Are there fictitious baraitot in the Babylonian Talmud?
Originally published in HUCA 42 (1971): 185-96.
Republ. in L. Jacobs. Rabbinic thought in the Talmud. London; Portland, OR: Vallentine Mitchell, 2005: 42-54.
Every student of the Babylonian Talmud is aware that not infrequently in this work a ruling is given by an amora and the very same ruling is then quoted from a baraita in identical or near-identical words. It will be sufficient for the purpose of stating the problem to give just one or two examples.
1. Berakhot 12a
R. Judah said in the name of Samuel: ‘They wished to read so [the Ten Commandments daily before the reading of the Shema] also outside the Temple but they [the Sages] had already prevented this because of the insinuations of the sectarians.’ We have also been taught so [i.e. in a baraita]: ‘R. Nathan said: “They wished to read so outside the Temple but they had already prevented this because of the insinuations of the sectarians”.’
2. Shabat 58a
Samuel said: ‘A slave may go out [into the public domain on the Sabbath] with a seal around his neck but not with a seal on his garments.’ We have also been taught so: ‘A slave may go out with a seal around his neck but not with a seal on his garments.’
3. Berakhot 40a
Raba b. Samuel said in the name of R. Hiyya: ‘Eat salt after every meal and drink water after every drink and you will never come to harm.’ We have also been taught so: ‘Eat salt after every meal and drink water after every drink and you will never come to harm.’
4. Ketubot 22a
Our Rabbis taught: ‘If a woman said: “I am a married woman” but later declared: “I am single”, she is believed.’ But did she not make herself into something forbidden? Raba son of R. Huna said, ‘the case is where she gave a reasonable explanation [amatla] to her [former] words.’ We have also been taught so: ‘If she said: “I am a married woman” but later declared: “I am single”, she is not believed. But if she gave a reasonable explanation [amatla] to her [former] words she is believed.’ And it once happened to a certain great lady, who was very beautiful, that when many men rushed to propose to her she said to them: ‘I am betrothed’. After a time she arose and married herself off. The Sages said to her: ‘Why did you do this?’ She replied: ‘At first, when unworthy men proposed to me, I said that I was betrothed; but now that I have been proposed to by a worthy man I married myself off.’ The law was brought before the Sages at Usha by R. Aha, Prince of the Palace, and they said: ‘If she gave a reasonable explanation to her words she is believed.’
How is this phenomenon to be explained? If there were only a few such instances it might have been possible to attribute it all to coincidence—the amora and the baraita arrived independently at the same ruling, though it would be stretching coincidence too far even here to explain the identical wording in this way. In fact, there are so many instances that coincidence can be ruled out. Moreover, although one finds proofs of the correctness of amoraic views from baraitot in the Jerusalem Talmud too, one never finds the remarkable correspondence in style and language there that one finds in the Babylonian Talmud.
In order to deal with the problem I. H. Weiss advances the startling thesis that suspicion is attached to many of the baraitot in the Babylonian Talmud and that many of them are fictitious, i. e. they are not authentic transmissions of tannaitic opinion but were invented by the Babylonian amora’im as alleged support for their views.  According to Weiss, the only alternatives are that either the amora’im invented baraitot to accord with their own views or else they had access to authentic baraitot but palmed them off as their own sayings! In any event, according to Weiss, something peculiar was going on that succeeds in putting a great question mark against the authenticity of many baraitot in the Babylonian Talmud. Weiss writes:
‘It is extremely hard to believe that both the amora and the baraita happened to say the same things in the same style and language word for word by coincidence. Even if the possibility of this kind of coincidence could be accepted in two or three places it is impossible to believe that such agreement should be found in hundreds of instances, as we do indeed find it in the Babylonian Talmud. And it is also no coincidence that baraitot of this kind are nowhere found in the Jerusalem Talmud. It is true that also in the Jerusalem Talmud we find support adduced for amoraic views from baraitot . . . but complete agreement in style and language between the words of the amora and the baraita, as we find in the Babylonian Talmud, is never found there anywhere at all. All this justifies doubts regarding the claim of authenticity and early date of these baraitot in the Babylonian Talmud.’
