Daf 3a - The Kal Va-chomer From Hoda'a to Eidim

Wordfile>>

Sources:

1. דף ג. "מאי שלא תהא ... שישנן בהכחשה והזמה".
2. רש"י ד"ה דליכא למימר הכי, חי' הרמב"ן ד"ה אבל עדים.
3. ב"ק דף עה. "איתמר מודה בקנס ... חייב".
4. משנה כריתות יא: עד מזיד הייתי פטור", תוס' ב"מ ד"ה מה לפיו, מה אם ירצה, רמב"ן ד"ה מה לפיו, מה אם ירצה.
5. תוס' ד"ה אשם, רמב"ן ד"ה אלא מה לפיו.

Questions:

1. What does the Gemara mean when it says: "Aval be-eidim de-leka lemeimar hakhi" ("But as regards the testimony of witnesses, where this argument does not apply"), according to Rashi? According to the Ramban? What is the difficulty with Rashi's explanation?

2. In the previous shiur, we cited the position of Mahari ben Lev that the law that a litigant's admission of a debt is binding is based on the idea that a person can accept upon himself a monetary obligation. How can we explain our sugya according to this approach?

3. What is the foundation of the dispute in Bava Kama regarding one who admits liability to a fine, and then witnesses appear?  According to which position can we understand the proof in our sugya that witnesses are stronger than admission?

4. What are the two understandings of the opinion of the Sages in Keritut? According to which opinion is it easier to understand our sugya?

1. What Does the Kal Va-chomer Teach Us?

One who altogether denies a monetary claim against him need not take a shevu'a de-orayta (oath mandated by the Torah), and must take only a shevu'at heset (oath instituted by Chazal).  If, however, he admits to owing part of the sum claimed from him ("modeh be-miktzat"), then the Torah requires him to take an oath that he does not owe the remaining amount.  According to Rabbi Chiya, such an oath is required even if the person denies the claim altogether but witnesses testify to his owing part of the sum.  In the initial stage, the Gemara explains that Rabbi Chiya derived this halakha through a kal va-chomer: if hoda'a (confession) to part of the claim obligates an oath, then certainly testimony to part of the claim, which is stronger than confession, generates a shevu'a de-orayta obligation.  The Gemara acknowledges, however, that despite the kal va-chomer, Rabbi Chiya's halakha nevertheless entails a chiddush, in that Rabba's explanation for the underlying reason behind the oath in a case of modeh be-miktzat does not apply in the situation of witnesses.  The kal va-chomer therefore comes to teach that the shevu'at modeh be-miktzat applies in that case as well.

            The Rishonim debate the meaning of this chiddush.  According to Rashi, we generally assume that somebody suspected of theft is not trusted with an oath, and therefore cannot take any shevu'at ha-dayanim (court-administered oath).  In a case of modeh be-miktzat, however, we do not suspect the defendant of theft, because we may assume he denies the remaining amount only temporarily, until he comes up with the money to repay the full sum.

In Rabbi Chiya's case, however, the defendant is under suspicion of theft.  After all, he denied the claim altogether, and therefore, seemingly, we cannot view his denial as a stalling tactic.  How, then, can we impose an oath upon him, regarding the money he is suspected of stealing?

            But according to this interpretation of the Gemara, it is difficult to understand how we overcome this problem through the kal va-chomer.  The strength of testimony has nothing at all to do with lowering the level of suspicion regarding the one against whom the witnesses testified.  The Ramban indeed pointed out this difficulty:

"I do not understand, what does the kal va-chomer come to teach us?  If you suspect his oath, how does the kal va-chomer make you not suspect his oath?"

The Ramban therefore suggests a different explanation of the Gemara.  Rashi's view, however, requires elucidation.  Perhaps we can explain on the basis of Tosefot's approach (s.v. aval).  Tosefot claim that, indeed, because of the fact that we cannot trust the defendant with an oath when witnesses testify that he owes some of the money he denied, we cannot derive Rabbi Chiya's halakha through a kal va-chomer.  However, the Gemara prefers to challenge the kal va-chomer itself by identifying possible "stringent" aspects within the "lenient" halakha on which the kal va-chomer is based.  Tosefot write, "This is the Gemara's method, to raise a refutation against [the basic logic underlying] a kal va-chomer, rather than questioning it due to other reasons.  It therefore does not bring the argument of 'he is stalling'."  Clearly, however, this is a very stretched reading of the Gemara.

