1. For a scholarly bibliography elaborating on intention's role in Judaism, see Eilberg-Schwartz, Howard, The Human Will in Judaism: The Mishna's Philosophy of Intention (Atlanta, Ga.: Scholars Press, 1986), 203 n. 9Google Scholar; and Botica, Aurelian, The Concept of Intention in the Old Testament, Philo of Alexandria and the Early Rabbinic Literature, (Piscataway, NJ: Gorgias Press, 2011)CrossRefGoogle Scholar. For an example of recent research on this topic in comparative religion, see Shaked, Shaul, “Religious Actions Evaluated by Intention: Zoroastrian Concepts Shared with Judaism,” in Shoshanat Yaakov: Ancient Jewish and Iranian Studies in Honor of Yaakov Elman, ed. Secunda, Shai and Fine, Steven (Leiden: Brill, 2012), 403–413CrossRefGoogle Scholar. For a historical overview, particularly of intention in Christianity, see Lacey, Thomas Alexander, “Intention,” in Encyclopedia of Religion and Ethics, ed. Hastings, James (New York: Charles Scribner's Sons, 1908), 7:380–382Google Scholar. For a summary of the philosophical discussion on intention, see Kieran Setiya, “Intention,” in The Stanford Encyclopedia of Philosophy (Spring 2011 edition), ed. Edward N. Zalta, http://plato.stanford.edu/archives/spr2011/entries/intention/.

2. B. Nazir 23a; B. Horayot 10a. All citations from Horayot and Nazir in this paper are from the Babylonian Talmud. For an overview of the rabbinic commentaries on RNBI's statement, see Rakover, Nahum, Matarah ha-mekadeshet ’et ha-’emẓa‘im (Jerusalem: The Library of Jewish Law, 2000)Google Scholar.

3. For RNBI's dates, see Cohen, Avinoam, Ravina ve-ḥakhmei doro (Ramat-Gan: Bar Ilan University Press, 2004), 69Google Scholar. For RNBI's unique linguistic style, see Margaliot, Reuven, Le-ḥeker ha-kinnuyim ve-ha-shemot (Jerusalem: Mosad Harav Kook, 1960), 29–37Google Scholar. On RNBI's methodology, see Epstein, Jacob N., Mevo'ot le-sifrut ha-’amora'im, ed. Melamed, Ezra Z. (Jerusalem and Tel-Aviv: Magnes and Dvir, 1963), 178Google Scholar. For a general overview of RNBI's life see Albeck, Chanoch, Mavo le-talmudim (Tel-Aviv: Dvir, 1969), 371–372Google Scholar. Regarding RNBI's cultural milieu and his connection to Raba, see Elman, Yaakov, “Rava as Mara de-Atra in Mahoza,” Ḥakirah 11 (2011): 69Google Scholar. On RNBI's unique character and his attraction to sin, see B. Shabbat 156a. There the Talmud reports that his mother was told that her son would become a thief. In order to prevent this, she told him to cover his head and pray for mercy. Once, while RNBI was studying Torah under a palm tree, his head covering fell off. He looked up and, noticing the tree, was immediately overcome by temptation for the dates. He climbed the tree and tore off a bunch of dates with his teeth. This story clearly reflects a Babylonian tradition concerning RNBI that attributes to him an inclination or attraction to sin from birth.

4. In other words, the transmission history will end with the medieval BT manuscripts and not venture beyond this point in time.

5. For a detailed discussion of these questions, see Yuval Blankovsky, “‘Averah li-shmah: le-korotav shel munnaḥ radikali ba-sifrut ha-talmudit” (PhD diss., Potsdam University, 2014).

6. Commenting on RNBI's statement, Nahum Rakover, “Matarah”, remarks that “due to its explosive nature, the principle has never become part of the mainstream of Jewish law, but has remained a relatively minor tributary” (p. xxii). This paper focuses on how the “explosive nature” of RNBI's statement influenced its talmudic sugya's formulation.

7. Kalmanofsky, Jeffrey, “Sins for the Sake of God,” Conservative Judaism 54, no. 2 (2002): 11Google Scholar.

8. Horayot, The Talmud of Babylonia: An American Translation, trans. Jaffee, Martin, ed. Neusner, Jacob (Atlanta: Scholars Press, 1987)Google Scholar.

9. In this context, it is worth mentioning that the word li-shmah meaning le-shem shamayim (for the sake of God) already appears in Y. Ḥagigah 2:1 (77c) see: Liebes, Yehuda, Ḥeta'o shel ’Elisha‘a (Jerusalem: Akademon, 1990)Google Scholar, 79 n. 22. Testimony to the interchangeability of the two terms may be found in a baraita that underwent the journey from Palestine to Babylonia: see T. Bikkurim 2:16; B. Pesaḥim 50b. Another option is to translate the word li-shmah as “proper intention” or “good intention”. The translation “for the sake of God” is preferred because of the linguistic connection we mentioned between li-shmah and le-shem shamayim in rabbinic literature.

