Probability and Statistics – Lecture 14
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
Table of Contents
- Statistical evidence versus eyewitness testimony in law
- Three types of reasoning for evidentiary policy
- “A person cannot render himself wicked,” self-incrimination, and the disqualification of relatives
- The relationship between intuition, rationality, and intellectual honesty
- The human legislator, the halakhic legislator, and the limits of “this requires further analysis”
- Creative derivations and supportive derivations as an analogy for “limping” explanations
- The article by David Enoch and his colleagues: blue/red buses and two kinds of explanations
- An epistemic explanation, the lottery paradox, and sensitivity to facts
- Rejecting the epistemic explanation and moving to a utilitarian explanation
- Summary of the direction: reasoning of the third type
Summary
General overview
The text argues that there is a fundamental distinction in the law of evidence between statistical evidence and evidence like eyewitness testimony, even though in both situations the level of certainty may be similar, and it suggests that the distinction does not stem from the strength of the evidence but from a different kind of reasoning. It divides reasons for evidentiary policy into three types: inadmissibility because of weak reliability, inadmissibility due to consequential legal considerations, and inadmissibility arising from legal intuition or a sense that it is “not appropriate,” even without a reason of the first two kinds. Through a discussion in Jewish law (“a person cannot render himself wicked,” disqualification of relatives) and through an article by David Enoch and his colleagues about “blue/red buses” and the lottery paradox, it is argued that many philosophers and legal scholars reject epistemic explanations or offer “limping” utilitarian explanations because they are unwilling to recognize this third kind of reasoning, and the direction at the end is to return to reasoning of the third type.
Statistical evidence versus eyewitness testimony in law
The text presents a case of one hundred prisoners, ninety-nine of whom attacked a guard, and argues that despite the 99% probability that the defendant participated, in such a case one does not convict on the basis of statistical evidence alone. It contrasts this with a case of eyewitness testimony, where even though the testimony is not one hundred percent certain, the default is to accept the testimony and convict unless it is shown that the witness erred or lied. It argues that the difference does not stem from the level of certainty but from something else, and later it retracts the idea that this is “really in the statistical sense” and prefers to see it mainly in terms of legal explanations.
Three types of reasoning for evidentiary policy
The text divides reasons for refusing to accept evidence into three types. The first type is defined as a problem of reliability or evidentiary strength, for example one witness who is not enough to convict. The second type is defined as a consequential legal consideration, such as “fruit of the poisonous tree,” where good evidence is excluded because of systemic consequences and negative incentives to obtain evidence by illegal means. The third type is defined as a “legal intuition” or legal sense of “it just doesn’t fit,” which has no explanation either in terms of reliability or in terms of future consequences, yet is still perceived as correct and rational as a source of normativity.
“A person cannot render himself wicked,” self-incrimination, and the disqualification of relatives
The text presents the Talmudic rule “a person cannot render himself wicked” and contrasts it with the usual legal explanations in general law, such as fear of torture or insanity, including the example of the Miranda ruling in which an American judge cites Maimonides. It argues that in Jewish law these explanations do not fit, because the Talmud is not dealing with interrogation rooms and torture, and it is even possible that a person confesses out of repentance and fear of Heaven, yet Jewish law still does not accept self-incrimination. It suggests that this functions like a scriptural decree in the sense that there is a normative feeling that “it is not appropriate to convict a person on the basis of his own testimony about himself,” and it cites the Ritva, who explains the disqualification of relatives from testimony in a similar way, as analogous to a person testifying about himself. It emphasizes that a scriptural decree is not a “matter without reason” in the sense of arbitrariness; rather, sometimes the reason is not of the usual types but a normative principle that is hard to formulate as a consequential or probabilistic explanation.
The relationship between intuition, rationality, and intellectual honesty
The text presents two possibilities: to reject feelings and act only according to “logic,” or to trust legal/moral intuition, which even without an explanation grasps something real. It argues that Pappino is an example of honesty in that he rejects explanations and admits that he remains with “this requires further analysis,” yet still agrees not to accept statistical evidence, instead of abolishing the policy. It claims that others offer “limping” explanations in order to justify an existing intuition because they are unwilling to recognize the category of valid intuition without rationalization of the first two types, and in this way they “force” explanations and even portray Jewish law as “broken” when the explanation does not fit its sources.
The human legislator, the halakhic legislator, and the limits of “this requires further analysis”
The text addresses the claim that the absence of an explanation simply means “this requires further analysis,” and emphasizes that the question is what the legislator had in mind when adopting the law. It argues that it is not plausible that human legislators had some deep explanation that no philosopher or legal scholar can discover, and therefore it is likely that they acted out of normative intuition. It adds that even in Jewish law the rule is not “written in a verse” but is built on the interpretation of the Sages, and even if there were a verse one should still ask why it is so; legal intuition is then seen as grasping what was called “religious values” even when there is no explanatory formulation.
Creative derivations and supportive derivations as an analogy for “limping” explanations
The text presents a distinction in the world of scriptural derivation between creative derivations, which generate a new law, and supportive derivations, which try to support a law already present in the tradition. It argues that there is a tendency to say that when a derivation is merely supportive, one may make do with a “limping” derivation, but this creates a difficulty: if the law was once generated in the past through a derivation, then that derivation was once creative, and it is not plausible that a newly innovated law would rest on a weak derivation. It suggests that sometimes it is preferable to remain with “this requires further analysis” regarding the derivation rather than adopt a derivation that “doesn’t hold water,” and it connects this to the tension between a law given to Moses at Sinai and “they learned it as a tradition,” according to the Netziv in Kedmat Ha’emek, as opposed to Rashi’s identification of the two, and to a unique view of Maimonides regarding support from Scripture that elevates the law’s status.
The article by David Enoch and his colleagues: blue/red buses and two kinds of explanations
The text opens the discussion with the article by David Enoch and his colleagues using the example of a city with two bus companies, blue buses 95% and red buses 5%, where an accident occurs and there is no direct testimony about the color of the bus. It argues that the legal intuition is that one does not require the blue company to pay on the basis of the statistical majority alone, but one does require payment when there is eyewitness testimony even if it is not certain. It notes that perhaps in religious law it is easier to accept reasoning of the third type because of “religious values,” but leaves open the possibility that even in secular law there may be value-based reasoning that is neither consequential nor probabilistic.
An epistemic explanation, the lottery paradox, and sensitivity to facts
The text presents Enoch and his colleagues’ distinction between an epistemic explanation and a practical-utilitarian explanation, and emphasizes that “epistemic” for them is not simply a matter of probabilistic strength. It describes their claim that the difference between statistical evidence and eyewitness testimony is that statistical evidence does not provide “knowledge” but only probability, whereas eyewitness testimony can count as knowledge. It brings the lottery paradox: a person with a one-in-a-million chance cannot say “I know I didn’t win” merely from the statistics, but if he read the result in the newspaper, he does say “I know,” even though there is a similar possibility of error. It details the analysis of “sensitivity” via counterfactuals: a statement based on statistics would not change even if the facts were the opposite, and therefore is not sensitive to the facts, whereas a statement based on an information source like a newspaper or an eyewitness should change if the facts were different, and therefore counts as knowledge, even though in practice it can still be mistaken.