            In order to explain Rashi's view, we must first discuss the concept of kal va-chomer.  Kal va-chomer is listed among the thirteen exegetical techniques used to extrapolate halakhot from the text of the Torah ("middot she-ha-Torah nidreshet ba-hen").  Therefore, a kal va-chomer operates not merely as a "sevara," a logical deduction, but also as a "derasha" – an exegetical inference on the basis of which Chazal determine Torah law.  Rabbi Tarfon and the Chakhamim debate the extent of a halakha derived through a kal va-chomer (Bava Kama 25a).  Generally, there applies the rule, "dayo la-ba min ha-din lihyot ke-nidon" – the law derived through a kal va-chomer cannot extend further than the original law, from which this law is derived.  However, according to Rabbi Tarfon, we do not apply the "dayo" provision if its application will undermine the entire kal va-chomer.  Even Rabbi Tarfon, however, upholds the rule of "dayo" when the kal va-chomer will not be undermined as a result.  The reason would seem to be that since, as mentioned, kal va-chomer works as an exegetical tool, rather than a process of logical deduction, it must, according to Rabbi Tarfon, be applied in all cases.  "Dayo" does, indeed, limit the scope of what we derive through a kal va-chomer, but it does so only if we can still sustain the kal va-chomer.  If "dayo" will effectively eliminate the kal va-chomer, then we do not apply "dayo" in such a case.  Clearly, then, Rabbi Tarfon views kal va-chomer as a "derasha" and not merely a "sevara."  Perhaps the Chakhamim agree with Rabbi Tarfon in principle, only they maintain that "dayo" sets a limit on the "derasha" itself, even if application results in nullification of the "derasha".

            Thus, it turns out that a kal va-chomer can be applied in two different contexts.  On the one hand, it is a "sevara," a line of reasoning that can be used to expand a given law - after its basic halakhot have already been determined – to other areas.  On the other hand, a kal va-chomer works as a "derasha," used on the primary level as a method of establishing the laws themselves.  In our sugya, Rabbi Chiya appears to make use of kal va-chomer on this primary plane, as a means of determining the parameters of shevu'at modeh be-miktzat.  Only after we implement the "derashot" and determine the rules of the given law, are we able to consider the rationale upon which the law is based.  Perhaps, according to Rashi, the kal va-chomer does not teach us directly that someone under suspicion with regard to money is not suspect with respect to an oath.  Rather, the kal va-chomer works as a "derasha," through which we establish that if witnesses testify that the defendant owes half the sum, he must take a shevu'at modeh be-miktzat on the remainder.  Without the kal va-chomer, we would have perhaps concluded that since this individual is now suspect with regard to money, we cannot trust him with an oath.  But once we derive Rabbi Chiya's halakha through a kal va-chomer, establishing the obligation of an oath when witnesses testify to the defendant's owing part of the amount, we have then determined that a shevu'a can be cast even upon someone who totally denied the loan, because even such a person may only be stalling (see Rashi 5a s.v. veha gazlan hu).

            It emerges from this discussion that one may, potentially, understand the kal va-chomer in our sugya in one of two ways.  According to one approach, the kal va-chomer here works as a "sevara," with which we try to expand the law of modeh be-miktzat to the additional case of witnesses.  To accomplish this, we must claim that witnesses' testimony is more trusted than a defendant's confession.  Logic then dictates that we demand an oath in situations where witnesses testify to the defendant's owing part of the sum.  According to the second approach (which we enlisted to explain Rashi's view), this sugya works on the primary level, trying to define the law requiring an oath for part of an amount claimed.  We use the kal va-chomer to show that the words "ki hu zeh" (Shemot 22:8 – the phrase from which we derive shevu'at modeh be-miktzat) introduce the obligation of an oath not only in a situation of modeh be-miktzat, but in cases of testimony on part of the claim as well.  Perhaps, according to this kal va-chomer, it suffices for us to know that witnesses generally carry more legal weight than a defendant's confession, not necessarily that they are BELIEVED more than a defendant's confession.