Abraham Weiss takes strong issue with I. H. Weiss.  The earlier Weiss’s work, Dor Dor Vedorshav, he remarks, has become virtually a popular work and since it accuses the talmudic sages of ‘forgery and plagiarism’ it needs to be vigorously refuted. After a comprehensive survey of the whole field, Abraham Weiss arrives at a very different solution from that of I. H. Weiss. According to Abraham Weiss the solution to our problem is to be found in the fact that the Babylonian amora’im did all their work in a literary atmosphere in which numerous sayings from earlier times had become the common property of the later scholars. It is not surprising, therefore, that a particular saying might be attributed both to a baraita and to an amora. The amora either quoted consciously from the earlier source without actually mentioning that it was a quote, or he stated that it was a quote but this statement was later omitted. The reason was that the saying had by then become attributed to him rather than to his original source.
Undoubtedly Abraham Weiss is closer to the truth than I. H. Weiss. Abraham Weiss’s account is adequate to explain the phenomenon; the process to which he calls attention can be observed generally in world literature. Every dictionary of quotations, for example, knows of famous sayings attributed to more than one author. For all that, Abraham Weiss’s explanation is too simplistic. The purpose of this essay is to demonstrate that even if Abraham Weiss is correct in his basic contention, this does not rule out the possibility that the redactors of the Babylonian Talmud, at least, occasionally reworked a baraita so as to fit it into their schematic arrangement of the sugya. There appears to be sufficient evidence that authentic tannaitic sayings were consciously given a fictitious wording in the form of a baraita for the purpose of providing a contrived argument which leads by stages towards its climax. 
We begin with the sugya in tractate Rosh Hashanah 24a-b (paralleled by the sugya in tractate Avodah Zarah 43a-b). The sugya begins with the report in the Mishnah that Rabban Gamaliel had diagrams of the moon’s phases on a tablet on the wall of his upper chamber for the purpose of demonstrating to the unlearned the various phases of the moon. This shows that there is no objection to making representations of the heavenly bodies, but the Gemara asks: ‘Is it not written: “Ye shall not make with Me” (Exod. 20: 20), which means: “Ye shall not make the likeness of My attendants?”’ To this Abaye replies that the Torah forbids only those attendants of which it is possible to make copies, i. e. the buildings and vessels of the Temple but not the making of a likeness of the moon.
A baraita is quoted in support in which it is stated that it is forbidden to make copies of the Temple or its vessels. The Gemara then objects, ‘but is it permitted to make the likeness of attendants of which it is not possible to make copies? Have we not been taught in a Baraita: “Ye shall not make with me”—Ye shall not make the likeness of My attendants which minister before Me on high?’ To this Abaye replies that the Torah forbids only the making together of the four faces in the heavenly chariot, i. e. the Gemara now goes back to the original reply that the verse refers to copying the Temple and its vessels. 
There follows a short digression in the Gemara which we omit as irrelevant to our particular analysis. The Gemara then objects, ‘but is it permitted to make the likeness of other attendants? Have we not been taught in a baraita: “Ye shall not make with Me”—Ye shall not make the likeness of My attendants which minister before Me on high, for example, the Ofanim, the Serafim, the Holy Beasts and the ministering angels?’ To this the answer is given that the Torah forbids only the making of the likeness of attendants in the upper sphere but not those in the lower sphere (such as the moon). There follows a digression in which an attempt is made to distinguish between making the likeness of the moon for purposes of worship and for other purposes. The Gemara then asks, ‘but is it permitted simply to make [i. e. not for the purpose of worship] the likeness of the moon? Have we not been taught in a Baraita: “Ye shall not make with Me”—Ye shall not make the likeness of My attendants which minister before Me, for example, the sun, the moon, the stars and the planets?’  To this the answer is given that the prohibition is only against making them, not against having them, and Rabban Gamaliel had them made for him by Gentiles.
A casual glance at this sugya shows that four successive interpretations of ‘Ye shall not make with Me’ are quoted, the last three in the name of a baraita. These are:
1) Ye shall not make the likeness of My attendants.
2) Ye shall not make the likeness of My attendants which minister before Me on high.
3) Ye shall not make the likeness of My attendants which minister before Me on high, for example, the Ofanim, the Seraphim, the Holy Beasts and the ministering angels.
4) Ye shall not make the likeness of My attendants which minister before Me, for example, the sun, the moon, the stars and the planets. 
Now it is surely stretching our definition of coincidence too far to suggest that there were four baraitot (or three, if the first quotation is not a baraita ), each of which conveys only part of the full meaning of the text and this so fortuitously that the baraitot could have been used in the sugya so as to work up to a climax the argument that is presented. The only convincing explanation is that originally there was only one baraita, which was divided up by the redactors (or elaborated on by them) in order to serve as bricks in the erection of their literary edifice.