Rejecting the epistemic explanation and moving to a utilitarian explanation
The text explains that the article itself raises the question of why it should matter at all whether we are dealing with knowledge or only probability, if the level of certainty is similar, and therefore their conclusion is that an epistemic explanation cannot justify the legal distinction. It argues that they conclude that the epistemic explanation lacks normative force because it fits neither the category of probabilistic reliability nor the category of consequential legal consideration, and they ignore a third possibility of normative intuition without explanation. It summarizes that they then move on to propose a practical-utilitarian explanation, and the speaker declares that he himself will later argue that the utilitarian explanation “doesn’t hold water” and is an example of a “supportive derivation” accepted only because there is a prior intuition against statistical evidence.
Summary of the direction: reasoning of the third type
The text states that the epistemic explanation is rejected both by Pappino and by Enoch and his group, and that the utilitarian explanation Enoch proposes will later be rejected as a “limping” explanation. It argues that if one rejects limping explanations even as supportive derivations, what remains is the possibility that there is not always an explanation in terms of reliability or consequences, but rather a normative reasoning of “this is not how one does things” that sustains the rule. It ends with questions from the audience about the availability of the article (only in English, with a link in column 228) and about the definition of “reasoning of the third type” as a situation in which “there is no explanation and yet it still must be upheld,” in the sense that there is no explanation of the first two types, and the speaker confirms that this is not irrational but an expression of the intuition that “it just doesn’t fit.”
Full Transcript
[Rabbi Michael Abraham] Okay, we’ll begin. At this stage in the series we’re dealing, yes, with doubt and statistics, but at this point we’ve been dealing with legal evidence, statistical evidence in law. And the basic distinction for which I presented an explanation is the distinction between two situations in which in both of them I have a very high degree of certainty regarding what I’m claiming, and nevertheless in one case it’s accepted and in the other case it isn’t. That means this distinction is not a distinction in the degree of certainty or in the strength of the evidence; there’s something else here. Right? The example was the one with the hundred prisoners, ninety-nine of whom attacked a guard and one did not. We have a 99 percent chance that the defendant before us, one of the ten—sorry, one of the hundred—we have a 99 percent chance that he participated in the incident, and therefore apparently we can convict him. But no, in such situations you don’t convict. On the other hand, if there’s an eyewitness who says that this prisoner participated, then even though with respect to the eyewitness there’s also some doubt—maybe he missed something, eyewitness testimony is not one hundred percent—still we accept it unless it’s proven that the eyewitness erred or lied or something like that, but the default is yes, to accept the eyewitness testimony. And the question is why, because in both cases the chance of error is the same chance. Why is it that in the second case we convict and in the first case we don’t convict?
So I presented various explanations; I won’t go into them again. Explanations on levels where at first I thought maybe it was even really in the statistical sense, but I retracted that, and I really think these are more legal explanations than statistical explanations. But now I want to sharpen this point further, and I’ll do that through a discussion of an article by David Enoch and his colleagues. They wrote an article exactly on this topic. And before I get into that, I’ll just remind you of an important point we discussed.
I said that when we come to justify some legal policy, say regarding the law of evidence, then the justifications can be divided into several types. When I don’t accept evidence, why don’t I accept it? One possibility is because it isn’t reliable enough. One witness is not enough; two witnesses are enough. Fine. One witness—it’s plausible that what he says is true, but it’s not enough to convict. We don’t accept it. Meaning, there is evidence that simply isn’t strong enough, and that’s why we don’t accept it. There is evidence that we don’t accept for legal reasons—for example, fruit of the poisonous tree. Right? In Israel it’s less so; in Israel they generally do accept it, I think, but in the United States, for example, they don’t accept evidence obtained illegally. If the evidence was obtained illegally, they don’t accept it.
Now we’re talking about a case where the evidence is good evidence. If the evidence weren’t good, then we wouldn’t accept it because it isn’t good. I’m talking about good evidence that we don’t accept because it was obtained illegally. What’s the idea here? After all, apparently the evidence is good evidence, and if you want to know whether so-and-so is guilty or not guilty, you have good evidence, so use it. The answer is that there are legal considerations there: if you accept such evidence, that gives people—or the police, or whoever is interested—an incentive to use illegal methods in order to obtain evidence. And therefore, if you think that law is a justified law and you want to make sure people have no motivation to violate it, then you have to impose some kind of sanctions on those who violate it—or not sanctions, but restrictions on those who violate it. And that’s why you don’t accept fruit of the poisonous tree. The consideration is what will happen in the legal system generally, what will happen in future cases. There is no explanation on what we might call the statistical level that this evidence is bad, like in the first type of explanation. This is a second type of explanation. Let’s call it, say, a legal explanation, a legal consideration.
The third explanation—I called it legal intuitions, so I need to find some term here that distinguishes them. So the second we’ll call a legal consideration. Meaning, there is some very clear logic, except that this logic is not about the reliability of the evidence but about the consequences of what will happen if I accept the evidence. And these are measurable consequences; we all understand them, they’re logical, you can formulate them, you can explain them. So it’s perfectly rational; there’s nothing here of a scriptural decree, right, there’s no problem here, it’s very, very logical.
The third possibility, whose existence I want to argue for, is some kind of legal intuition. I called it a legal reasoning or a legal intuition, as distinct from a legal explanation. Right? The previous one was a legal explanation; here I’m saying I have some kind of intuition that this is simply not the right way to act. Now I don’t have an explanation in the consequential sense, meaning I don’t have an explanation of the form: if I do this, then such-and-such will happen, and I want to prevent that, like I said before about fruit of the poisonous tree. Certainly I don’t have an explanation in terms of the reliability of the evidence, and yet I still have some kind of feeling, some kind of intuition that says—take, for example, “a person cannot render himself wicked.”
So “a person cannot render himself wicked,” in the legal aspect, in the legal context, people often bring various legal, evidentiary explanations of one kind or another. The claim is that if a person renders himself wicked, then maybe he went insane—it can’t be; why would a person convict himself and put himself in jail? Or alternatively, that he was tortured by the police and therefore confessed. Since that’s so, maybe we question his confession and we are unwilling to accept self-incrimination. Various explanations more or less of that sort. But in Jewish law—and I said, I already mentioned this—I said that in the Miranda ruling, right, that’s the Fifth Amendment, I think, or the second, I don’t remember anymore, to the American Constitution, with the Miranda warning that you always see in movies: you can remain silent, we’ll appoint you a lawyer if you don’t have money, and so on—when someone is arrested, the officer always recites the Miranda warning to him. So it’s called the Miranda warning because it began with the Miranda ruling, and there the non-Jewish American judge cited the Talmudic rule that a person cannot render himself wicked, and brought Maimonides, who rules that way, and everything. But he explained it with legal explanations—maybe torture, maybe he went insane, or things like that.
By the way, maybe he went insane—that’s an explanation that appears in Maimonides, but Maimonides himself is contradictory on this matter, so I don’t want to get into that. In any case, he gave legal explanations, and therefore Israeli jurists too, when they bring the rule that a person cannot render himself wicked, always bring it—not always, but in the cases I’ve seen—they bring it from the Miranda ruling through the gentiles indirectly, and they pass it along as some kind of legal explanation, yes, the explanations I mentioned earlier.
