial intelligence. There may be inaccuracies in the transcribed content and in speaker identification.

🔗 Link to the original lecture

🔗 Link to the transcript on Sofer.AI

[Rabbi Michael Abraham] No. In a place where there is only possession, without the presumption that whatever is in a person’s hand is his, it could be that fifty-one percent would indeed suffice. In a place where we say, for example, that we do not say a migo in order to extract property, then it’s not simple. If there is no presumption against that migo, then it’s not certain that we would say a migo cannot be used to extract property if all we have is mere possession. The problem is that in most cases, alongside possession there is also a presumption. But for example, there is the case of “those goats that ate peeled barley,” right? The goats that are found in my yard, and there is a dispute between me and someone else over whose goats they are. Now, goats normally wander around. The fact that they are in my yard does not prove that they are mine, because goats roam. And still, they give me some preference, and the burden of proof is on him. Why? Because right now the goats are by me, so why should a religious court take the goats away from me if there is no reason? But in that case, it could be that fifty-one percent really would suffice, because here there is no real evidence in my favor, no real weight. Okay, so that really is already a topic we won’t get into now. The Talmud in Yevamot, when it rules the Jewish law—I remind you that there is a dispute between Rabbi Yehuda HaNasi and Rabban Shimon ben Gamliel whether a three-time presumption is established after two times or after three times. Right, a three-time presumption—what color do you want the silver stripe to be? Right, a three-time presumption is established after two times according to Rabbi Yehuda HaNasi. So the Talmud rules the Jewish law and says: Rav Yosef son of Rava said to Rava: I asked Rav Yosef: Is the Jewish law like Rabbi Yehuda HaNasi? And he said to me yes. Is the Jewish law like Rabban Shimon ben Gamliel? And he said to me yes. Is he making fun of me? Right? So Rav Yosef says Rav Yosef laughed at me. Why? Because Rav Yosef son of Rava asked Rav Yosef, and he asks him, tell me, is the Jewish law like Rabbi Yehuda HaNasi? He says yes. He says, is the Jewish law like Rabban Shimon ben Gamliel? He also says yes. Meaning, Rav Yosef laughed at me—that’s what he says to Rava. Rava’s son says Rav Yosef laughed at me, says to his father. I’ll tell on you to my father, so to speak. So he said to him: No, he is blind and he answered you straightforwardly: marriage and circumcisions are like Rabbi Yehuda HaNasi; menstrual cycles and an ox forewarned for goring are like Rabban Shimon ben Gamliel. What does that mean? What distinguishes these contexts from those? So the medieval authorities (Rishonim) write—Nimukei Yosef and Ritva and Rashba and Nachmanides, many medieval authorities (Rishonim) write—that essentially we remain in doubt whether Rabbi Yehuda HaNasi is right or whether Rabban Shimon ben Gamliel is right. Do you need two, or do you need three? So what then? In a place where we are dealing with stringency, we go like Rabbi Yehuda HaNasi. In short, we always go stringently. If Rabbi Yehuda HaNasi is the stringent view, then two times creates a presumption. If Rabban Shimon ben Gamliel is the stringent view, three times creates a presumption. Now what happens in the case of menstrual cycles, which are rabbinic? So why do we rule like Rabban Shimon ben Gamliel? Because we want to know whether after two times a woman is already obligated to be careful about the cycle the next time. So here we go stringently and therefore I say you need three times. No—menstrual cycles are rabbinic, that’s a different story altogether. Menstrual cycles are rabbinic, and there we go leniently in a doubt. But an ox forewarned for goring, for example—in the case of an ox forewarned for goring, you are coming to extract money from a person, you want to collect full damages rather than half damages as with a non-forewarned ox. In order to extract money, you need three times. Okay? Meaning, when in doubt you cannot extract money, so after two times there is a doubt whether the ox is forewarned, after three times it is certain. So in doubt you go with three times. Usually the approach is, and so on, and regarding marriage and circumcisions we go stringently like Rabbi Yehuda HaNasi, meaning that because we are dealing here with danger, we say two times is enough to be concerned and not let her marry the third groom because he may die, or to circumcise the third son, and so on. Now the accepted view is that we are dealing here with the laws of doubt, and therefore we follow the laws of doubt. So in a possible danger we are stringent, in a rabbinic-level doubt lenient, in a Torah-level doubt stringent, and so on. But these are the laws of doubt. The way I presented it earlier, it can be explained differently. It’s not the laws of doubt. In a place where you are trying to dislodge an existing presumption, then you need strong evidence to dislodge the presumption. Strong evidence is only after three times, but certainly not because of doubt. This ruling is decided as a definite ruling, not as part of the laws of doubt. When you come to change a presumption, you need good evidence. Good evidence is after three times. If you are not coming to change presumptions, then weaker evidence suffices, and therefore two times is also enough, exactly as I said earlier regarding the common denominator. Because in fact Keren Orah asks—he asks: what happens with an ox forewarned for goring that later reverts and becomes a non-forewarned ox? It doesn’t gore. Some other ox, or dog, or something, comes before it and it doesn’t gore it, so it reverts to being a non-forewarned ox. Now if with an ox forewarned for goring we go after three times—because with an ox forewarned for goring the Jewish law is like Rabban Shimon ben Gamliel, you need three times. Three times is to turn it into a forewarned ox—why? Because I am coming to extract money from the owner of the ox; instead of half damages he will pay full damages. To extract money you need evidence, so in doubt you cannot extract money, therefore you need three times. He says: what happens in the opposite direction? Now I want to know whether the ox has reverted to being non-forewarned. Now this is really to keep possession, not to extract. So we should have been satisfied with two times that it did not gore in order to return it to the presumption of a non-forewarned ox. Right? Because now we are talking about whether it gored—let’s say it did not gore twice and now it gored, and now the question is whether it is a non-forewarned ox or a forewarned ox. If it is a forewarned ox, I need to pay full damages; if it is a non-forewarned ox, I do not need to pay full damages. So what does that actually mean? After two times it should revert to being a non-forewarned ox, because in doubt you cannot extract from me more than half damages. Meaning, it should be ruled oppositely when I turn the ox into a forewarned ox and when I return the ox to being non-forewarned. When it becomes forewarned, it needs three gorings for that. In order for it to return to being non-forewarned, it should be enough that twice it did not gore. But we don’t find that anywhere. The straightforward understanding is that it takes three times both to make it forewarned and to return it to being non-forewarned. That’s Keren Orah’s question. In fact, in Or Gadol on the Mishnah, he claims that no—according to those medieval authorities (Rishonim), it follows