The most probable reading of the original baraita was: ‘Ye shall not make the likeness of My attendants which minister before Me on high.’  In stage 1 only the first part of this original baraita is quoted (leaving out bamarom, ‘on high’); in stage 2 the full baraita; in stage 3, the addition of the gloss beginning ‘for example’; and, finally, in stage 4 the different gloss which serves to clinch the argument. At each stage a little more is revealed until the full force of the objection is conveyed. 
Further evidence that the whole truth does not lie with Abraham Weiss’s solution to our problem is provided by the following considerations. Very frequently in the Babylonian Talmud a mishnah is emended in order to avoid a difficulty presented by the unemended text. The standard formula for this is: hesore mehasra vehakhi katana—‘the mishnah is incomplete and should read as follows.’  Now, historically considered, it is clear that this kind of emendation is artificial and is a device adopted by the amora’im. If, therefore, the Gemara quotes a baraita which accords in every detail with this kind of amoraic emendation this must be an anachronism and the inescapable conclusion is that the baraita quoted is fictitious, at least in the form quoted. In tractate Bava Kama 16a, for instance, Ravina makes such an emendation and this is supported by a baraita which uses the very same words as Ravina! It is impossible to explain this phenomenon on the lines of Abraham Weiss’s solution, i. e. that in a literary atmosphere sayings can easily be attributed to teachers of different generations. This can only be applied to sayings which could have originated in the tannaitic period but it obviously cannot apply to the use of a device unknown until the amoraic period. 
In this sugya there is a debate between R. Huna and R. Yohanan whether the benediction over bread should be recited over a whole loaf or over a broken piece (i. e. where the broken piece has some other advantage such as size or quality ). R. Nahman b. Isaac said that the God-fearing man satisfies both requirements. The ‘God-fearing man’ is said to be Mar son of Rabina , who used to place the broken piece underneath the whole loaf and break bread over both of them.
We are then told that a tana taught in the presence of R. Nahman b. Isaac: ‘He should place the broken piece underneath the whole loaf and break bread and recite the benediction over both of them.’ R. Nahman b. Isaac then asks the tana what his name is and he replies: ‘Salmon’. Whereupon R. Nahman says: ‘Thy name is peace and thy teaching is perfect for thou hast made peace among the disciples.’ Now if the baraita quoted is an authentic tannaitic statement we would have the astonishing coincidence that the tannaitic procedure should have anticipated the practice which was only adopted generations later in order to satisfy the requirements of both amora’im, R. Huna and R. Yohanan. Even if it be argued that already in tannaitic times the matter was debated and the compromise adopted (although this is extremely unlikely) it is surely stretching our definition of coincidence too far to say that after the compromise had been suggested by R. Nahman b. Isaac the tana Salmon should fortuitously have produced a baraita with exactly the same compromise. The most probable explanation is that the baraita is fictitious and was invented by Salmon. It is not impossible that the meaning of R. Nahman’s pun on the tana’s name is itself a hint that the master had seen through his pupil’s subterfuge but approved of it nonetheless!
Of course, the type of baraita we are considering here—quoted by a tana of the amoraic school—is in a case of its own. Baraitot of this kind are frequently quoted in the Babylonian Talmud itself as inauthentic.  But here R. Nahman b. Isaac does appear to accept the baraita and does not resort to the usual formula of rejection. Hence, if we are correct, we have here a fictitious baraita given nonetheless the semblance of authenticity in order ‘to make peace among the disciples’. 
‘R. Huna said: “A man with running eyes [zebalgan] should not lift up his hands [to deliver the priestly benediction].’” The Gemara asks: ‘But was there not such a one in the neighbourhood of R. Huna who used to lift up his hands?’ The Gemara replies: ‘That one was familiar to the townspeople [dash be’ira],’ i. e. the objection is because the people might be distracted by his unusual appearance but this would not apply to one with whom the townspeople were familiar.
The Gemara then says: ‘We have also been taught so in a Baraita: “A zebalgan should not lift up his hands but if he was familiar to the townspeople it is permitted.”’ The Gemara then remarks: ‘R. Johanan said: “A person blind in one eye should not lift up his hands.”’ The Gemara asks: ‘But was there not such a one in the neighbourhood of R. Yohanan who used to lift up his hands?’ To this the Gemara replies: ‘That one was familiar to his townspeople.’ The Gemara continues: ‘We have been taught so in a baraita: “A person blind in one eye should not lift up his hands but if he was familiar to the townspeople it is permitted.”’