In Jewish law it is very clear that this is not correct. The source for “a person cannot render himself wicked” contains no concern at all that the person went insane. On the contrary, maybe he decided to repent, and therefore he confesses his sin and is even willing to bear the punishment. A righteous person—there is an Eye in Heaven that sees, right, I don’t know exactly what—fear of Heaven can cause a person to return and confess the sin he committed. And as for the concern about torture, that’s obviously irrelevant, because the Talmud is not talking about a situation with police interrogations. There is no police force. In Jewish law there is a court; the court hears the witnesses, yes, examines them in the religious court. There’s no torture, there’s nothing; there are no situations in which they put them in jail and don’t let them sleep unless they admit whatever they’re supposed to admit. None of that exists. Everything takes place in the courtroom itself; there are no dark interrogation rooms where nobody knows what is happening inside. So all those concerns do not exist, and nevertheless the Talmud says that a person cannot render himself wicked.
And apparently, if I understand the passage correctly, this is called a scriptural decree. But what stands behind this scriptural decree? What stands behind this scriptural decree is some kind of feeling, I think—I at least have such a feeling, such an intuition—that says it is not proper to convict a person on the basis of his own testimony about himself. It is not proper to accept self-incrimination. If you ask me why—I have no explanation. Certainly I have no statistical explanation in the sense that the evidence is bad; on the contrary, a confession is the queen of evidence. I have no legal explanation, what I called the second type; I don’t see what the problem is if I accept self-incrimination. I don’t see any consequence that I am preventing by refusing to accept self-incrimination. I said earlier why in Jewish law all those concerns are irrelevant, so what remains? So what is this scriptural decree? Is the scriptural decree a matter without reason? The Torah just decided this because it felt like it? After all, if the Torah said this, then presumably there is some sort of logic behind it. Things are not determined for no reason, even in Jewish law.
Years ago I spoke about scriptural decrees and explained that a scriptural decree is not a matter without reason. It is a matter without a reason that I understand, perhaps—and even that is not always true. Certainly there is some reason behind these things; the Holy One, blessed be He, does not simply exercise judgment without justice, right? He doesn’t just decide things arbitrarily because He feels like it. Why would He suddenly produce a twisted legal rule just because He feels like not accepting self-incrimination? After all, you want justice to come to light. If the person is guilty, then why not accept his testimony about himself and act accordingly? There’s no logic in that—unless there is an explanation either on the level of the underlying idea or on the legal level. But if there is no explanation, then you say: you have a feeling? Take a pill, get over the feeling. What is this feeling?
No. It turns out that people have some kind of intuition that this is simply not the right thing to do. I don’t have explanations of the first two kinds, neither statistical nor legal, but I trust my intuition that if it says this, then apparently it is grasping something real. Not that there is an explanation and I just don’t know it, a legal or probabilistic explanation. No. Rather, this is simply not the proper way to behave. Not because there are such-and-such consequences, and not because this testimony is false, but because it is not proper to convict a person on the basis of his own testimony about himself.
I already mentioned that Ritva, who explains this way also the testimony of relatives. That the testimony of relatives is also disqualified because it is like a person’s testimony about himself. A relative is in some sense part of me, so when the relative testifies about me, it is as though I myself am testifying about myself. And that seems to him a good explanation for why the testimony of relatives is disqualified, while the Talmud says that this is a scriptural decree, the disqualification of relatives from testimony. We see that what is called a scriptural decree in this context is some kind of explanation based on a certain legal intuition. It is clear to me that it is not right, legally, to act this way, even though this has no problematic consequences and there is also no problem with the reliability of the evidence. It is neither of the first two explanations.
What is this thing? It’s a bit hard to define, but a kind of “not appropriate,” it is not appropriate to act this way. It just doesn’t fit, right? You don’t do things like that. I’m repeating myself and waving my hands—I once told you what my study partner says when I start waving my hands. But that’s how it is; there are feelings like that. And the question is: we have two possibilities here. One possibility is to say, okay, feelings are feelings, take a pill, get over the feeling, and we are supposed to act according to logic, not according to feelings. Feelings may be built into us in some way—so what? So because of that I won’t act according to logic? A person is supposed to act rationally. That’s one possibility.
The second possibility says no: maybe there really is something here that is not proper. I don’t know how to conceptualize it and define it or justify it, but it’s still not right—you just don’t do it. This is not how one acts. And I trust this legal or moral intuition, and therefore I will go with it even though I have no explanation. And I don’t see this as an irrational move. Why? I already said these things earlier, but I want to sharpen them because I want to show you that this operates for many people, philosophers and legal scholars alike.
The best example is Pappino’s article, the article I mentioned last time. Pappino, who discussed why statistical evidence is not accepted. He raised various possibilities, rejected them all, and remained with “this requires further analysis” as to why he himself agrees that such evidence must not be accepted. But it never occurred to him to say, okay, true, this is accepted policy in the world, but it’s all nonsense, we should abolish it. It’s not logical. So what if we have that feeling? We should overcome it. If I found no logical explanation, then the conclusion ought to be: let’s abolish it. So yes, we should accept statistical evidence. No—he never even entertains that. He remains with “this requires further analysis”: really, why? I don’t understand why not, but I myself also think no. And I think that’s a very nice demonstration of the phenomenon I was just talking about. He just didn’t know how to define it in this way.
The claim is that there are things you simply don’t do. Like, I don’t know, cutting in line. One can perhaps offer some kind of explanation; I’m not sure it will always work, and I’m not sure the explanation is “it’s unfair.” You arrived later, so stand behind those who came before you. Don’t cut in line. Why? If someone asks why, I’m not sure I have such a good explanation to give him. I have this natural sense of justice that whoever arrives first should go in first. In that sense I mean to say also in the legal context: I have some sense that this is the right way to act, or that this is the wrong way to act, and I don’t dismiss that sense. The fact that I didn’t find an explanation for it doesn’t mean it isn’t correct. Not because I’m not smart enough and could have found one—no. It may be that there is no explanation. There is no explanation in consequential terms or in terms of reliability. But I still think it is right to act this way. It’s a subtle point, but one has to notice it: it’s a third kind of explanation, and I think that’s what it is.
And once people ignore this possibility, what they do is actually something that isn’t very intellectually honest. What do I mean? You have a feeling, say, that no, it is not right to rely on statistical evidence. You didn’t find a sufficient explanation. Pappino was honest; he said, listen, I didn’t find one. Apparently we ought to abolish it, and I don’t understand why I myself don’t agree that it should be abolished. So he remains with “this requires further analysis.” But there are many others, in my opinion, who do offer explanations, and the explanations are limping. But because they have the intuition that this is the right way to act, they offer that explanation despite the fact that it limps. Or they look for some consequential explanations of one kind or another; they explain why relatives are inclined to lie—that’s how the disqualification of relatives from testimony is often explained, while the Talmud itself says explicitly—I’m speaking in Jewish law—that the Talmud itself says explicitly that this is not true. So they force things here and there because they are unwilling to accept this third type of explanation. Therefore it has to be mapped onto one of the first two. Ah, but we don’t find a good explanation of one of the first two types—so they wriggle. So they do something strained, but they have to explain something here. Because on the one hand they won’t give up this intuition even though they have no good explanation, and on the other hand they are unwilling to acknowledge the existence of such a category of intuition without explanation. So what do they do? They find some limping explanation and say, fine, apparently that’s the explanation. And afterward, of course, if you investigate Jewish law, they offer some limping explanation and then say, but it limps, therefore clearly Jewish law here doesn’t work properly. There’s something defective in halakhic thinking. To me it’s the same thing. It’s simply an unwillingness to recognize this possibility—that there are some intuitions that have validity even if you haven’t found a rationalization for them in terms of the first two types of explanation. As an illustration of that.
