Q&A: That a Person’s Own Admission Should Not Be Greater than the Testimony of Witnesses by an A Fortiori Inference
That a Person’s Own Admission Should Not Be Greater than the Testimony of Witnesses by an A Fortiori Inference
Question
Hello Rabbi,
The Talmud at the beginning of Bava Metzia discusses Rabbi Chiyya’s novelty, according to which witnesses who obligate a defendant for part of the claim obligate him to take an oath regarding the remaining amount. The rationale presented for Rabbi Chiyya’s law is the heading appearing above: “that his own oral admission should not be greater than the testimony of witnesses,” by an a fortiori inference. (One might perhaps wonder whether this is an original part of the baraita or not, but that is not our concern here.)
From there, as is well known, the Talmud moves on to explain why an a fortiori inference is needed to establish Rabbi Chiyya’s law and why it is not enough to rely on a common-denominator derivation. For this purpose, the Talmud brings Rabbah’s famous statement regarding the oath of one who admits part of a claim. After the Talmud explains why the basic reasoning that allows imposing an oath in the case of a person’s own admission does not exist in the case of witnesses’ testimony, and therefore one must use an a fortiori inference, it clarifies what that inference is—and in the end it turns out to be a common-denominator derivation after all.
A. There is a logical difficulty underlying the Talmud’s move. Essentially, what the Talmud is claiming is that one cannot derive it through a common denominator, so we will derive it through an a fortiori inference. That claim is a bit difficult, because if a certain law cannot be learned through a common denominator, then all the more so it should not be possible to learn it through an a fortiori inference, no? After all, the fact that there is no common denominator means that there is some relevant stringency in case A that does not exist in case B. Unless one finds a stringency in case B that is more severe than the stringency in case A—but then perhaps the question would not have arisen in the first place.
B. According to almost all the explanations of the medieval authorities, aside from the Ritva, who seems to have sensed this though I am not clear what he says in the end, and perhaps also the Rashba, who understood the refutation differently from Rashi and Tosafot, it is not clear how the a fortiori inference solves the problem that arose at the earlier stage of the Talmud’s discussion. (This is somewhat connected to the previous question.) Rashi explains that the problem is that in the case of witnesses’ testimony, the lenient aspect—saying not to impose an oath—is that the person is more suspect than one who admits part of the claim. So what feature that exists in witnesses’ testimony solves the problem on that plane? Unless we say that in order to impose an oath one must consider two questions: (1) is it justified to require an oath? (2) will the oath be effective? And in addition say that one always weighs those two factors together, and in the case of witnesses the combined weight of those two factors is no less than in the case of his own admission. But this still requires examination.
C. The second Tosafot on the page, and even more clearly in the Shitah Mekubetzet, say that one does not challenge an a fortiori inference on the basis of reasoning alone. That definitely answers the question I asked before, but it creates another one. What does it mean that one does not challenge an a fortiori inference on the basis of reasoning? That it is a hermeneutic rule that works according to laws regardless of their relevance to what is sought? Is this connected to something I heard in your name, that an a fortiori inference is not a logical rule?
D. Does the baraita, or the Talmud, settle the debt at the end of the passage? In other words, is a paradigm derivation really just an a fortiori inference? Or must one say that the Talmud retreats from its original assumption?
I would be happy if the Rabbi would answer at least some of the questions. In any case, thank you very much in advance.
Answer
Hello,
This is a very tangled passage, and it is hard for me to get into the details here. I will try to comment on your questions.
First I should note that, on the face of it, the basic a fortiori inference itself requires explanation. Simply put, it seems completely unfounded. The stringency in witnesses as opposed to admission is not because of credibility, and the obligation of an oath does not stem from the strength of the credibility regarding the half that he admits.
As for the question whether an a fortiori inference is logical or not, I discussed it in the article “A Good Measure” for Parashat Bereshit, 5765–66, which is on the site. Rabbi Chaim argues that it is, but I have no doubt that he is mistaken. I do not know where you heard from me that this is not a logical rule. I hold the opposite.
As for your questions:
A–B. If I remember correctly, the Talmud’s conclusion is that the a fortiori inference is itself the common denominator, not that one needs an a fortiori inference because there is no common denominator. For Rabbi Chiyya, the law of conspiring witnesses is not a refutation, and therefore he remains with the common denominator. So your two questions here are not clear to me.
Only parenthetically I would add, regarding your basic question itself—which as far as I understand does not relate directly here—that according to at least some views about a paradigm derivation, even a very slight refutation is enough to undermine it (see the passage in Chullin 115, if I remember correctly), whereas to undermine an a fortiori inference you need a real refutation. Therefore, if there is some slight refutation against a common-denominator derivation, there may still be room to derive it through an a fortiori inference. And as is well known, the rule-makers dispute whether a common denominator that is created out of a fortiori arguments, rather than a paradigm derivation, is itself a paradigm derivation or whether it remains an a fortiori inference. I seem to recall Atzmot Yosef at the beginning of Kiddushin discussing this, among others.
C. I thought perhaps this is what the Tosafot you cited means: one does not refute an a fortiori inference with some slight logical consideration, as one would with a paradigm derivation, but only with a substantive refutation. I have not checked it right now, but every refutation of an a fortiori inference is a refutation based on reasoning.
But now I am reminded that, as I recall, Tosafot means something entirely different. In several places throughout the Talmud, Tosafot repeats the principle that there is no refutation of an a fortiori inference based on reasoning—that is, a refutation that points to some conceptual advantage of the source case over the derived case does not undermine the a fortiori inference. Refutations of an a fortiori inference have to be based on law, not on reasoning. And the logic of this is very simple: the a fortiori inference is built on a law that exists in the derived case but not in the source case. We see that despite the conceptual advantage of the source case, the law does not exist there, whereas in the derived case it does. So we see that the derived case is stricter despite the reasoning that gives an advantage to the source case. This is what Tosafot calls “absorbing the refutation.” See, for example, Tosafot s.v. “I will not infer,” Bava Kamma 25a, and many other places.
D. As I wrote to you, at the end of the passage the Talmud understands that the a fortiori inference Rabbi Chiyya intended is itself the common denominator described there. So why is it called an a fortiori inference and not a common denominator? That depends on the dispute I mentioned, from Atzmot Yosef: if the common denominator is also considered an a fortiori inference, there is no problem. According to the views I mentioned, that a common denominator composed out of an a fortiori inference is itself not an a fortiori inference but a paradigm derivation, you would have to explain that Rabbi Chiyya’s a fortiori inference is the one from his own admission, and for the refutations he relies on additional supports—which in the end creates an overall structure of a paradigm derivation.