that returning to a non-forewarned ox is two times. But I think that is incorrect, because the point is exactly the point. The point is: this is not from the laws of doubt. I claim this is a definite ruling, not a ruling based on doubt. So therefore the question is not who is extracting money from whom, but rather what? Once the ox had the presumption of being forewarned, in order to remove it from the presumption of forewarned and turn it into the presumption of non-forewarned, it needs three times that it did not gore, because I am changing the presumption. Its status. It doesn’t matter at the moment that I am leaving the money with the defendant. There is no extraction of money here; on the contrary, it is retaining money. But as I said earlier, this is not from the laws of doubt, where I say that in doubt I cannot extract money. Rather, to change a presumption requires good evidence. Now, both changing the presumption from non-forewarned to forewarned, and from forewarned to non-forewarned, are changes of presumption. It means deciding that the status of the ox is no longer what it had been until now. For that you need good evidence, because I had good evidence that it was indeed a forewarned ox until now. In order to remove it from good evidence that it is forewarned, you need good evidence that it has become non-forewarned. So in my opinion that question is not correct. That question assumes we are dealing here with the laws of doubt, but I claim that this is a definite ruling, not from the laws of doubt. It is a definite ruling that says that in order to change a presumption, you need good evidence. No, there is no intermediate state. Meaning, if it did not revert to being non-forewarned, then it is forewarned. It is what it was until now. If it reverted to being non-forewarned, then it reverted to being non-forewarned. The presumption always continues as it was so long as it has not changed, so it remains what it was. So this connects to what I said earlier with the common denominator, that two times is enough, and so on—that if we understand it this way, then it explains the Talmud very well. A bit differently from what people usually understand. We need to look at the wording of the medieval authorities (Rishonim) to see whether all of them fit this. In some of them, in my opinion, it fits—that this is not the laws of doubt, but a definite ruling. It reminds me of—there is the same discussion in the topic of retroactive clarification. Whether there is retroactive clarification or there is no retroactive clarification, yes. Let’s say I have a barrel of wine of one hundred logs and I want to separate terumah, two logs, one out of fifty. I want to separate terumah, so I say: I will drink the wine, and the two logs left at the end will be terumah. And then it turns out retroactively that they had already been the terumah from the beginning. And therefore I drank the wine even before I separated the terumah. I separate unspecified terumah; after I drink, the last two logs will retroactively turn out to be those that became terumah from the beginning. So the question is whether we say that or not say that, whether there is retroactive clarification or there is no retroactive clarification. This appears in several Talmudic passages. In practical Jewish law, we rule that in rabbinic matters there is retroactive clarification, and in Torah-level matters there is no retroactive clarification. So the Ran explains that this is from the laws of doubt. Why? Because usually to say there is retroactive clarification is the lenient approach, and to say there is no retroactive clarification is the stringent approach. Sorry—to say there is no retroactive clarification is stringent, and there is retroactive clarification is lenient. For example, in the context of the wine, to say there is retroactive clarification means that I may drink—it is lenient. To say there is no retroactive clarification means that one may not drink, because I did not separate terumah. One has to separate terumah; you cannot rely on what will be clarified at the end. So therefore there we need—there, since it is a Torah-level law to drink untithed produce, therefore we rule there is no retroactive clarification, because that is the stringent approach. But in rabbinic laws, when it is in rabbinic laws, we rule there is retroactive clarification because that is lenient. And in a rabbinic-level doubt we go leniently. But there are those who say no, we rule that in rabbinic matters there is retroactive clarification and in Torah-level matters there is no retroactive clarification definitely—not out of doubt. It is not because of the laws of doubt. A practical difference: what would happen if there were a Torah-level law where ruling there is no retroactive clarification would specifically be the lenient side? You can imagine such a case. If this is about doubt, then we would go and rule that there is retroactive clarification even in Torah-level matters. But if this is a definite ruling that in Torah-level matters there is no retroactive clarification and in rabbinic matters there is, then it does not matter whether this ruling is lenient or stringent. It is a definite ruling; it is not that in the laws of doubt we go leniently or stringently. It is a definite ruling: in Torah-level matters there is no retroactive clarification, and in rabbinic matters there is retroactive clarification. It is somewhat like the discussion I am talking about here. There is—you could understand, yes, I’m just saying this in parentheses—there is another answer I thought of to Keren Orah’s question. Why, when I return the ox from the presumption of forewarned to the presumption of non-forewarned, do I still need three times that it did not gore? And I think this connects to the earlier explanation I gave. What I am saying is this, the formulation is this: basically these are the laws of doubt, and in doubt we do not extract money. Therefore, when the ox becomes forewarned, it needs three times that it gored in order to become forewarned. If it gored only twice, then it is in doubt whether it is non-forewarned or forewarned, because the question is whether Rabbi Yehuda HaNasi is right or Rabban Shimon ben Gamliel is right, whether there is a presumption after two times or after three. And if you come to extract money from him, in doubt you cannot extract money. And then Keren Orah asks, yes, but what happens if the ox had been forewarned, then did not gore twice, and now I ask whether it has reverted to being non-forewarned or not. In that case, if it has reverted to being non-forewarned, then the damager will pay half damages. And if it has not reverted—if it is still forewarned—he will pay full damages. So here, ruling that the presumption has changed leaves the money in his possession. Therefore here, if this is about the laws of doubt, then I would have expected that in doubt two times would suffice. Two times for it to revert to being non-forewarned. Yes, that is Keren Orah’s question. To that one can formulate, based on the Chazon Ish—you can see such a view. The Chazon Ish says—and this really is not directly connected to us, but it is an interesting point—the Chazon Ish argues as follows. There is an inquiry in monetary law of damages. When my ox gores another ox, then I have to pay. There is an inquiry as to why I have to pay. Is my obligation to pay because I did not guard it properly, I was negligent in guarding it, or not? Is my obligation to pay because my ox caused damage, and I am responsible for what my ox does? Only if I guarded it properly am I exempt. But my liability is not because of negligence; my liability is