[Speaker B] Rabbi, I didn’t understand why this is a new category. Why do you say it’s a third category? It’s just that you don’t have an explanation and it requires further analysis, that’s all.
[Rabbi Michael Abraham] No, it doesn’t require further analysis. I’m not remaining with “this requires further analysis.” If I were remaining with “this requires further analysis,” then I wouldn’t use this category—or at least not sufficiently. But no, I do use this category. I won’t accept self-incrimination or I won’t accept statistical evidence despite the fact that it requires further analysis, so it’s not just “requires further analysis.”
[Speaker B] Because there’s a law not to accept self-incrimination, but the explanation for the law—you, I’m talking about the legislator.
[Rabbi Michael Abraham] There’s a law about that. About why the legislator adopted this law.
[Speaker B] Yes, but once he adopted it, fine—
[Rabbi Michael Abraham] I’m asking about the legislator, not once he adopted it. I’m not asking about you. I’m asking why the legislator adopted this law.
[Speaker B] It requires further analysis. You don’t have an answer, so it requires further analysis.
[Rabbi Michael Abraham] No, it doesn’t require further analysis. The question is about the legislator himself, not about me. What did he tell himself? I may not have understood him—that’s not the point. I’m talking about what the legislator himself thought when he adopted it. After all, the legislator isn’t smarter than all the jurists. He didn’t have some explanation that all the jurists in the world couldn’t find. So why did he do it? Something here can’t be right. This is not merely remaining with “this requires further analysis.” You, as a legal scholar, can remain with “this requires further analysis,” because you are not the legislator; you don’t make the law and you don’t determine whether it should be obeyed. But the legislator cannot remain with “this requires further analysis.” If he is in that state, he simply won’t legislate it, that’s all.
[Speaker B] No, but he had a reason that I don’t know.
[Rabbi Michael Abraham] So I’m saying: if the legislator is flesh and blood like me and like you, and all the jurists and philosophers in the world don’t find a reason—let’s say, I’m not sure David Bitan, I have such great confidence in the laws he legislates, sorry that I picked on him in particular, but never mind, any one of the hundred and twenty—do I have such confidence in their supreme wisdom that even though all the jurists and philosophers in the world can’t find an explanation, they must have thought of something deep that no one else came up with? Not likely. And they themselves probably also didn’t think they had an explanation, but in my opinion they had some intuition that this is the right way to act, this is the right thing to do.
[Speaker B] Yes, that’s true regarding a human legislator, but regarding the rule of no self-incrimination, that comes from the Holy One, blessed be He, because it’s a scriptural decree.
[Rabbi Michael Abraham] No, this thing is not from the Holy One, blessed be He. This thing comes from the interpretation of the Sages, which is a Torah-level interpretation, but it is not written in any verse. If the Sages had not decided this, it would not exist in Jewish law. There is no verse that says this. The Sages understand: a person cannot render himself wicked, and a person is considered close to himself, and there are major disputes among the medieval authorities (Rishonim) about what exactly that whole double expression means. But there is something here. And even if it were from the Holy One, blessed be He Himself and written in a verse, you would still have to ask yourself why He did it.
[Speaker B] Yes, but that can’t be known. He has religious values that I don’t know.
[Rabbi Michael Abraham] Exactly. So I’m saying: what my legal intuition is really sensing is what you just called religious values, only I never know how to explain them. Maybe the Holy One, blessed be He—I suppose—did know how to explain them. I don’t know how to explain them, but that doesn’t mean I’m left with “this requires further analysis” about them. I think it’s correct.
[Speaker B] No, the “requires further analysis” is not about whether it’s correct, but about the explanation behind it.
[Rabbi Michael Abraham] The explanation—I don’t care. But the correctness is correctness. So that’s why this is a third type.
[Speaker B] No, Rabbi, I didn’t understand. It’s simply not a third type. It’s just that you don’t have a reason. You don’t have an explanation for it, but you obey it because—
[Rabbi Michael Abraham] There is no explanation for it, not just that I don’t have an explanation for it.
[Speaker B] Yes, no, you don’t have an explanation for it.
[Rabbi Michael Abraham] No—there is no explanation for it. Not an explanation in terms of the first two types.
[Speaker B] Who says? Maybe you just didn’t find it.
[Rabbi Michael Abraham] So I’m saying, again: look at other legal systems, not the Torah’s. Other legal systems are legislated by human beings, and there too legal scholars operate in the same way. So there is no reason to say that in the halakhic context it works differently, even though our legislator is wiser. And even that isn’t certain, because it’s not the Holy One, blessed be He, but rather the overwhelming majority of Torah-level laws were established by the Sages, not by the Holy One, blessed be He. Human beings established them. So I’m saying there is such a move here; you can’t ignore it.
There may be situations where there is a very wise legislator and he has explanations that I haven’t arrived at—explanations, I mean in the usual sense, consequential explanations or explanations about the admissibility of evidence and all that. But there are other situations as well. And I began to say that I think there is an interesting analogy to this issue with regard to derivations. In the world of derash they divide between two kinds of derivations: there are creative derivations and supportive derivations. Creative derivations are derivations that generate a new law. I expound a verse and from that I get a new law that wasn’t known until now; it is created now as a result of the derivation. A supportive derivation is a law that I received by tradition, and I’m trying to think what the derivation was, because it isn’t stated explicitly in the Torah, so it must have come from a derivation, but I don’t know from which derivation. There is a tradition from earlier generations that this is the law, but I have no idea what derivation they made. So I look for a derivation that will support the existing law. It doesn’t create a new law; it supports an existing law—or preserve a law, support a law, that’s what it’s called, never mind, same thing.
Now, it turns out in many places—and you even see this among the medieval authorities (Rishonim)—that since the derivation is a supportive derivation, it can also be pretty limping. Meaning, if there is a derivation that is problematic, then the medieval authorities (Rishonim) basically say: so apparently it is a supportive derivation. As if the law was already known and the derivation comes to support the law, because it is not plausible that they built a new law, one now binding on all of us, on such a limping derivation. Who says it’s correct if the derivation is so limping? But if it’s a supportive derivation, then maybe even a limping derivation is enough for me.
Now pay very close attention—this is not simple at all. Suppose this derivation really is limping in my view, and I received this law by tradition. It was innovated, let’s say, five hundred years ago. Fine? And then they made some derivation, but I don’t know what derivation they made. I try to reconstruct it and arrive at some fairly limping derivation. So what am I really saying? Since the law is correct, apparently this is the derivation, even though it seems limping to me. But this is still strange, because that generation, when it created this law, relied on some creative derivation. For me it’s a preserving derivation, but for the one who created the law, that derivation was a creative derivation. Now if he made the derivation that I think he made, then he used a limping derivation to create a new law. That can’t be. Meaning, I may allow myself a limping derivation because I know this law is a correct law, I received it by tradition, but that still cannot be right. Because if I’m correct that this is the right reconstruction of what the generation that created this law did, then it turns out that the generation that created this law relied on a limping derivation. But limping derivations are not enough to innovate a new law. That’s the proof that the derivation I found is probably not the correct reconstruction of what they did then.
In other words, it is hard to accept a limping derivation even when it is supportive. Better to remain with “this requires further analysis.” We received the law by tradition. I trust that the Sages who did this then probably knew what they were doing. But as for me, the best derivation I found is this limping derivation, so apparently it is not correct and I need to throw it out. And I will remain with “this requires further analysis”; I don’t know what the derivation was on which they relied. But to adopt the limping derivation I found just because it is supportive—that isn’t logical.