10. This phrase means that the transmission of the sugya reflects increased cultural activity. As an indicator of this increase, compare the length of the shortest version of the sugya in MS Munich 95 of Horayot (around 330 words) to the longest version of the sugya in the Bomberg Talmud Horayot or any of the Nazir manuscript versions, which hover around 550 words. The sugya expanded by two-thirds, an increase that seems far greater than one would expect to find in comparing the average sugya's versions in the BT's manuscripts. The vast difference in scale between the manuscripts is exceptional; this article seeks to explain this difference. There are other sugyot in the BT where similar scales of difference occur. These other sugyot also deserve an explanation.

11. The analysis of the late additions is based on Shamma Friedman's principles on judging insertions: see “Perek ha-’ishah rabbah ba-bavli,” in Meḥkarim u-mekorot, ed. Dimitrovksi, Haim (New York: Jewish Theological Seminary, 1977), 1:301–308Google Scholar. An English translation of the definition of these criteria appears in the summaries at the end of the book (no page numbers given).

12. Nazir's textual witnesses are MN = MS Munich 95; V = MS Vatican 111; G = Ginzburg 1134; BN = Bomberg Talmud for Nazir. N designates these textual witnesses when taken as a single group.

13. Horayot's four textual witnesses are M = MS Munich 95; P = MS Paris 1337; I = Modena––Archivio Storico Comunale 26.1 (fragment from the Italian Geniza); B = Bomberg Talmud for Horayot. H designates these textual witnesses when taken as a single group.

14. The translation follows the text printed in the Vilna edition. See below for a comparison of the dialectical structures of the differing sugyot in N and H. Textual variants that contribute to an understanding of the preliminary sugya's structure and contents are discussed in the notes. These variants contribute significantly to the proposed dating of the additions introduced into this sugya.

15. In N the wording is “’eḥad ‘akholo le-shem/le-shum ’akhilah gasah,” but in H a briefer version omits the word le-shem/le-shum (with the intention of); the briefer version is more likely to reflect the preliminary sugya's form.

16. In N we find le-mishtei ḥamra, but in H, a briefer version omitting the word ḥamra (wine) appears. The briefer version is more likely to reflect the preliminary sugya's form.

17. MN, V, and G add the words beino u-bein Yisra'el (between him and Israel), “him” referring to Lot. In H the division is between Israel and Ammon (bein Yisra'el le-'Amon) and not between Lot and Israel.

18. In N the contentions are described as bars (ki-vriḥim), but in H they are not; in B the contentions are between Israel and Ammon, and in M and P the bars separate Israel and Ammon.

19. The Vilna printed edition (following BN) reads ve-'armon, which should be translated “and a castle,” not “of a castle.” MN and G read le-'armon, which the translation above follows. Perhaps BN adopted ve-'armon to signify that the word 'armon is not associated with the last derashah about the contentions but rather introduces a new derashah based on the phonetic similarity between 'armon and 'Amon.

20. P and B add she-nifrad me-’Avraham (who separated from Abraham), similar to line 56. In N and M that phrase is missing. It was probably a late addition to the passage, which explains why it is only found in some of Horayot's textual witnesses.

21. The Italian Geniza fragment on the sugya (I) starts at this point.

22. This line does not appear in H; see comment below.

23. BN, V, and G have ha-rasha‘a, (wicked) which does not appear in MN. In Horayot we find another description of Balak: M, P––“King of Moab”; B––“on the altar”; I––“King of Moab on the altar.”

24. This translation follows the version in N and I. B and P add an explanation: B––“from here” (me-hakha); P –– “from the daughter of Lot.” These are probably late glosses, which did not appear in the earlier sugya.

25. This translation follows the version in BN. In the Nazir manuscripts the explanation following the verse varies: MN, G, V ––the word kelal (at all) does not appear. The parallel line in Horayot's textual witnesses reads as follows: P, D––only the word kelal appears; I––no explanation follows the verse.

26. This is the tradition regarding the transmission of the derashah in N and B. In I, R.Yoḥanan is added to the chain of transmitters: R. Ḥiyya b. Abba in the name of R. Yoḥanan in the name of R. Yehoshua b. Karḥa; P is similar to I, but lacks R. Yehoshua b. Karḥa.

27. This translation follows the version in BN––be-Yisra'el la-malkhut. There are various versions in the other textual witnesses: NM, B––only la-malkhut (royal house); G, V––only be/le-Yisra'el (of Israel); I––le-malkhut ‘Oved, Yishai, David, u-Shelomo; P––be-Yisra'el le-malkhut ‘Oved, Yishai, David, u-Shelomo.