responsibility for what my ox did. A person has responsibility for what his property does. Usually they bring a practical difference for this question—a dispute between Pnei Yehoshua and the Chazon Ish. Pnei Yehoshua argues—what is the dispute? The dispute is over the question where my ox gored your ox, and there is a dispute over whether I guarded it properly. I claim I guarded it properly, and therefore I should be exempt. The injured party claims no, you did not guard it properly. On whom is the burden of proof? Who has to bring the evidence? So Pnei Yehoshua argues that the injured party has to bring the evidence. Why? Because the injured party wants money from the damager, he wants the damager to pay him. The one who seeks to extract from another bears the burden of proof. You want to be paid, you are the plaintiff. The one who seeks to extract from another bears the burden of proof. And the Chazon Ish argues that the burden of proof is on the damager. So the later authorities (Acharonim) explain regarding the Chazon Ish—he is a later authority (Acharon), right? But the later authorities (Acharonim) after him tend to explain the Chazon Ish as holding like the view that liability is for negligence in guarding. Liability for damages is for negligence in guarding. And then what? So long as you have not proven that I was negligent in guarding, you have no grounds to obligate me. Prove that I was negligent in guarding in order to be able to sue me. Okay? If you haven’t proven it, the burden of proof is basically on—well, that is actually Pnei Yehoshua, sorry. The burden of proof is on the injured party who claims the money, because if he does not prove that I was negligent, there is no cause of action. So what if my ox caused damage? If I was not negligent, you cannot sue me. In contrast, the Chazon Ish argues—as the later authorities (Acharonim) explain—that the monetary obligation to pay is not because of negligence in guarding but rather responsibility for the acts of my ox. So what? Then once my ox gored, first of all I am liable, I am responsible. True, if I prove that I guarded it properly I will be exempt, but that is a claim that comes to exempt me; it’s not that there is no reason to obligate me. There is reason to obligate me because my ox caused damage. If I prove that I guarded it properly, then they will exempt me. In other words, says the Chazon Ish, then obviously the burden of proof is on me, because I am the one trying to exempt myself. There is a basis to obligate me. If I want to be exempt, then I should prove that I really deserve exemption. So I need to prove that I guarded it properly. So apparently the dispute between the Chazon Ish and Pnei Yehoshua is over the question why one pays for one’s property that caused damage: is it because of negligence in guarding, or is it responsibility for the acts of my ox? Except that Rabbi Blumentzweig, the rosh yeshiva in Yeruham, once told me that everyone keeps repeating this all the time and nobody looks at the Chazon Ish himself. When you look at the Chazon Ish himself, he explicitly writes the opposite. He writes that the burden of proof lies on the damager because the claim that the ox caused damage even though you guarded it properly is not a reasonable claim. Not because the burden of proof is on you because you need to exempt yourself and the negligence issue is an exempting claim—no. The opposite. The Chazon Ish apparently learns like Pnei Yehoshua: that negligence in guarding is what obligates him, and therefore in principle the burden of proof should be on the injured party who sues him. Except that, says the Chazon Ish, but what are you basically claiming here? That you were not negligent, you the damager. Because you are claiming: I was not negligent, therefore I am exempt. You are basically saying that your ox caused damage even though you guarded it properly? That is not reasonable. To guard it properly means precisely that it is not reasonable for the ox to cause damage. Meaning, the claim—it can happen, and if you prove that you guarded it properly and nevertheless the ox caused damage, you will be exempt. But it is a weak claim. It is an unreasonable claim, and therefore the claim you are making—that I guarded it properly and the ox caused damage—is an unreasonable claim, and therefore the burden of proof shifts to you. A major novelty in the Chazon Ish: he is essentially saying that the burden of proof is not always determined by the question of who is the plaintiff and who holds the money. Usually the burden of proof is on the plaintiff—the one who seeks to extract from another bears the burden of proof. The one in possession of the money, the burden of proof is on the other party. He says—the Chazon Ish says that is all true if the one holding the money is making a reasonable claim. But if you, who are holding the money, are making an unreasonable claim, then the burden of proof shifts back to you. You need to prove it. Because you are making an unreasonable claim. And this is a novelty: possession is not determined only by the question of who is holding the money; one also factors in the plausibility of the claims of the two sides. Whoever makes a very implausible claim, the burden of proof will be on him. If so, then basically the claim is—that what I want to argue is that even if with a three-time presumption, where if it is after two times or after three times we really are in doubt and in doubt we do not extract money, therefore when the ox reverts to being non-forewarned, it should be enough that it did not gore twice in order to return to the presumption of non-forewarned. Okay, that is Keren Orah’s claim. My claim is that this is incorrect. Why? Because if the ox had been forewarned, then it is not reasonable that all of a sudden, by not goring, it is now non-forewarned. It is probably just a case—that it happened not to gore. To prove that this is really essential, not incidental, you need three times. Even though here, in terms of who is extracting the money, the other side is extracting the money. I am holding the money. But still, it could be that the burden of proof is on me because my claim is not plausible. Now you understand that this is an explanation even if we understand that this really is a ruling from doubt. But the logic of this explanation is exactly the logic I suggested earlier. Because why is this claim not plausible? Because it comes to depart from the current status of the ox. Right now the ox is presumed to be forewarned. Now suddenly you are telling me that a forewarned ox doesn’t gore? Not plausible. Show me three times that it didn’t gore and I will be convinced that its nature is truly different. But if you do not show me that, I will say fine, it was incidental, it was distracted, I don’t know exactly what. Because you are making a weak claim. Once you make a weak claim, the burden of proof is on you, and therefore two times will not suffice; you need to bring three times that the ox did not gore. So for our purposes, what we learn from here is that the difference between two times and three is basically the difference in what strength of evidence you need. If you are coming to dislodge an existing presumption—not necessarily a monetary presumption, but to change an existing status—you need strong evidence, and that will require three times. If you are just trying to decide between two equally balanced possibilities, two times is also enough, because two times still gives you a slight preference for one explanation over two explanations, and that is exactly the situation in the common denominator. So here I close the question that was asked last time.