There are situations—it’s not completely categorical—there are situations where the derivation is limping, but I feel that maybe I missed something and someone smarter than I am could explain to me why this derivation is actually a good derivation. So maybe there it could work. But where it is clear to me that this derivation really doesn’t hold water, and I just didn’t find anything else, there I think that even as a supportive derivation it is not right to accept such a derivation. And medieval authorities (Rishonim) and later authorities (Acharonim) often explain: fine, this derivation is true, it’s not compelling, but it’s a supportive derivation. You can’t accept such a thing.
[Speaker B] But a supportive derivation would never have been a creative derivation.
[Rabbi Michael Abraham] Why not?
[Speaker B] It’s not that someone created it with this derivation and afterward it became a supportive derivation. It was never a creative derivation, because all the laws we received—not all, but the laws for which there is a supportive derivation—they were transmitted orally from the Holy One, blessed be He, to Moses and so on.
[Rabbi Michael Abraham] No, no, no. You’re making an incorrect assumption here. You’re talking about a derivation that supports a law given to Moses at Sinai. But I’m talking about a derivation that supports a law created in earlier generations. I once mentioned the Netziv in Kedmat Ha’emek, his introduction to his commentary on She’iltot, Ha’emek She’elah. So in that introduction, called Kedmat Ha’emek, a very long, very detailed, very fundamental introduction, there he distinguishes between a law given to Moses at Sinai and “they learned it as a tradition”—between that and an accepted interpretation, no?
[Speaker B] What? Between that and an accepted interpretation? No.
[Rabbi Michael Abraham] I don’t know—“accepted interpretation” is Maimonides’ phrase—between that and “they learned it as a tradition.” “They learned it as a tradition” and a law given to Moses at Sinai. Now “they learned it as a tradition”—usually Rashi, wherever it says “they learned it as a tradition,” Rashi says: a law given to Moses at Sinai. But Maimonides—as the Netziv argues—explains “they learned it as a tradition” differently from a law given to Moses at Sinai. A law given to Moses at Sinai comes from the Holy One, blessed be He. “They learned it as a tradition” is a law created in earlier generations and passed down to us by tradition, but the source was not passed down to us. We have the law, we know that this is the law, I don’t know how they derived it, but it is not a law given to Moses at Sinai. Okay? Now I’m talking here about laws of that kind.
And if derivations support a law given to Moses at Sinai—those derivations, sorry, support a law given to Moses at Sinai—then indeed I don’t really care if it’s a limping derivation, though I don’t see the point of it. Why do it? You have a law given to Moses at Sinai and that’s it. A derivation where you are trying to reconstruct what the earlier generations did—that makes sense, because you also want to learn how to derive, you want to understand how they operated. So there is a point in doing that. But here, accepting a limping derivation is very problematic, even if the derivation is supportive.
[Speaker B] So then I don’t conceptually understand what a supportive derivation is. Because if we’re talking about a law we received from Moses, a law given to Moses at Sinai, then basically we don’t need it, because we received it by tradition. And if we’re talking about a law created by earlier generations, then it’s not a supportive derivation; it’s a derivation that once was a creative derivation and we are reconstructing it.
[Rabbi Michael Abraham] That’s “they learned it as a tradition.” Right?
[Speaker B] So why call it a supportive derivation?
[Rabbi Michael Abraham] Because it is not creating the law now. It is not creating the law; rather, I am using it to support an existing law. No—but there are two types of supportive derivations.
[Speaker B] Supportive derivations are only derivations that now support the law, and I don’t know exactly…
[Rabbi Michael Abraham] Not only that—also. There are derivations that are only supporting derivations, and there are derivations that are, in essence, supporting derivations. In Maimonides himself, by the way, he has a unique approach on this issue, because Maimonides argues that a law given to Moses at Sinai for which they found textual support is a Torah-level law. Even though a law given to Moses at Sinai is a rabbinic law, and a law created through a derivation is also a rabbinic law. But if it is a law given to Moses at Sinai for which we found a derivation, it becomes Torah-level. Nachmanides attacks him on this, but never mind. So according to Maimonides there is a point to making this kind of supporting derivation even for a law given to Moses at Sinai. Fine, but I’m saying this is just an example, so we won’t get into it too much. The claim—what I want to claim—is that basically I want to make the same move about this issue. Suppose I have some legal intuition. There’s a rabbi here, okay… mute yourselves, so anyone who isn’t speaking should mute because there are noises here that are disturbing. So we said that sometimes I have a legal intuition, and people tend to think: well, without an explanation it isn’t rational, so obviously there must have been some explanation, and then they start looking for explanations. They don’t find any. So the explanation ends up being a lame explanation—they find some lame explanation. Like with the analogy to a supporting derivation, right? So they find some lame explanation. And the claim is that since I have such an intuition, then even a lame explanation can be accepted. That was probably the explanation. And I say no: if that explanation can’t generate this law, then it also can’t support it—this law, whatever, this halakhah or this… right? Then it also can’t support it. Because if it’s a lame explanation, then it can’t be the explanation for the law, because a lame explanation is not an explanation. The only alternative is to say: there is no explanation. The intuition grasps here what was earlier called a religious value, or I don’t know, some principle that isn’t of the first two kinds, and it tells me that this is the proper or improper way to act, that’s all. And that is sufficient justification. That I will accept, and it is rational even though it has no explanation. By “no explanation” I mean in the first two senses. Yes. Fine, so this is just an introduction to why—because we saw this in Pappino, but I want to see it in our discussion too. So let’s start with the article by Enoch and his group. They begin with their own example of this issue, a similar principle. They say: in some city there are two bus companies operating, blue buses and red buses. And the blue buses are ninety-five percent of the buses in the city. Okay? The red ones are five percent. Now there was a bus accident and someone was injured. Now the question is whether I can sue the blue company, even though I have no witnesses that it was a blue bus—maybe it was a red one? Since ninety-five percent of the buses are blue, I want to sue the blue company so that it compensates the injured party, the person who was hurt. Right? The question is whether I can do such a thing. The answer is no. This is exactly statistical evidence, like the prisoners in jail and so on. We’ll later see that it’s not exactly the same thing, but for now at least it’s like the prisoners in jail—statistical evidence. Statistical evidence is not accepted. Okay? What happens if there is eyewitness testimony that the bus belonged to the blue company? And again, eyewitness testimony also isn’t one hundred percent. It could be that they missed something there and it wasn’t exactly blue but red, because it was far away and they didn’t see exactly, or there were two buses there and they thought the blue one hit him but maybe the red one did, I don’t know. There can be all sorts of possibilities. So let’s say there is some percentage of doubt even in the eyewitness testimony. That testimony they will accept. And it will be possible to sue the blue company so it compensates the injured party. Yes, the same question we asked earlier. Now again, the claim is that everyone agrees that a distinction is made between these two situations. And maybe I want to add one more remark before I—actually, one more remark I forgot earlier. I’m not sure about it, but one more remark. In the context of halakhic law or religious law, it is easier to adopt the third type of reasoning. I can definitely understand people who work in secular law, ordinary law, where they are unwilling to accept the third type of reasoning. Why? Because suppose this stems in some sense—I don’t know which sense, something elusive—for which I have no explanation. A legal system, all that is imposed on it is to bring the truth to light and maintain order. So I am willing to accept either of the first two types of explanation: the statistical explanation, or that the evidence is weaker, or the legal explanation in terms of consequences, like the fruit of the poisonous tree. Anything else is some kind of religious explanation, I don’t know exactly what. Secular law does not tend to accept that sort of explanation. Therefore my claim is mainly when we deal with Jewish law, because in Jewish law there is what Eliyahu said earlier—religious values. And it could be that because of the considerations—let’s call them religious—it is not proper to do such a thing, even though I don’t have explanations of the first two kinds. Why did I say I’m not sure about this? Because something like this could also exist in secular law. If you think that such a thing is not done in some sense—not religious but human, moral, value-based, I don’t know what—then I can also hear a secular legal commentator or legal philosopher adopting this kind of reasoning. But clearly within a religious world it is much easier to adopt these reasonings; it is much more natural. Okay, that’s just the addition I forgot to mention. Now I return to the article by Enoch and his group. They