2. Hoda'at Ba'al Din – the Defendant's Confession

In any event, the sugya proceeds to clarify this kal va-chomer from confession to testimony, questioning why the ability of the former to generate an oath necessarily dictates that the latter does so, as well:

"What is the kal va-chomer?  If his confession, which cannot yield an obligation of payment, yields an obligation of an oath, then it certainly follows that witnesses, who can yield an obligation of payment, yield an obligation of an oath.  His confession cannot yield an obligation of payment?  Is not a defendant's confession equivalent to the testimony of one hundred witnesses?"

At first glance, this passage is very difficult to understand.  How could the Gemara possibly think that a litigant's confession does not obligate him to pay?  Even small schoolchildren know that "hoda'at ba'al din ke-mei'a eidim" – a defendant's confession is legally equivalent to the testimony of one hundred witnesses.  Indeed, Rashi suggests that we should not seriously consider the Gemara's original suggestion.

            By contrast, we might propose a different explanation, which we will introduce by analyzing the concept of hoda'at ba'al din.  The previous shiur dealt with the halakha equating a litigant's confession with testimony, and noted the debate between the Ketzot and the Mahari Ben Lev (see Ketzot, 34:4).  According to the Mahari Ben Lev, hoda'at ba'al din operates on the basis of "hitchayvut" – the ability of a person to take upon himself a given obligation.  The Ketzot, however, disagrees, and claims that we believe a litigant with respect to his confession more than we believe witnesses.  Now, if the halakha of hoda'at ba'al din works on the basis of self-obligation, then we have no indication regarding the believability of confession and we are compelled to explain that our sugya compares hoda'a and testimony in a general sense, claiming that the strength of witnesses generally exceeds that of hoda'a.  We then understand the Gemara's refutation, that confession yields a payment obligation just as witnesses do, and the basis of this obligation – either our acceptance of the witnesses' testimony, or the litigant's taking upon himself an obligation to pay – is inconsequential.  If, however, the kal va-chomer in our sugya deals strictly with the issue of trust, then it seems difficult to accept the Mahari Ben Lev's opinion, that hoda'at ba'al din evolves from the litigant's self-obligation, for what does his self-obligation have to do with the question of trust?

            According to the approach that the kal va-chomer deals with the relative levels of trust afforded to witnesses and the defendant's confession, we might explain that the stages in our sugya discuss precisely this very point.  Initially, the Gemara thought that a defendant's confession is binding only by virtue of his having taken upon himself an obligation to pay.  Understandably, then, the Gemara determined that "his confession cannot yield an obligation of payment" – meaning, his confession is not believed as testimony is.  In the end, however, the Gemara retracts this argument and views hoda'at ba'al din as based upon our trust in the litigant's confession, and we therefore have no basis – at this point – to claim that witnesses are believed more than the defendant.

3. "Kenas"

Thereafter, the Gemara tries basing this kal va-chomer on kenas (fines, which can be cast upon an individual only through testimony): "What do we mean by 'money'?  Kenas.  If one's confession, which cannot yield an obligation of kenas, yields an obligation of an oath, then it certainly follows that witnesses, who can yield an obligation of kenas, can yield an obligation of an oath."  One's confession cannot yield a kenas obligation because of the principle "modeh bi-knas patur" – one who confesses to a crime normally punishable with a kenas is exempt from the kenas payment.  One question, however, remains.  Does the fact that one's confession cannot yield a kenas obligation reflect a lack of trust in the confession, as opposed to testimony, or do we, perhaps, fully believe the confession, and the Gemara means simply that witnesses generally possess more power than a litigant?

            This question requires us to analyze the halakha exempting one from kenas payment when he confesses.  The Gemara in Masekhet Bava Kama (75a) brings two views as to whether one must pay kenas if witnesses testify to his crime after the confession.  These two positions appear to reflect two different understandings of the modeh bi-kenas exemption.  The view that the exemption continues even after the arrival of witnesses presumably believes that the confession not only is insufficient to generate a kenas obligation, but actively exempts the perpetrator from kenas.  Meaning, kenas is fundamentally a punishment; on the strictly legal plane, this individual does not owe this money to the litigant.  The Torah cast this fine only upon a sinner; one who confesses and repents is exempt from paying kenas, even if witnesses come thereafter.  The other view, which allows for the introduction of a kenas obligation if witnesses come subsequent to the confession, maintains that confession cannot generate a kenas obligation, but does not provide an exemption from the kenas payment.  According to this view, since a person is not believed to incriminate himself, he is not liable for kenas payment through his own confession; only testimony of witnesses can yield a kenas obligation.