[Speaker E] Rabbi, once again, Rabbi—once again, according to this nice innovation of the Chazon Ish, that according to the Chazon Ish even with possession, if there is a strong claim against it then we don’t follow “the one who seeks to extract from another bears the burden of proof”—so if, let’s say, once again I’m some righteous man from Bnei Brak and I have a dispute with someone from Sodom, and the Sodomite is holding the money. Everyone there is robbers, Sodom plus-plus. And he is in possession, and this Sodomite who is in possession is making some claim. But on its face, as a claim, it is very weak, because everyone there—

[Rabbi Michael Abraham] —are robbers and so on. In such a case, in principle, it could be. Meaning, if the claim is considered problematic enough, then it could indeed be that we would reverse the burden of proof according to the Chazon Ish. Okay. Although I would say that there is a difference between the problematic nature of a claim in itself and statistical problematic nature. What do I mean? When you tell me that you guarded the ox properly and nevertheless it caused damage, the claim is internally implausible. If you guarded it properly, how did it cause damage? Now I’m not saying it can’t happen, but it’s not plausible. But with the claim of the Sodomites—the fellow says, what do you want from me, I’m a righteous man in Sodom. So there are many wicked people around me—so what? Do you have evidence that I too am wicked? Meaning, if there were evidence that he himself is wicked, then you would be right. But if you are only making claims against him just because most of his neighbors are wicked, it’s not clear that that would suffice. I gave an example of this from the world of law; in legal theory they give several examples. Let’s say there are two bus companies in a city. The blue bus company and the red bus company. Now the blue buses are ten percent and the red buses are ninety percent of the buses; the red bus company is larger. Now a bus ran over a person, injured someone. Now we don’t know—there is no way to know—whether it was a blue bus or a red bus. The question is which company should compensate him. Seemingly there is a ninety percent chance that it was a red bus, so the red company should have to compensate him even though I have no proof that it caused this, because I know it was a bus but I don’t know which color. So ninety percent of the buses—and let’s assume we really do follow the majority. Right now I’m ignoring—according to Rav, that we do follow the majority in monetary matters. Fine? So the red bus company should have to pay. The answer is that in the legal world they don’t accept that. The red company does not have to pay; you have to bring witnesses that it was a red bus. Now think of another case. Two witnesses come and say, listen, it was a red bus. Two witnesses say it was a red bus. Now, however, we have statistics showing that witnesses sometimes make mistakes. Witnesses don’t always see accurately. Here the red company will pay. Now even assuming it is the same probability—let’s say the witnesses are ninety percent reliable, and ninety percent of the buses in the city are red. Still, in the first case the red bus company will not pay, and in the second case the red bus company will pay, and in both cases it’s ninety percent. The question is why. And this seems to be accepted probably in every legal system in the world, if I understand correctly; I read an article by David Enoch about this. Many others have dealt with it; it is a famous question. In my opinion the explanations that were given are not good; I propose another explanation on my website. But what really is the difference? The difference is that the red company can say: true, ninety percent of the buses are red, but my drivers are careful. In order to obligate me, you need to bring evidence about the matter itself, not circumstantial evidence that seems to point in my direction. So why with witnesses who testify that it was the red company—there I am not concerned? Why not? That too is only ninety percent. Because there is positive evidence about the red company in a concrete way. True, the evidence may perhaps be wrong, but the evidence at its base speaks about positive guilt of the red company. It is not circumstantial evidence; it is direct evidence. It is direct evidence that may not be correct. The first evidence is circumstantial evidence, even though the strength is the same strength. Meaning, the percentage that you are right is the same percentage. But with circumstantial evidence you cannot obligate a person, you cannot tell him you are wicked because in your neighborhood most people are wicked. Fine, but I’m righteous—what do you want from me? What does that have to do with me? But if you tell me: you are wicked—true, sometimes you don’t do a wicked act, but you yourself are wicked in ninety percent of cases—that would indeed be a good claim. Because then you have a concrete claim about him; it is not circumstantial evidence. Even though that claim too is only ninety percent, because sometimes he does not behave wickedly. But once you bring evidence about the person himself, that he behaves wickedly, that is good evidence. Okay? Therefore I say that in this Sodom case too, the fact that someone came from Sodom is not necessarily enough to make his claim implausible and place the burden of proof on him. But if he himself is known as a Sodomite in the sense of how he behaves, not where he lives, right? That is something else. It could be that there indeed the Chazon Ish would say that the burden of proof shifts to him.

[Speaker E] What if, let’s say, let’s say the sainted Chazon Ish were suing Shalom Domrani over some 100 shekels, and the Chazon Ish has no evidence at all? But he says, “He robbed me, he passed by my house and snatched the 100 shekels from my bag.” There’s no evidence at all, and Domrani is in possession. So what would we say?