argue that this phenomenon with the buses—yes, this phenomenon can have two kinds of explanations. A priori, even before getting into the topic. An epistemic explanation, as they call it, and a practical-utilitarian explanation. What does that mean? An epistemic explanation is one that depends on our knowledge of reality—yes, epistemology is the theory of knowledge—and from the standpoint of our ability to know reality. It is an explanation tied to knowledge of reality. And a practical-utilitarian explanation is basically what I earlier called a legal explanation. But—and this is an important point—on the face of it, in their article it seems that what they call an epistemic explanation is not parallel to the first kind of explanations I mentioned earlier. The explanations—let’s call them statistical explanations—explanations that the reliability level of this evidence is insufficient and therefore I don’t accept it. That is not what they call an epistemic explanation, but rather something connected to human cognition. In a moment I’ll give that more content. Except that precisely because of this they reject that possibility and say such an explanation cannot be accepted. So let’s now see how this works. But in fact they ignore the third type of reasoning—that is what is important for us here. And they do not see at all the possibility of an explanation of the third kind. I have a legal intuition even though there is no explanation of the first two kinds. So they basically say this. They begin with the epistemic explanation, and they basically argue that the indication I have for whether the explanation is epistemic or legal-utilitarian will be whether we find the same distinction in non-legal contexts, in other contexts. If we find the same distinction in non-legal contexts, then apparently the explanation for this distinction is not rooted in the legal considerations I mentioned earlier. The legal utilitarianism, like the fruit of the poisonous tree, right? The fruit of the poisonous tree is a legal consideration. If we suddenly find in philosophy that philosophy casts doubt on evidence obtained by improper means, then it will be clear to us that the explanation is probably not one of legal utility but something else. All right? So that is only an indication. Now they basically argue as follows. They say—their claim, basically—is that the difference between convicting or obligating the blue bus company to pay when there is eyewitness testimony, and the situation where only statistics work against them, where in that case we exempt them, is because saying that a blue bus hit this person without eyewitness testimony does not count as my having knowledge that a blue bus hit the person. I have statistical probability, but that does not count as knowledge. It is not epistemic, okay? By contrast, when I have eyewitnesses who saw it, I can say that I have knowledge of the matter. That is their claim. Later they reject it, but that is their claim for now. Meaning: legal conviction should be made only on the basis of knowledge—or legal liability should be imposed only on the basis of knowledge. Indications that this is the truth, when they are not based on knowledge, are not enough to impose legal liability on someone. That is basically the claim. And Enoch says: let me show you that this explanation is not a legal-utilitarian explanation, because I will show you that the same distinction itself is made also in philosophical areas unrelated to law. And the example they bring is the following. Think—this is what is called the lottery paradox, that’s what it is called in the literature. Think of a person who bought a lottery ticket, and the chance of winning this lottery is one in a million. Now a drawing was held, the winning number came up, and in reality I did not win. Fine? But I don’t know that. Now they ask me: tell me, did you win the lottery or not? So I can say, look, the chance that I win is one in a million, but I have no information whether I won or not. But I know statistically that the chance I won is one in a million. In such a situation you cannot say that I know I did not win. You can say I assume I didn’t win, I have a high probability that I didn’t win—you cannot say I know I didn’t win. You don’t know. Okay? Not because it is not certain, notice. It is not certain in either of the two possibilities. But I cannot use the verb “know” for a situation where I have no information about the situation except some statistical calculation that tells me what the result will be. Therefore I cannot say that I know that this is the result. By contrast, think of another situation. A person bought a lottery ticket. The chance of winning this lottery is higher, one in a thousand. Now the drawing was held, the winning numbers were published in the newspaper, and the person saw in the newspaper—he read the newspaper—and this person reads the newspaper and sees that he did not win. Okay? Can he say, “I know I didn’t win the lottery”? The answer is yes. He has information, he knows, assuming the newspaper is reliable and so on. Yes, I can say I know that I won the lottery. Now notice: there is, after all, a one-in-a-thousand chance that there was a printing mistake, that the newspaper printed a winning number but not my number because they made a mistake; it could be that it really was my number, only there was a printing error in the paper or an error in the information the newspaper received, a printing error, it doesn’t matter right now, or an error in my reading of the newspaper. There can be errors—let’s say the chance of that error is one in a thousand. So the chance of my winning times the chance of that error, both one in a thousand, one in a thousand times one in a thousand is one in a million. So here too the chance that I am mistaken is one in a million, and nevertheless here I can say I know I didn’t win. Even though there is a one-in-a-million chance that I was mistaken and actually did win. Why is it one in a million? One in a thousand that I won and one in a thousand that the newspaper did not report correctly and really it should have printed my number. By the way, this is not correct probabilistically in light of what we saw in one of the previous lessons. I’m not asking what the chance is that if I won, then the newspaper would write it correctly. I’m asking the reverse conjecture, reverse conditional probability: what is the chance that if the newspaper wrote this, I did win or did not win. That is the opposite question, and we already saw that those probabilities are really not the same at all. They make a statistical mistake there in the article, but never mind for now. The idea is not important—the idea doesn’t matter. There is some probability that the newspaper made a mistake, that much is true; it’s just not necessarily one in a thousand. But in any case for our purposes there is a difference between these two situations. And I think if you asked people on the street, if someone read in the newspaper that he didn’t win, he would certainly say “I know I didn’t win,” and everyone would accept that as a legitimate statement. But if someone said, only because of the statistics I know I didn’t win because it’s one in a million, they would tell him, listen, you can’t say that you know—you assume; maybe there is a reasonable chance that that’s so, but you can’t say that you know. You have no information, you don’t know. You have a calculation that says that. And in the second case it is uncertain information; in the first case it is not information but some inference not through information—not direct information, I mean. Okay? So their claim is that at the philosophical level there is a difference between statistical evidence—because you understand this is exactly the same thing as in our case—because when I have statistical evidence that I didn’t win, I cannot say I know I didn’t win, but if I have information that I didn’t win, even though the chance that this information is wrong is the same chance as in the statistical error, here I can say I know I didn’t win. What do we see? And here this has no legal significance at all; this is not a legal context, no one is convicting anyone and nothing is being done with this. The whole question is only what you can say. The fact is that in language there is a distinction between these two situations. In the first situation it is called “you assume you didn’t win”; in the second situation it is called “you know you didn’t win.” Meaning: if in language we distinguish between these two situations, that means there really is a difference between a decision based on information and a decision based on a statistical calculation. And that difference also appears in contexts that are not legal at all. So that means that apparently this difference, the explanation for this difference, is not rooted in the legal sphere. If it were rooted in some legal calculation like the fruit of the poisonous tree or something like that, then it should exist only in the legal sphere and not in the philosophical sphere or everyday life or whatever you want to call it. That is their claim. And therefore they say that the explanation they propose—that conviction can be based only on knowledge—is an explanation of the epistemic type and not of the legal-utilitarian type. Now notice: epistemic—you already see what they mean when they say epistemic. Epistemic is not parallel to what I called a probabilistic explanation. They are not claiming that the chance of error is higher in this case than in that case. The chance of error is the same in both. My degree of certainty that I didn’t win is one in a million in both situations. So there is no explanation here of the first type. Now they have proved that the explanation is not of the second type, right? That they proved: that it is not a legal-utilitarian explanation, because this distinction appears also in contexts that are not legal. So what is left? What is left is something they call an epistemic explanation. What does that mean? Words like this—that only on the basis of knowledge can one convict. Now what happens in the next stage? Look what happens in the next stage. In the next stage they argue.