Now first we note the curious coincidence that both R. Huna and R. Yohanan should have given a ruling having to do with the priestly benediction that was contradicted in their own neighbourhood. The most probable explanation is that there was doubt in the time of R. Huna and R. Yohanan whether a zebalgan or a person blind in one eye may lift up his hands but these teachers did not object to such persons doing so in their vicinity. In all probability these teachers pointed out that they only permitted it because the persons concerned were familiar to their townspeople but that otherwise it would have been forbidden. The Gemara simply restates this in the form of question and answer. Or it is possible that R. Huna and R. Yohanan simply tolerated these persons lifting up their hands because they saw no objection at any time to this and it is the Gemara which makes the distinction (between one familiar to the townspeople and one unfamiliar) and places this in the mouths of R. Huna and R. Yohanan.
In any event it seems certain that the distinction in these two cases arose out of actual incidents which occurred in the amoraic period. In that case it is surely stretching our definition of coincidence too far to suggest that quite fortuitously there happened to be two baraitot which made this distinction: one with regard to the zebalgan, the other with regard to the man blind in one eye. The most plausible explanation is that the baraitot are, in fact, fictitious, and postdate the amoraic distinction.
Of particular relevance to our discussion are the very frequent quotations of baraitot in the form: ‘In one baraita it was taught . . . but in another it was taught. . .’ and then an attempt is made to reconcile the baraitot. Higger gives a complete list of instances of this phenomenon.  Higger notes that most of these baraitot are not found elsewhere in the tannaitic literature! It is plausible to suggest that this is, in fact, an amoraic device, i.e. that the baraitot, at least in the form in which they are quoted, are fictitious, and are introduced as a pedagogical aid, i.e. in order to qualify the statement or law under consideration. As a bare statement this may sound unlikely but the following examples make it more convincing.
One baraita is quoted in which it is said that it is praiseworthy to bow during the thanksgiving benediction, but in another baraita it is said to be blameworthy. The Gemara declares that there is no contradiction: the one baraita refers to the beginning of the benediction (when it is praiseworthy to bow), the other to the end. The Gemara goes on to say that Rava used to bow at both the beginning and the end of the thanksgiving benediction. When the Rabbis asked him why he did this he replied that he had witnessed R. Nahman and R. Sheshet doing this. The Gemara objects: ‘But have we not been taught in a baraita that one who bows during the thanksgiving is blameworthy?’ I.e. one of the previously quoted baraitot says this and the proposed resolution of the contradiction by distinguishing between the beginning and the end is evidently not accepted by Rava and his teachers. To this the reply is given that the baraita does not refer at all to the thanksgiving benediction (in the Amidah) but to the thanksgiving in Hallel: ‘Give thanks unto the Lord for He is good’ (Ps. 118: 1). The Gemara objects: ‘But have we not been taught in another baraita that one who bows during the thanksgiving and during the thanksgiving of Hallel is blameworthy?’ I.e. this demonstrates that the first reference cannot be to the thanksgiving in Hallel. To this the reply is given that the reference is to the thanksgiving in the grace after meals.
We have in the sugya the following baraitot:
1) One who bows in the thanksgiving is praiseworthy.
2) One who bows in the thanksgiving is blameworthy.
3) One who bows in the thanksgiving and in the thanksgiving of Hallel is blameworthy.
In the Tosefta  it is said that one should bow during the thanksgiving (modim)  at the beginning and the end, i.e. as Raba and his teachers did. This passage from the Tosefta is quoted in the Jerusalem Talmud  but more significantly it is quoted at the beginning of this very sugya in the Babylonian Talmud. 
If the baraitot in the sugya are all authentic tannaitic productions we have the strange phenomena that they use a different term than the Tosefta for ‘thanksgiving’; that they appear to contradict the Tosefta; that this contradiction is not noted, only the contradiction with Rava’s practice: that one baraita simply refers to the thanksgiving while the other refers to the thanksgiving and the thanksgiving of Hallel; and that it all fits so conveniently into the argument as developed in the sugya. But if it is accepted that the pedagogical device of the fictitious baraita is common to all such instances in the Babylonian Talmud, all the difficulties are removed. The whole sugya can then be seen as a comment in the form of a sustained argument on the original ruling of the Tosefta. It is as if the sugya is saying that while the Tosefta rules that one should bow at the beginning and the end of modim, one should not conclude from this that it is similarly praiseworthy to bow at other forms of thanksgiving. On the contrary, one must not bow at the thanksgiving of Hallel and grace after meals.