[Rabbi Michael Abraham] We would automatically say the Chazon Ish—

[Speaker E] —would win because…

[Rabbi Michael Abraham] So I’m saying: if I had clear evidence about Domrani that he’s a robber, I’m not sure such evidence exists, because otherwise he’d be sitting in prison. But let’s say I did have clear evidence that Domrani is a robber. In such a case, I assume the Chazon Ish would say that the burden of proof shifts to him. And I assume that in that situation, at the very least they would make him take an oath according to the Chazon Ish. Meaning, they wouldn’t just take money away outright, but they would let the Chazon Ish win the money by oath, without bringing direct proof on the substance of the matter. Okay, so that’s regarding the superiority of three over the superiority of two, a greater superiority and a lesser superiority. But for our purposes, what does this actually mean, if I go back to our line of thought? That when you have several examples, and of course the more examples you have the better, the generalization works that way in science too: the more experiments you bring that you performed, the theory is considered more corroborated, stronger. And that is exactly the mechanism here. If you have three pieces of evidence that a person drinks, that is stronger than two pieces of evidence that he drinks. Even though both with three pieces of evidence and with two pieces of evidence there is still an interpretive option to explain him as not a drinker. But the inferiority of that interpretive option, compared to the interpretive option that he is a drinker, is greater. The inferiority, yes, the gap in plausibility grows as the number of cases increases. Exactly like in scientific generalization. I said that the common denominator is generalization; the presumption established by three occurrences is all scientific generalization. It’s exactly the same logic. So that’s really the point, and therefore I argue that this is a halakhic example of using abduction and the induction derived from it in order to reach general conclusions. That’s basically what we’re doing here. Exactly as they do in the scientific context, this is really learning from experience. It’s a classic example of how in Jewish law we learn from experience. If something happened three times, my assumption is that there is some general law behind those three cases. Okay.

Now I’m moving to the next part, still within the presumption of three occurrences. When we speak about presumptions established by three occurrences, I’ll assume for now three times; let’s ignore the two for the moment, because in most cases, simply speaking, it’s three times. The later authorities discuss the question—I may have mentioned this last time—how to understand a presumption of three occurrences. You can understand it as an evidentiary presumption. That is, if the ox gored three times, that is evidence that it is a goring ox, that it is forewarned. So it’s an evidentiary presumption. And you can understand it as a presumption of habituation. If the ox gores three times, that ingrains in it the nature of being a goring ox. It becomes a goring ox; not that this is evidence that it is a goring ox. And I talked about this last time too, that even according to the view that this is evidence, it doesn’t work retroactively; it works only from here on. That doesn’t matter right now, but those are the facts. So the later authorities discuss the nature of presumptions established by three occurrences: are they presumptions of habituation or presumptions of evidence?

Now, in Kehillot Yaakov on Taharot, section 66—in the old edition I think it was 47, but in the updated edition it’s 66—there’s a long section there on presumptions established by three occurrences. He discusses at length exactly this point: whether these presumptions are presumptions of habituation or presumptions of evidence. And he says that he asked his brother-in-law, the Chazon Ish, what he says about it, and the Chazon Ish says there are both kinds. Meaning, it’s not true that every presumption established by three occurrences has the same character. There are presumptions that are presumptions of habituation, and there are presumptions that are presumptions of evidence.

Now, he doesn’t mention there the big difficulty that led to this hesitation, because after all, the Talmudic text in Yevamot that we saw last time compares these presumptions to one another. It says that whoever holds that a presumption is established by three occurrences will say that in all cases the presumption is established by three occurrences. Whoever says that a presumption is established by two occurrences will say that in all cases it is by two occurrences—Rabbi and Rabban Shimon ben Gamliel. In practical Jewish law we split them, but as far as Rabbi and Rabban Shimon ben Gamliel are concerned, the dispute runs across the board. Now, if there is no connection between the presumptions—these are presumptions of conduct and those are presumptions of habituation—then what room is there to compare them?

For example, there are presumptions of menstrual cycles in the case of a woman. Right? Or you know what, not menstrual cycles; maybe there one could still somehow argue, although it’s not likely. But what about the presumption of “his brothers died because of circumcision”? I circumcised two children, three children, and they all died. So there is now a presumption, and therefore I do not circumcise the fourth child. Or a woman whose three husbands died, and therefore she no longer marries a fourth husband, because there is concern that he too will die. Is that presumption an evidentiary presumption or a presumption of habituation? Here it’s obvious that we’re talking about evidence, right? It’s not that the woman gradually becomes deadly over time; rather, if she killed three husbands, apparently she has some kind of lethal nature. There it is clearly an evidentiary presumption, right? Now, if it is an evidentiary presumption, how can you compare that, for example, to the presumption of a forewarned ox, where there apparently it is a presumption of habituation? After it gores three times, apparently it becomes forewarned. What basis is there for comparison—if this is three times, then that too is three times? Why? Maybe habituation works one way and evidence works another way? Why does the Talmud compare them? If the Talmud compares them, that means it sees all these presumptions as something with the same character, and therefore if it’s three times in this case, then it’s also three times in that case. If it’s two times in this case, then it’s also two times in that case. And the Chazon Ish doesn’t answer that. Meaning, he says there are both kinds, and it doesn’t necessarily have to be the same character for all presumptions established by three occurrences. So how does the Talmud compare them?

I think the point is that in fact the Chazon Ish is right, and there are presumptions that are presumptions of habituation and presumptions that are presumptions of evidence. That’s clear. But, but, the number three—even if, let’s say, we talk only about presumptions of habituation, so let’s say all presumptions were habituation. Or evidence, it doesn’t matter. Why in all presumptions do three occurrences constitute evidence or constitute habituation? Why not four, two, twelve, twenty-four? Why specifically three? Maybe an ox gets habituated to being a goring ox after three times, but menstrual cycles enter some fixed routine only after five? Why three times? Why in all places does it have to be three times?

Yes, have you ever thought about it—it reminds me—have you ever thought about the number three in another context? Why do all university degrees take three years? Leave engineering and exceptional cases aside, but almost all degrees are three years. Maybe literature has more material than philosophy, or less material, and psychology—doesn’t matter—every field, physics, mathematics, all fields require three years of study. Why? Obviously it’s not because reaching a certain level in a profession requires the same amount of study time in all professions. Nonsense; obviously that’s not true. So what is it? You had to set some standard and measure that says: from when do you deserve a bachelor’s degree? That has meaning for salary, for—doesn’t matter. You made a certain effort, you reached a certain level of specialization of a bachelor’s degree. Okay, so you have to put the line somewhere. So they put the line after three years. That’s all. So however much material fits into three years in literature, philosophy, physics, psychology, everywhere—that’s what’s called a bachelor’s degree. We didn’t happen to arrive at three years because in every place we examined what needs to be studied to get a bachelor’s degree, and by chance in every place it came out to three years. It works the other way around. We are looking for a standard for what can be called a bachelor’s degree, because, say, the salary level is for a bachelor’s degree: in any field in which you did one, you deserve the salary of a bachelor’s degree. So it makes sense to demand roughly the same amount of study, the same level—I don’t know, however you define it—some kind of uniform measure. It’s not reasonable that you should get a bachelor’s degree in education—for which, after all, two days would be enough; that’s already too much in my opinion—compared to physics, where after three years I don’t know how much you know even after three years, or mathematics, or things of that sort. So if you’re talking about the question of how much is needed to reach a reasonable level of engagement in the profession, that will vary enormously from profession to profession. If you want to say this one has a bachelor’s degree, that one has a bachelor’s degree, and therefore he deserves a certain salary grade, then it makes sense to set a standard for how much one must study quantitatively in order to reach that standard. And the number three—in this case, three years for a bachelor’s degree—we did not arrive at it by chance. Rather, we established a standard of three years, because you have to establish the standard somewhere, so we set it at three years.