[Speaker B] Rabbi, regarding their proof that this isn’t a legal explanation—it’s like, it’s an explanation, right, but also a legal explanation. It’s called a legal explanation because it’s applied in the legal sphere, but it can also be applied in other spheres.
[Rabbi Michael Abraham] No, but the legal consideration won’t appear in the other spheres. After all, if the consideration is a legal one, then it is relevant to the world of law, but what does that have to do with the question of what I read in the newspaper? What—will there be police torture afterward? Will legal problems arise as a result of this? Nothing. The whole question is only how you relate to acquiring knowledge of this kind and acquiring knowledge of that kind, and to this kind of conclusion and that kind of conclusion. Now at the next stage they try to explain the concepts more. This is a very careful and interesting analytical analysis, because Enoch has very strong analytical ability. So they say they want to argue that the concept of knowledge—what we call knowing—the concept of knowledge is connected to what they call sensitivity. What does that mean? Let’s try to understand it this way; let’s look at the lottery example. When I claim that I didn’t win because there is a one-in-a-million probability that I will win, is there a difference? The drawing already took place yesterday; I just don’t yet know the results. I claim that I didn’t win because the probability that I will win is one in a million. Okay? Now what would happen if in the drawing held yesterday I actually did win? Would I now say something different? I don’t have the information. What I am saying now does not depend on the question of what the true information is, right? Whether I won yesterday or did not win yesterday, whatever the results of the drawing were yesterday, today I am supposed to say the same thing, that I didn’t win, because the chance that I won is one in a million—even though that statement could be plainly false, assuming that yesterday I won. The chance is small, but if I did win yesterday, then what I am saying now is of course incorrect. If I did not win yesterday, what I am saying now is correct. But still, from my point of view I will say the same thing no matter what happened yesterday. That means that my statement has no sensitivity to the facts. Sensitivity to the facts means that a change in the facts would change my statement. Here a change in the facts would change nothing about my statement; I would have exactly the same justification for saying that I didn’t win whether in fact I did win yesterday or in fact I did not win yesterday. The facts make no difference regarding my statement and the validity of my statement. So there is no sensitivity. By contrast, they argue, what happens in the second case? I read in the newspaper that I didn’t win, and now I declare: I know I didn’t win. Okay. They say: if yesterday I had won the lottery, suppose yesterday the facts were different—this is called in philosophy a counterfactual, statements contrary to fact. Counterfactuals, okay? Statements—I look at a contrary-to-fact situation; that’s not what happened, but let’s see what would have happened had something else occurred. So we look at counterfactual statements, and reality is like this, but let’s think what would happen if. Okay? So in the first case there is no sensitivity to a change in the facts. Meaning, in the counterfactual situation the result would be the same result. So it doesn’t matter whether reality is like this or the opposite; the result is the same result: I say that I did not win the lottery. It is not sensitive to what happened. By contrast, the second statement, they argue, is sensitive. Why? Because if I had won the lottery yesterday, then almost certainly my number would also have appeared in the newspaper. The chance of an error is small, but my number probably would have appeared in the newspaper. And then when I read the newspaper I would discover that I had won. In that case I certainly would not say that I didn’t win, right? I would say that I won. That means that my statement based on reading the newspaper is sensitive to the real-world fact. A different real-world fact would lead to a different statement. Meaning, the statement of the second type is sensitive to the facts. And their claim is that this is what defines the second track as information or as knowledge. Therefore I can say “I know I didn’t win” in the second case, because information is something that is sensitive to the facts; it is drawn from the facts. You can say that you knew. Because if the facts had been different, then after all you would have thought something different. So that means that the statement you are making now is in fact a statement that expresses knowledge. I know I didn’t win. The statement of the first type has nothing to do with the facts at all. You cannot say that you know you didn’t win. To know is to know facts. All you can say is: I assume I didn’t win because the chance of that is very small. Meaning, this is how they explain the concept of knowledge. What is the difference between drawing some conclusion and knowing that conclusion? The question is whether the method you used was sensitive to the facts or not sensitive to the facts. That is basically the proposal they make. Okay. Now, one has to understand according to their definition, according to the definition of these guys, of Enoch and his group, what would happen if in practice the person did not win and the newspaper mistakenly wrote that he had won? Yes, we said there is some small chance that the newspaper made an error in the information it conveyed—a typing error, an information error, not important. And it said that I won. And I would read the newspaper and say: I won. According to their definition I certainly can say, “I know that I won,” right? Because that statement is sensitive to the facts even though it is not true. Now notice: to say that something is information as opposed to saying that something is a conclusion has nothing to do with whether it is true or false. It only has to do with how I obtained or arrived at that conclusion—whether through some connection to the facts or through some consideration not connected to the facts themselves. Okay. That is just a point to sharpen the meaning of what they are saying. Now their claim is that this is the explanation of the legal distinction, this one, about the blue and red buses. If the one that hit this woman, this injured person, was a blue bus, I would convict the blue company because—because, because I am not on the track of—suppose I adopt a policy that convicts on the basis of statistical evidence. What would happen here is that you are convicting not on the basis of information. Why? Because after all this conviction is not affected at all by the question of what happened. Even if the bus that ran over the person was red, I would convict the blue one. I would convict the blue one because most likely—ninety-five percent of the buses are blue. This is not sensitive to the question of what really happened. The considerations are statistical considerations; they do not touch what really happened, they are not sensitive to what happened. Now, if that is indeed the case, then that means I convicted here on the basis of something that is not information. That we do not do. But if someone saw the bus and says, “It’s a blue bus, the bus that hit the victim—it’s a blue bus.” True, there is some chance that this eyewitness is mistaken, but if the bus were not blue but red, most likely the eyewitness would not say that the blue one ran him over but would say that the red one did. Meaning, the witness’s statement is sensitive to reality. He can make a mistake—again, like with the newspaper. But in principle it is sensitive to reality. If reality were different, he would probably testify differently as well. So that means that I convict here on the basis of information, because I convict here on the basis of a consideration that is indeed sensitive to the facts in reality. That is called convicting on the basis of information. Okay? That is basically the claim. Their conclusion, basically, is that the difference between statistical evidence and direct evidence or circumstantial evidence as opposed to eyewitness testimony or something like that is based on an epistemic consideration. And the epistemic consideration is what they described here: the question whether it is on the basis of knowledge or on the basis of statistical inference. And “epistemic” basically means that it is on the basis of knowledge. And notice, this is not parallel to saying that in one case the evidence is weaker than in the other. No—it is the same evidentiary strength. It is a philosophical difference. The question is whether it is based on knowledge, but the level of certainty is the same level of certainty. Okay? That is their claim, that is the epistemic explanation. Except that in the next chapter of the article they really come to the obvious objection. Why on earth does it matter to us whether it is knowledge or belief? What is this, a decree of Scripture? If I have ninety-five percent and