            It stands to reason that if one's confession actively exempts him from kenas, this does not result from a lack of trust in his confession, but rather stems from the fact that one who confesses is not considered a "rasha" deserving of kenas.  If, however, hoda'a does not exempt the perpetrator but merely lacks the wherewithal to generate a kenas obligation, which can be generated only through witnesses, then we may indeed prove from kenas that witnesses are trusted more so than a litigant's confession.

            It thus turns out that if we explain that hoda'a, as opposed to testimony, is not believed sufficiently to generate a kenas obligation, we may then understand the kal va-chomer as proving that witnesses are trusted more than the defendant's confession.  If, however, hoda'a actively exempts one from the kenas payment, then we must explain our sugya as dealing not with the issue of trust, but rather with the general relative strengths of hoda'a and testimony.

4. Korban Chatat

            The Gemara then tries refuting the kal va-chomer on the basis of korban chatat (the sin offering brought for an inadvertent violation): "One's confession is different, in that it can obligate him in a sacrifice – will you apply this to witnesses, who cannot obligate him in a sacrifice?" The ensuing discussion in the Gemara indicates that this refutation assumes the position of the Chakhamim in Masekhet Keritut (11b), who, in opposition to Rabbi Meir, exempt one from a korban chatat if two witnesses testify to his having mistakenly committed a violation and he denies having done so.  The Gemara in Masekhet Keritut (12a) brings two approaches to understanding the Chakhamim's view:

"The question was raised: What is the Chakhamim's reasoning?  Because a person is trusted with regard to himself more than one hundred people, or perhaps because we say, since if he had said [that he committed the sin] intentionally, he would have been exempt [from a korban], therefore, if he says, 'I did not eat,' he is likewise believed and exempt?"

The commentaries note that our sugya works off the assumption that according to the Chakhamim, a person is believed with respect to himself more than one hundred witnesses.

            We should point out, however, that were it only for this halakha, we can prove only that there exists one case where the strength of hoda'a exceeds that of witnesses.  After all, Rashi explains that for purposes of a korban chatat, we do not take into account witnesses' testimony because of the clause used in the verse introducing this obligation: "or if it is known to him [that he inadvertently committed a severe violation]."  This implies that he must realize his transgression independently, and not through the testimony of others.  It would seem, therefore, that in terms of objective trustworthiness, we fully believe the witnesses.  But for purposes of korban chatat, the knowledge itself that so-and-so unintentionally ate forbidden fat, for example, does not suffice.  In order for one to become obligated in a korban chatat, the sinner must himself become aware of his violation, and in this regard, the subjective knowledge of the alleged sinner carries more weight than the testimony of one hundred witnesses, for despite the testimony, the person is convinced that he never sinned.  Therefore, we exempt him from a korban.  The Meiri writes:

"That which we wrote regarding witnesses who said, 'You ate forbidden fat' and he denies it, that he does not bring a sacrifice – some apply this even when he said, 'I do not know,' since it is written, 'He confesses' – [implying that he becomes obligated] specifically through his verbal confession.  Nevertheless, if he remains silent, he brings a sacrifice, for silence is tantamount to confession."

Clearly, if the individual did not deny the witnesses' account, and he says that he does not know if he committed the sin, his exemption does not result from our trusting him more than we trust the witnesses.  Rather, we exempt him due to the absence of his personal knowledge of the sin's occurrence.  Necessarily, then, proving the fact that the sin occurred does not suffice to generate an obligation to bring a korban chatat; we require the knowledge (or confession) of the sinner.

            In order to prove that the individual is trusted more than the witnesses with respect to the korban chatat obligation, we must examine the opposite situation, where the individual claims that he transgressed, while witnesses testify to the contrary.  Indeed, Tosefot (s.v. ma le-fiv) conclude on the basis of our sugya that even in such a case we believe the individual over the witnesses.  Commenting on our sugya, Tosefot write, "Meaning, for it [one's confession] can obligate him in a sacrifice even if the witnesses oppose him and say, 'It was not forbidden fat, but rather permissible fat'."  It appears from Tosefot that, in fact, the individual is believed over the witnesses with respect to korban chatat, for we cannot allow bringing a chatat merely on the basis of personal conviction, without any backing in objective fact.  After all, the actual occurrence of the sin is what generates the obligation to bring a chatat; one who offers a chatat without having sinned violates the prohibition against bringing chulin (a non-sacrificial animal) into the Temple grounds.  Tosefot's view, however, requires explanation; after all, "on the basis of two witnesses shall a matter be established," and throughout Halakha we rely absolutely on the testimony of witnesses.