The same thing applies to presumptions. Let’s say, for the sake of discussion, that all presumptions were evidentiary presumptions, okay? Fine. But still, the evidence that a field is yours comes after three years of uncontested use, whereas the evidence that the ox gores may perhaps come only after seven times that it gored. And the evidence that a woman is deadly—maybe twice is already enough to make it clear that she is deadly. So why, why do we decide that everywhere it has to be three? And if Rabbi says two, then everywhere it’s two? It’s pretty obvious that here too it’s not because we’re trying to reach the same evidentiary strength, say sixty percent. Because in goring it could be that you get to sixty percent after four gorings, and in menstrual cycles it’s sixty percent after two times she sees blood. Each case on its own. Therefore it’s not that we want to arrive at the same evidentiary force, and therefore in all places it’s three times, and elsewhere it’s two times, or according to Rabbi it’s two times. No, it works the other way around. The evidence grows stronger as the number of occurrences rises. That’s obvious. Now somewhere you have to draw a line. He says: if it happened three times, from my perspective that is evidence. And Jewish law is supposed to operate simply, like a normal legal system. So there’s no point in every case entering into statistics and checking what percentage of probability there is after two times, three times, four times, and so on. He says: let’s set a standard—three times is sufficient evidence. That’s it; that’s my standard.

And therefore the number three in this case—like the three years of a bachelor’s degree—does not indicate that there is really similarity among all the contexts such that all of them need three times in order to create a presumption. But if that is true even if all presumptions were evidence or habituation, then it can also be true if some presumptions are evidence and some presumptions are habituation. Because we are setting a halakhic standard. If you need an accumulation of cases to create habituation or evidence, you have to draw a line. How many cases are needed? Rabbi says two is enough. Rabban Shimon ben Gamliel says three. But obviously this is an arbitrary standard. It is not some determination that in every place the same evidentiary force, the same force of habituation, is created by three times. That is certainly not true. Rather, you have to put the line somewhere, because each place is different; you can’t function like that. So Jewish law says: okay, let’s establish that if something repeats three times, the halakhic assumption is that it is apparently a general law. That’s it. Either habituation or evidence. And then I have no problem with the Chazon Ish saying that there are presumptions that are habituation and presumptions that are evidence, and nevertheless the Talmud compares these to those. It says: if in this case we go after three times, then in that case too it will be after three times. If in this case it’s two times, then in that case too it will be two times. And why? Because the arbitrary standard according to Rabbi is two times, and according to Rabban Shimon ben Gamliel it is three times. Somewhere you have to cross the line, put the line—yes, like the heap paradox. So according to Rabban Shimon ben Gamliel you put the line at three, according to Rabbi at two.

Yes, an example of this—I think I mentioned it, didn’t I?—about a three-year-old girl whose virginity returns. The Shakh brings a Jerusalem Talmud. The Talmud says that if someone raped a girl under the age of three—or not raped, had relations, it doesn’t matter, because in any case it’s rape, yes, intercourse with a minor—then her virginity returns, meaning he has not damaged her. Because the fact that he tore the hymen doesn’t matter, because until age three it returns. Beyond age three it does not return, and then he has to pay and all those things. The Jerusalem Talmud asks: what happens if the court intercalated the year? So let’s say the girl’s third birthday is on the 10th of Adar, except now there is Adar I and Adar II; the court intercalated the year. So now age three is really in Adar II, on the 10th of Adar II, which chronologically is really three years and one month. If the court had not intercalated the year, then she would have been three on the 10th of Adar I, because there’s only one Adar. So the Jerusalem Talmud says, bringing some verse—how does it go? “I call to God Most High…” something like that—that everything is determined according to the court. What the court established is what determines the presumption. The Shakh says: and from this Jerusalem Talmud you see that, yes, the heart of kings and princes is in the hand—not in the hand of God, but in the hand of the court. Meaning, what the court determines changes nature. If the court did not intercalate the year, then the virginity returns until the 10th of Adar. If it did intercalate the year, then the virginity returns until the 10th of Adar II.

Now, with all due respect, I don’t want to say anything too sharp, but this is really totally absurd. That’s the mildest way I can put it. Totally absurd. There is no doubt that it’s not true. So what is going on? And you don’t even need to learn this from the Jerusalem Talmud. Rather, after all, girls—when does the virginity return? For some girls it will be at age two, for some at two and a half, two years and ten months, three, three and a half, four. There is some distribution, right? Now we don’t know, and we’re not going to perform medical tests—we also don’t know how to do medical tests for every single girl—to determine whether her virginity will return or not. So you have to draw a line. The line was set at age three. Take some average, or I don’t know exactly what, a median, whatever—it was set at age three. Now, if the court intercalated the year, nothing in physics or physiology changes. But still, since the line is an arbitrary line, it makes sense to say that when she reaches her third birthday, that is when her status changes. That’s all. And every year with the calendar, we’re not going to keep checking whether it’s a leap year or not a leap year. We go by the birthday. And when her birthday arrives, until then our assumption is that her virginity returns; from then on the virginity no longer returns. But not because some actual physiological event occurred—for after all, in any case we’re dealing with an arbitrary line. Who said this girl herself even reaches that point at age three? Maybe for her it happens at two years and ten months, or at four? There are girls like this, and girls like that. We set an arbitrary line at—again—three, by the way. Interesting, it’s always three. Right? Because there really is some tendency that if one has to set some arbitrary line, one goes with three. Maharal surely has all sorts of explanations for these matters, the special qualities of the number three. A well-known numerologist. But the claim is that we are dealing with an arbitrary line, and therefore don’t think that shifting this line changes something in reality. Rather, we simply want a simple theory, we want a simple legal or halakhic mode of conduct. The simplest way to proceed is to decide that it always happens on the third birthday. That’s all.