The Mishnah  rules that if accidental homicide is caused by a downward movement (e. g. by someone coming down a ladder) the one guilty of it must go into banishment but if it is the result of an upward movement (e. g. by someone going up a ladder) he does not go into banishment. In the course of the discussion the Gemara suggests that both these rulings embrace, too, the case of a butcher cutting meat, i.e. if he kills in a downward movement he goes into banishment but not if he kills in an upward movement. Four baraitot are then quoted. They all begins with the words: ‘If a butcher [killed] while cutting [the meat]’ but they then give different rulings, namely:
1) In front he is liable, behind he is exempt. 2) Behind he is liable, in front he is exempt. 3) Whether in front or behind he is liable. 4) Whether in front or behind he is exempt. There is no contradiction, observes the Gemara. Ruling 1) deals with a downward stroke in front and an upward stroke behind; ruling 2) deals with a downward stroke behind and an upward stroke in front; ruling 3) deals with a downward stroke both in front and behind; and ruling 4) with an upward stroke both in front and behind. The Tosefta  states simply that if a butcher killed someone behind him he goes into banishment.
But further on in the Tosefta  it is stated that if the butcher killed while cutting meat he does not go into banishment. It is clear from the Tosefta that the distinction between normal cutting (in front) and killing behind him while swinging his chopper is this, that in the first case so little care has been taken that it comes near to intentional homicide (karov lemezid) and banishment is only for unintentional homicide. That this is so can be seen from the Tosefta’s comparison of the one case to that of a man who sits on a couch knowing full well that an infant is lying on it (and he smothers the infant) and the other to that of a man who sits on a couch without knowing that an infant is there.
Thus the distinction between a downward movement in front and a downward movement behind, etc. is not based at all on the Tosefta and would appear to be purely academic and amoraic. This only serves to increase the unlikelihood that there should be four actual baraitot, each with a different ending, which can be fitted so neatly and conveniently into the amoraic scheme. But if the baraitot are considered to be fictitious—in the nature of an amoraic pedagogical device to which we have called attention—everything falls quite naturally into place.
A baraita is quoted beginning with the words: ‘Be betrothed to me with this cup’.  It is said that there are three baraitot (i.e. three different versions of the original). In one it is said: ‘With it [the cup itself] and with its contents’, i.e. the marriage is valid if the cup together with its contents amounts to the value of a peruta (the minimum amount for kidushin to be effective) even if either is worth less than a peruta. In another baraita it is said: ‘With it and not with its contents’, i.e. the contents are disregarded and the cup on its own must be worth a peruta for the marriage to be valid. In the third baraita it is said: ‘With its contents and not with it’, i.e. the cup is disregarded and the contents on their own must be worth a peruta for the marriage to be valid. There is no contradiction, observes the Gemara. One baraita refers to a cup containing water, one to a cup containing wine, and one to a cup containing oil. 
Now it is not beyond the realm of possibility that there really were three actual Baraitot from the tannaitic period which the Gemara seeks to reconcile. But is it not more plausible by far to suggest that the concern of the Gemara is to state that there is no uniform rule in this matter since it all depends on the contents of the cup and that this is expressed by postulating three fictitious baraitot? Abraham Weiss would evidently call this suggestion an accusation of ‘forgery’ but if, in fact, as we have suggested earlier, this was an established and acknowledged pedagogical device in such circumstance, there is no need to put forward anything at all resembling an accusation of forgery on the part of the editors but simply to call attention to a well-known means of pointing out distinctions in the form of question and answer. 
To sum up: while I. H. Weiss goes too far in postulating that so many of the baraitot in the Babylonian Talmud are fictitious, Abraham Weiss is similarly in error in going to the opposite extreme. The evidence we have adduced goes to show that occasionally the redactors did use fictitious baraitot for the purpose of literary device and as a pedagogical means. The talmudic pilpul seems to have demanded at times the working up of rules and subtle distinctions into the question and answer form for the sake of greater effectiveness. There is more than a little truth in the old gibe about the yeshiva student who said that he had a marvellous answer and was now looking for a suitable question.
1. Dor Dor Vedorshav, ii. 242-4.
2. Leheker Hatalmud (New York, 1954), pp. 35-63.
3. On the general question of the use of literary device in the Babylonian Talmud see my Studies in Talmudic Logic and Methodology (London, 1961).