            The Ramban indeed questioned this halakha, and therefore had difficulty with our sugya: "I find this very difficult: do not witnesses, too, obligate him in a sacrifice if he does not oppose them?  And his confession obligates him only when there are no witnesses opposing him; when they oppose him, it does not appear [that he would be obligated to bring a korban]."  At first glance, one might claim that the Ramban argues with Tosefot, for throughout Halakha we place full trust in the testimony of witnesses.  However, the Ramban proceeds to give the following explanation to resolve his difficulty: "This is what it says: We find that witnesses cannot obligate him in a sacrifice, such as when he says, 'I did not eat,' whereas his mouth obligates him through confession, as well as through silence.  But we say this because here his mouth is stronger than witnesses."  Meaning, when witnesses testify that he committed the violation, if he accepts their testimony by remaining silent and must therefore bring a sacrifice, his obligation is due not to the testimony, but rather to his confession, insofar as silence is considered an implicit confession.  The Ramban here implies that even he agrees that establishing the occurrence of the violation to require a korban, demands the sinner's consent, and witness' testimony does not suffice.  He is merely uncertain as to whether we can rely on his confession if it contradicts the witness' report.

            Later, the Ramban explains the position of the Chakhamim:

"The reason for this is that since this matter concerns himself, he is believed, for if he did not want to bring a korban, we do not seize it from him.  Therefore he is believed with regard to himself more than one hundred people, who have nothing to do with the person.  Rashi z"l explains here very well that this is because it is written, 'or if his sin is known to him' – and not that others inform him."

Meaning, Bet-Din relies on witnesses only in the legal realm, which demands objective standards by which it reaches its judicial decisions.  In the legal framework, therefore, Halakha determines that we can take witnesses' testimony – after the process of interrogation – as absolute proof.  Korban chatat, however, is not under Bet-Din's authority, but is rather the responsibility of the sinner himself.  Therefore, he can rely on his personal knowledge to determine the objective facts, perhaps even when it runs contrary to the testimony of witnesses.  It turns out, then, that according to the Chakhamim, there is a realm within which the "litigant" is trusted over the testimony of witnesses.

5. Asham

The Gemara dismisses this refutation, claiming that Rabbi Chiya follows the view of Rabbi Meir, that witnesses are believed even with respect to korban chatat: "Rabbi Meir said: If two can bring a person to capital punishment, which is severe, then should they not bring him to [the obligation to bring] a korban, which is more lenient?"  But the Gemara then attempts a different refutation, from korban asham (guilt offering): "His confession is different, for it can obligate him in an asham."  The Gemara then responds, "Asham is the same as korban [chatat]."  Meaning, just as Rabbi Meir allows for an obligation of korban chatat on the basis of two witnesses, so does he allow for an obligation of korban asham based on their testimony.  At first glance, this response is obvious: why should we distinguish between a chatat and asham?  The question then arises, what does this Gemara come to teach us?

            In truth, however, within this brief discussion lies a very big chiddush.  Tosefot (s.v. asham haynu korban) note that according to the Gemara in Bava Kama, Rabbi Meir agrees to the Chakhamim that we cannot require one to bring a korban asham on the basis of two witnesses.  Tosefot write, "The sugya there, which holds that according to Rabbi Meir, witnesses cannot obligate one in asham, [says this] because regarding asham it is written, 've-hitvada' ['he shall confess']."  Meaning, there exists a basic difference between a korban chatat and korban asham.  What generates the obligation to bring a korban chatat is the transgression, and Rabbi Meir and the Chakhamim argue only with regard to the means by which we establish the occurrence of this act, if we do so through testimony, or if the individual's knowledge takes precedence.  But what generates the obligation of korban asham is not the occurrence of the sin per se'; the individual bears no obligation to bring an asham so long as he has not confessed.  And although Rabbi Meir maintains that witnesses can provide the necessary information to obligate a korban chatat, they undoubtedly cannot take the place of confession, which is required to generate the obligation of korban asham.  Therefore, until the individual comes forward and confesses, we cannot obligate him to bring an asham.  This difficulty was raised by the Ramban, who writes:

"We find it difficult, for even according to the Rabbanan, two witnesses obligate him to bring a korban [chatat] if he does not oppose them, and yet, they do not obligate him to bring an asham, for regarding an asham it says, 've-hitvada,' and the confession is what obligates him.  How, then, can we obligate him without confession, even if Eliyahu would come and tell us?  This does not resemble at all the case of one who eats forbidden fat and the like, for at the moment of eating one becomes obligated [to bring an offering], and the witnesses serve merely to reveal the facts.  But regarding an asham – one has no obligation until the moment of his confession!"

            In explaining this distinction, I heard from HaRav Aharon Lichtenstein zt"l that what generates the obligation of chatat is the transgression itself, whereas what generates the obligation of asham is the personal status of "chotei" (sinner).  Hence, so long as he does not confess and acknowledge that status, he is not obligated to offer a korban asham.  In light of this distinction, it becomes very difficult to understand the attempted refutation based on asham.  For if, indeed, we distinguish in this regard between chatat and asham, then this difference does not stem from the fact that regarding asham we believe the individual over the witnesses.  Rather, we need the individual's confession as a basic requirement which generates the obligation of asham.  We can explain the Gemara's refutation only if we deal here with the respective strengths of the litigant and witnesses in a general sense.  If, however, we deal with the issue of trust, what does confession regarding an asham have to with trust?  Or, perhaps, one might suggest that our sugya does not distinguish between chatat and asham, as indeed it appears from the Gemara's response – "asham is the same as korban."

Summary

            We dealt here with the Gemara's discussion concerning the possibility of deriving the halakha requiring a shevua when witnesses testify regarding part of the claim, based on a kal va-chomer from the law of modeh be-miktzat.  We proceeded through the different stages in the Gemara, and we came up with two different understandings of this kal va-chomer.  According to one approach, the kal va-chomer works as a "sevara," a logical deduction that expands the law of modeh be-miktzat to a situation of witnesses, by virtue of the fact that witnesses are believed more so than a litigant's confession.  According to the second approach, this kal va-chomer is an exegetical inference, predicated on the fact that generally speaking, testimony possesses greater strength than hoda'a.  In conclusion, the Gemara determines that we cannot deduce this halakha based on this kal va-chomer, because of a different argument: "One's mouth is different, because it is not subject to hakchasha [opposition] or hazama [accusation of falsehood]; can you apply it to witnesses, which are subject to hakchasha and hazama?"  In the end, Rabbi Chiya derives his halakha from a "ba-meh ha-tzad," with which we will deal in our shiur next week.

Sources for the next shiur:

Sources:

1. גמרא ג: "אלא אתיא מעד אחד ... תורת הזמה לא פריך" (ד.). רש"י וריטב"א.
2. רמב"ן ד"ה 'הכא גרסינן', רשב"א ד"ה 'הכי גריס רש"י'. 
3. תוספות ד"ה 'הצד השוה', תוספות רבינו פרץ ד"ה 'שעל ידי טענה וכפירה וכו''.
4. שבועות מ. "אמר רב נחמן בר יצחק אמר שמואל ... מחייבו שבועה", חידושי הרמב"ן על אתר "איכא דדייקי...".

Questions

1. Pay attention to the various stages in the attempt to understand how Rav Chiyya's derived his halakha. On what basis does the Gemara reject the possibilities of learning the din from single witness and gilgul shevu'a?

2. The Gemara concludes that Rav Chiyya bases his halakha on the common element found in two sources tzad hashaveh. The Spanish Rishonim had a different reading than Rashi regarding the final position that the Gemara proposes for Rav Chiyya. What are these two sources according to Rashi and according to the Spanish Rishonim? How does this difference affect the final understanding of the derasha?

3. What may be learned from the Tosafot regarding the factor that obligates an oath? Does the Ramban agree?

4. The Ramban cites a disagreement between the Rif (Shevu'ot 28a in the Rif's pagination) and Rabbenu Efrayim (and Ri Migash), on the issue whether the testimony of a single witness obligates an oath when the plaintiff is himself is uncertain that anything is owed to him. How is this disagreement reflected in the various understandings regarding the criteria for taking an oath? 

Translated by David Silverberg