Now I say the same thing about presumptions. When we say that after three times there is a presumption, it is not because in all places three times bring me to evidence of seventy percent. And two times is evidence of sixty percent, and the argument is whether you need evidence of sixty percent or seventy percent. No. There are places where two times is forty percent and three times is fifty; there are places where two times is seventy and three times is eighty. So you have to set some line, and the line must be universal because conduct has to be simple, that’s all. Therefore the line—Rabbi says after two times, because two times is preferable too: even with two different explanations, they are less good than one explanation for two cases, so some preference is enough for me. Rabban Shimon ben Gamliel says: I want a more significant preference, I want three times. Okay? But not because he is aiming at some given percentage of probability in all contexts; that won’t give you the same percentage of probability. Rather, one has to set a line, so the line is at three cases.

So there is no difficulty for the Chazon Ish as to why the Talmud compares these presumptions to those, and if here it is three then there too it is three. Because whoever says that the arbitrary line is three will say that in all places. Not because they are really similar, but because he chooses to draw the arbitrary line after the number three, that’s all. Therefore there is no problem at all with comparing one thing to another. This is also not a biblical decree or anything of the sort; rather, one has to draw some line somewhere, so they draw it after the number three.

Now, it has to be understood that clearly there is some reasoning behind this determination. It is not arbitrary in the sense that it says nothing. Obviously the more cases there are, the more plausible it is. But how much more plausible, or to what percentage of probability I arrive with each number of cases—that changes from context to context. Just as with testimony. One witness is not believed; two witnesses are believed. Why not five? Obviously with five I have a higher level of certainty than with two, and with two I have a greater level of certainty than with one. So why did they establish two? And maybe Moses and Aaron are more reliable, so there one would be enough? And people who are not Moses and Aaron—maybe there should have to be seven? Why two? Obviously it’s an arbitrary line. They draw a line—but an arbitrary line with logic behind it. Two witnesses are better than one because there is cross-checking, like two pieces of evidence. If two witnesses bring testimony, that is stronger evidence than when one witness says the thing. So obviously it is stronger—but who said that it is enough? The line is drawn at two witnesses. Not only that, they even insist that “two are like a hundred.” Meaning, even if there are a hundred against the two, we will not recognize the superiority of the hundred over the two. A hundred is like two. It’s like two against two. To that extent Jewish law insists that this arbitrary line has to be preserved. Two witnesses are the maximum degree of evidence halakhically. Of course that is not factually true, but halakhically we regard it as maximum evidence. Jewish law often draws arbitrary lines because it has to decide. But that does not really mean that essentially in all places it proceeds in exactly the same way.

[Speaker C] Now I want to move to the next part, to go back to the hypothesis, and there we begin to touch on abductions, and problematic abductions, through the presumption of three occurrences. So let’s begin, because I’m not sure we’ll get through all of it, but let’s start. There is a Mishnah in Bava Kamma 23b: “What is an innocent ox and what is a forewarned ox?”

[Rabbi Michael Abraham] “A forewarned ox is one concerning which testimony was given on three days, and an innocent ox is one that refrains for three days.” You see here too: both the return and the going-forward require three. And that’s the difficulty I mentioned in the name of Keren Orah—why does the return also require three? Those are the words of Rabbi Yehuda. Rabbi Meir says: “A forewarned ox is one concerning which testimony was given three times,” and the rule for an innocent ox is something else. So what is the dispute between Rabbi Meir and Rabbi Yehuda? Rabbi Yehuda requires that there be three testimonies about three separate days on which the ox gored. Rabbi Meir says that as long as testimony was given about it three times, it becomes forewarned, even if those three times were all on one day. It does not have to be on three different days.

The Talmud says: What is Rabbi Yehuda’s reason? Abaye said: he learns it from a verse. Abaye said: “‘Yesterday’—one, ‘the day before’—two, ‘the third day back’—three, and ‘its owner did not guard it’—this brings us to the fourth goring,” for which he pays full damages. Rava said: “‘Yesterday and the day before’—one, ‘the third day back’—two, and ‘its owner did not guard it’—today he is liable.” Fine, it doesn’t matter right now; it’s derived from a verse. And what is Rabbi Meir’s reason? Why does Rabbi Meir say that even on the same day there is a presumption? We see from the verses that it has to be on three different days. As it was taught, Rabbi Meir said: “If when he spaced out his gorings he is liable, then if he concentrated his gorings all the more so.” Rabbi Meir says: I agree with the exposition of the verse. The verse says that if it happens on three different days the ox becomes forewarned. But if on three different days it becomes forewarned, then certainly three times on one day it becomes forewarned. What the Torah teaches me by saying that over three days it becomes forewarned is not that only over three days it becomes forewarned, but that even when it happens over three days it is forewarned. But if it happens on one day, then obviously it becomes forewarned. It becomes forewarned even more on one day than over three days. Okay?

Now that is the dispute between Rabbi Meir and Rabbi Yehuda. The Tur, in Orach Chayim section 114, brings a dispute among the medieval authorities (Rishonim) regarding saying “He causes the wind to blow and the rain to fall.” In a moment it will sound like some kind of riddle, but soon you’ll see the connection. “Jerusalem Talmud,” says the Tur. “One who prays and does not know whether he mentioned it or not”—he doesn’t know whether he said “grant dew and rain” or “He causes the wind to blow and the rain to fall”—“before thirty days, the presumption is that he mentions what he is accustomed to. From then on, he mentions what he ought to.” Right? Meaning, within thirty days after the wording changes—whether on the fifteenth of Cheshvan or the seventh of Cheshvan or on Shemini Atzeret, in Tishrei or in Nisan—then for the month after the date on which the wording changes, within that month the assumption is that you are still accustomed to the previous period. After that month, apparently you are already saying what you ought to say, not what you were accustomed to, but what is actually correct to say.