4. The version in Avodah Zarah adds here: ‘which minister before me’.
5. Maimonides, Yad Hahazakah, Beit habekhirah 7: 10, in fact concludes that, according to the final argument of the Gemara, the baraita regarding the copying of the Temple and its vessels has nothing to do with ‘Ye shall not make with Me’ but is because of the different principle of ‘fearing the Temple’—mora mikdash; see the commentary of Me’iri on this passage in Rosh Hashanah.
6. The version in Avodah Zarah adds here ‘on high’.
7. In Mekhilta on Exod. 20: 20, ed. Horowitz, p. 239, the reading is: ‘Ye shall not make the likeness of My attendants which minister before Me on high—neither the likeness of the angels nor the likeness of the Ofanim nor the likeness of the Cherubim.’
8. Unless the original reading was detanya the quote here is best understood on the lines of Abraham Weiss’s analysis that this interpretation of the verse was so well known in the Babylonian schools that it was always automatically associated with the verse itself.
9. As in the first part of the Mekhilta quoted in n. 7.
10. The procedure would be that of the novelist today who works out his plot first and then develops it gradually in order to keep the interest of his readers. It has been said that the detective story, for example, ‘is written backwards’.
11. The standard methodologies differ as to whether the intention in such instances is really to emend the Mishnah or that this is simply a way of explaining the Mishnah, i. e. the Mishnah does not have to be read differently but this is what it means. For the first view: Rashi on Berakhot 11b, s. v. af ligemara; Rashi on Megillah 28b, s. v. he tzana and the other sources quoted by Higger, Otzar habaraitot, vol. x (1948), 130-1. (Some of these are by no means conclusive.) For the opposite view see Shelah, Torah shebe’al peh, s. v. bekhamah mekomot.
12. Note the use of the Aramaic verabanan for the Sages instead of vehakhamim. Cf. the emendation and the baraita quoted in its support in Berakhot 13b-14a. There, however, the solution is far less forced and artificial.
13. See the standard commentaries to this passage.
14. Cf. Shabat 61a, where Mar son of Ravina is similarly said to have satisfied both requirements when an issue was in doubt. The chronology of our passage is difficult. The meaning can only be that the later redactors applied R. Nahman’s saying to Mar son of Ravina.
15. Both I. H. Weiss, Dor Dor Vedorshav, ii. 242, and Abraham Weiss, Leheker hatalmud, loc. cit., discuss these baraitot; see especially the complete list in Higger, Otsar habaraitot, vol. i (1938), chapter 8, pp. 74-108.
16. The use of Aramaic in many baraitot is not really germane to our problem. Even when it seems clear that these are amoraic additions in Aramaic to baraitot worded in Hebrew these are in all probability conscious elaborations by the amora’im and hence have nothing to do with our problem of the fictitious baraita, i. e. these additions were not presented in the first instance as part of the baraita but as amoraic elaborations. Higger, Otzar habaraitot, vol. i (1938), chapter 9, pp. 109-24, quotes a number of such baraitot, e.g. Berakhot 27b-28a; Berakhot 54a; Berakhot 62b; Pesahim 50b; Pesahim 112b. Higger in these instances (we omit Aramaic popular sayings, incantations and the like which, as Higger observes, may well have been recorded originally in Aramaic) expresses doubts whether these were originally part of the baraita or whether they are amoraic elaborations. It seems, in fact, certain from the style of these passages that they are amoraic elaborations and were never intended to be read as part of the baraita quoted, but see note 12.
17. Otzar habaraitot, vol. ix (1946), pp. 113-24.
18. Berakhot 1: 2.
19. In the quote from the Tosefta given in the Gemara at the beginning of the sugya (Berakhot 34a) the term used is hoda’ah.
20. Berakhot 1: 5.
21. Berakhot 34a.
22. Makot 2: 1.
23. Makot 2: 4.
24. Makot 2: 6.
25. The source is Tosefta Kidushin 2: 3.
26. See the standard commentaries in order to determine which is which.
27. Relevant to our theme is the question of the comparison of baraitot in the Babylonian Talmud with versions in the Jerusalem Talmud. Further investigation is required but see, for example, the story of R. Eliezer and the Sages in Bava Metsi’a 59b compared with the version in Moed Katan 3: 1. In the Jerusalem Talmud the story is told quite differently, is not quoted as a baraita, and contains none of the little asides found in the Babylonian Talmud and typical of it. It might also be noted that there is less of the miraculous in the Jerusalem Talmud version and this fits in with our knowledge of the style and approach of the Jerusalem Talmud. But if this is so the only conclusion to be drawn is that the baraita in the Babylonian Talmud in the form we have it is fictitious.