ninety-five percent is enough to convict, then in both cases I have ninety-five percent. Why should I care where I got it from? In truth it is ninety-five percent. It’s not that I have doubt—I do have doubt—it’s ninety-five percent in both cases. Philosophically or linguistically I call this knowledge and that not knowledge. So what? Why should I care what it is called? If I have ninety-five percent certainty and the assumption is that such certainty is legally sufficient to obligate the blue company to pay, then why should I care how this certainty was created? You understand that their claim here is that it cannot be that the explanation is an epistemic explanation because it does not fit into one of the first two categories. It does not fit into the category of reliability, because after all in both cases the probability is one in a million, or five percent in the case of the buses, or in the lotteries one in a million. So statistically there is no difference; the statistical explanation does not exist. A legal explanation, as they said earlier, does not exist here either, and the proof is that the distinction between information and conclusion also exists in non-legal contexts, therefore it is certainly not a legal explanation. So they say: then what is left? Nothing is left. Therefore it is obvious that it simply cannot be right. Okay? That is basically their argument. They made quite a convoluted move. Since it is accepted that there is some explanation, they try to define it all the way through, to give it its due, yes, according to the principle of charity. They go all the way with it, explain that this is knowledge and that is not knowledge, and in the end they say, wait a second, fine, that is the explanation according to those who indeed tie it to the epistemic issue. But it cannot be that an epistemic explanation is relevant at all to such questions. Why should I care about philosophical distinctions between knowledge and not? Do I have ninety-five percent that he is guilty? Let him pay. Is ninety-five percent not enough? Then let him not pay in either case. And what is supposed to determine it is only the percentages. Or, as I said earlier, legal considerations. Fine, that too could be, like the fruit of the poisonous tree. But here there is neither this nor that, so what is left? You already understand where I’m heading. They ignore the third possibility, say it cannot be because for them there is no third possibility. And as a result they now move on and propose a practical explanation. I won’t have time to get into the practical explanation because it would be a shame to begin it in the few minutes that remain. I only want to summarize where we stand for now. Basically it comes out like this: they formulated, I think, in the clearest and sharpest way the explanations of the epistemic type, but they reject them outright. Now, unlike Pappino, who didn’t find such an explanation and also says knowledge and not knowledge—what do I care? Maybe you defined the concept of knowledge, he says, but why should I care whether this is called knowledge or not? What matters is the statistics. And he remains with the matter unresolved. He does not give up on this rule that one does not convict on the basis of statistical evidence, but he remains unresolved. He cannot find an explanation. It is clear to him that there is some explanation—there has to be some explanation—and apparently he is missing it. He does not take into account the third possibility, that there is no explanation: intuition without explanation, which is also a sufficient basis. Enoch goes most of the way with him. They basically end up, after making a greater effort than he did to define the concept of knowledge, still in the end agreeing with him that even after we define it, why should I care? Why should that matter for legal liability, tort liability in this case? Okay, and therefore they too are basically in the same position he is in. Except that afterward they go one step further and propose an explanation of the second type, the legal-utilitarian one. And they also explain why it appears in non-legal contexts even though the explanation is legal. It is a beautiful analytical exercise, and that is what they do. Now here I am already saying in advance: I will argue, after we study that explanation, that this explanation doesn’t hold water; it is lame. And the reason they adopt it is because for them it is a supporting derivation, because there is a clear intuition that this is the right way to act. The only explanation they find, after they reject the epistemic explanation—and there is no statistical explanation—the only explanation they find is basically a utilitarian explanation. And then they say, okay, then apparently that is the explanation, because they too are unwilling to give up their intuition that it is wrong to impose liability based on statistical evidence. So they choose a lame explanation, exactly like the whole introduction I gave earlier. You can simply see it happening right in front of your eyes. And after I show that this explanation is lame, I will argue that lame explanations do not help even for a supporting derivation. If this is not a correct explanation and there has to be an explanation—you do not accept the possibility that there are legal rules without explanation—and this explanation is lame, then throw out the explanation, and then throw the legal rule in the trash. But no one suggests doing that. They prefer a lame explanation. And all this for only one reason: because they are unwilling to accept an explanation of the kind I am talking about—or not an explanation, but a justification without an explanation—which by the way is exactly what I did in the previous lesson. In the previous lesson I proposed explanations of the third type. In this lesson and the next one I am trying to show you why we have to arrive at those explanations, because all the other explanations do not really hold water. So the epistemic explanation—they themselves have now dropped it, and Pappino also does not accept it. The explanation they will propose, which is utilitarian, I will later explain why I do not accept it, and we will go back to the explanations I gave in the previous lesson, which are the explanations of the third type. Not explanations—justifications. There are no explanations, justifications of the third type. Sounds like the title of a movie: Justifications of the Third Type. Fine, that’s it. If there are comments or questions?
[Speaker D] Okay, regarding supporting derivations, could there be another explanation—that even though I have an intuition regarding the law, it could be that earlier generations also decided the law by intuition, and I am only trying to conceptualize it? Meaning, for example, with the tools of the hermeneutical principles that are available to me and were not available to earlier generations, I am trying to conceptualize their thinking?
[Rabbi Michael Abraham] That is my theory about earlier generations too.
[Speaker D] No, for this purpose, say, the Rabbi’s explanation about the conceptualization of the hermeneutical principles across the generations. It could be that an earlier generation said that this was an a fortiori inference somehow.
[Rabbi Michael Abraham] And that is not my explanation, but even according to my explanation the earlier generations also made use of the tools of derivation; they just did not define them. They did not define the rules for themselves. It’s like someone who speaks a language naturally: he does not know the rules of grammar, but he speaks accurately. He knows how to speak. But beyond that I say: it is clear that the Sages did not understand this the way I do. When the Sages spoke about the hermeneutical principles, they understood that it came through tradition. That is at least clear to me. So how did they conceive of their own activity? They really should think that the previous generation also made use of these tools of derivation, and the question is how that can be with such a lame derivation. Fine, that is for another discussion.
[Speaker B] Rabbi, is this article available not in Hebrew? Or only in English?
[Rabbi Michael Abraham] Only in English, at least as far as I know, only in English. It appears—I have a column about it, column 228, and there there is also a link to the article. So it’s in English.
[Speaker B] Okay. And a justification of the—I want to go back to this—a justification of the third type is basically: there is no explanation, but still it has to be upheld.
[Rabbi Michael Abraham] There is no explanation, but—sorry?
[Speaker B] But it still has to be upheld, but still it has to be done.
[Rabbi Michael Abraham] Correct. And there is also logic to doing it, not because… When I say there is no explanation, I mean there is no explanation of the first two types. Like not cutting in line. It’s just not done; you don’t do such a thing. Now, I do not call that irrational behavior even if I haven’t found an explanation for why not to cut in line, because I have an intuition that it is not right to behave that way. That is not called irrational action. Okay? All right. Shabbat shalom. Goodbye. Thank you.