Where is the practical difference? So he says: “Within thirty days”—I’ll keep reading here—“if he is uncertain whether he mentioned it or not, he must repeat, because presumably he did not mention it.” And the same applies thirty days after Passover: presumably he did mention it, and therefore he must repeat. “And Maharam of Rothenburg”—okay? So the Talmud says that the practical difference is in cases of doubt. If you’re in doubt, you don’t remember whether you said “grant dew and rain for blessing” or “He causes the wind to blow and the rain to fall,” or whether you said “He causes the dew to fall.” You don’t remember which of the two you said. Right, those with chariots and those with horses, as is well known. You don’t remember what you said, so you’re in doubt. So the Talmud says, the Jerusalem Talmud says: if you’re in doubt, you can assume that you said what you’re accustomed to. Within thirty days, you’re not yet used to the new wording, so assume you said the incorrect thing, what belonged to the previous period. After thirty days have passed, the assumption is that now you are already saying the correct thing. And if you are in doubt, assume that you do not need to repeat; assume that you said the correct thing.

Now he says: that is the Jerusalem Talmud. “And Maharam of Rothenburg would customarily say on Shemini Atzeret the blessing ‘You are mighty’ ninety times up to ‘He causes the wind to blow and the rain to fall,’ corresponding to thirty days in which he says it three times each day. And now, if he were in doubt, he would not need to repeat.” Maharam of Rothenburg says: why wait thirty days and every day say “He causes the wind to blow and the rain to fall” three times? On the first day, I’ll say ninety times: “He causes the wind to blow and the rain to fall, He causes the wind to blow and the rain to fall, He causes the wind to blow and the rain to fall,” ninety times. I will already enter into the presumption that I am accustomed to saying “He causes the wind to blow and the rain to fall,” and now there’s no problem at all. If I’m in doubt, I don’t need to repeat; it solves all the problems. All the laws of doubt are resolved.

And his proof is from the chapter “How the Foot [Causes Damage]” in Bava Kamma, which says regarding a forewarned ox: “If when he spaced out his gorings he is liable, then if he concentrated his gorings all the more so.” What does Rabbi Meir say? If doing it over three different days makes it forewarned, then three gorings on one day certainly make it forewarned. Right? So Maharam of Rothenburg says: the same here. Since after thirty days, if he is in doubt, he need not repeat because he has become accustomed, then all the more so ninety times in one day. “If when he spaced out his gorings he is liable, then if he concentrated his gorings all the more so.” He compares it to the presumption of a forewarned ox. This is basically the three times over. Only here it’s not three—it’s thirty times three. Three prayers each day, thirty times. By the way, it’s not even three times each day, because on Sabbaths, for example, the formulations are different, and there are also more prayers, so that’s a separate issue. It’s clear here that it is a bit arbitrary. But never mind, the idea is like the presumption of three occurrences, only here the unit is—call it ten days, let’s say—and that’s the equivalent of one time, so three times gives you thirty days. Fine? So Maharam of Rothenburg says: just as with a forewarned ox, instead of waiting thirty days, each with three prayers, let me say it ninety times on the first day and I’ll be out of doubt immediately; I’ll already be accustomed to saying the correct wording.

And Rabbi Rabbeinu Peretz, of blessed memory, wrote: “I have not seen the elder rabbis of France do this.” Right, our rabbis in France did not do this. Maharam of Rothenburg is Germany, yes, so our rabbis, the masters of Tosafot in France, we did not see doing this. Now, the fact that they didn’t do it—maybe they just didn’t think of the idea? Rabbeinu Peretz apparently understood that they maintain it is ineffective—not simply that they don’t do it, but that they don’t do it because it doesn’t work, not because they didn’t think of the idea. Why not? So he says: “The case is not comparable to the proof.” Right, what was Maharam of Rothenburg’s proof? From the forewarned ox: if spaced gorings make him liable, then concentrated gorings all the more so. He says: the case is not comparable to the proof. Why? “For there, the reason regarding the forewarned ox is because it has been established as one that gores, and if it was established by three spaced-out gorings, all the more so by three close-together ones. But rain, which was instituted in prayer, and the matter depends on the habit of his tongue, we do not say this.” And my master my father, the Rosh, of blessed memory, tended toward the reasoning of Maharam. And his student too, of course.

What is the argument really about? Rabbeinu Peretz says: why are you bringing me proofs from an ox goring and “grant dew and rain for blessing”? In the case of the ox goring, we are discussing the question whether it is established as one that gores. What does “established as one that gores” mean? So the common explanation—we’ll come back to this, but I’m saying it briefly—is that this is basically evidence that it has a goring nature. And there are three times. So if it gored on three spaced-out occasions, that is evidence that it is a goring ox; then if it gores on the same day itself, certainly that is evidence that it is a goring ox. But in the prayer for rain, and in “He causes the wind to blow and the rain to fall,” there it is clear that these repetitions are not evidence that by nature I am accustomed to saying “He causes the wind to blow and the rain to fall.” There it is habituation, not evidence. Who said that in habituation too we say that if it happens at a distance then all the more so it will happen when close together? For evidence that is true, but who said that for habituation it is true? So on the face of it, this is how the later authorities usually explain it: that the dispute between Maharam of Rothenburg and Rabbeinu Peretz is over the question of how to understand the presumption of the forewarned ox—whether it is evidence or habituation.

[Speaker C] Okay, I’ll stop here, because from here it starts becoming calculations again. I’ll go over that last passage of the Tur next time and continue from there. That’s it for now. Any comments or questions? So, if you want. Okay then, goodbye, have a peaceful Sabbath.

[Speaker B] Good night, peaceful Sabbath.