28. About this phenomenon in the BT and its connection to an oral culture see Elman, Yaakov, “Orality and the Babylonian Talmud,” Oral Tradition 14, no. 1 (1999): 86Google Scholar.

29. The phrase “similar literary structure” denotes that all three derashot divide the verse into its component parts and offer their commentary on each of these parts.

30. See Ruth Kaniel Kara-Ivanov “'Imahot u-paytanut be-mitos holadat mashiaḥ mi-beit David” (PhD diss., Hebrew University, 2010), 145 n.48.

31. Rabbinovicz, Raphaelo Nathan, Dikdukei sofrim (Munich: ‘A. Huber, 1879)Google Scholar, 10:34 n. 22.

32. The formulation that includes parts I and II appears in B. Pesaḥim 50b, B. Sotah 22a; B. ‘Arakhin 16b, as well as in our sugya. The formulation that includes parts I and III appears in B. Sotah 47a, as well as in our sugya. The formulation that includes parts I–III appears in B. Sanhedrin 105b.

33. Thus is the version in N. Although B quotes Rav's statement twice in the sugya it does not present Rav's second statement using this stylistic device. R. Joshua Boas, in Masoret ha-shas in the printed edition of B. Horayot 10a, noticing this absence, offers to fill it by adding the word gufa to the text. B's text may be explained as a partial copy of the parallel sugya in Nazir. This influence can also explain the different placement of the question on RNBI's statement in B (after line 89).

34. In another attempt to explain the variant textual witnesses, one might suggest that the omission of Rav's statement's first appearance in the Horayot manuscripts was the result of a copyist's error. According to this theory, after the mistake occurred, the next copyist tried to “correct” the text and did not copy the word gufa because in the text he had in front of him Rav's statement appeared only once. We must then also assume either that for some reason this copyist did not copy part II of Rav's statement or that the text he was copying from lacked part II. This line of thinking seems less probable.

35. The question and answer on RNBI's statement seem to be late additions to the earlier version of the sugya. These additions appear in all of Nazir's textual witnesses and we have no reason to suspect that this part of the sugya was not part of the Nazir sugya when the tractate was formulated.

36. For a list of rabbinic commentaries containing this deduction, see Nahum Rackover, Ends That Justify All Means, 42–45.

37. Surprisingly, this late addition to Horayot (B, I, P) is similar to the passage in Yevamot, not to the one in Nazir:

1. 1.

   1. In Yevamot (as in B, I, and P), the statement is attributed to R. Yoḥanan in the name of R. Shimon b. Yoḥai, but in N the statement is attributed to R. Yoḥanan alone.

2. 2.

   2. The wording bi-shlama ra‘a le-ḥayye in the majority of Yevamot's textual witnesses (l. 97) is similar to that in I (the only Horayot version that contains this line). In N the version is bi-shlama ra‘a shafir; this version only occurs in Yevamot's MS Munich 95.

These findings indicate that the transmitters of Horayot who included this passage took it from Yevamot or from a Nazir version unavailable to us. The differences between B, I, and P with regard to the scope of this addition indicate that none of these manuscripts is merely a copy of the other; nevertheless, all three manuscripts have one common denominator—their similarity to Yevamot, not to Nazir. It is difficult to date this addition; however, the fact that B, I, and P do not represent an independent version of the passage bolsters the assumption that this addition did not precede the formulation of these two variants (Yevamot and Nazir); on the other hand, the fact that the transmitters of Horayot took this passage from Yevamot or from a Nazir version that we do not possess indicates that it was introduced early on, before the prevalent version of the Nazir sugya was established or widely disseminated and accessible.

38. The fact that all three times these statements appear in the BT they are quoted together indicates that they were considered one literary unit. Comparing the versions of those statements in B, I, P, and N, and those versions and this passage in B. Bava Kamma 38b can shed light on the origins of this passage in those textual witnesses of Horayot that include it. The main differences between the various versions of the passage occur in the explanatory additions to the statements (lines 117, 120, and 125):

1. 1.

   1. The parallel of line 117 in B. Bava Kamma is “milḥamah hu de-lo ‘avid ha ’angarya ‘avid be-hu” and in correlation to that in line 120, “ve-’afilu ’angarya lo ta‘avid be-hu.” The word ’angarya means “the seizure of people or goods for public service” (Sokoloff, Michael, A Dictionary of Jewish Palestinian Aramaic of the Byzantine Period [Ramat-Gan: Bar-Ilan University Press, 2002], 64)Google Scholar. In N the wording is “milḥamah hu de-lo ha ẓa‘aurei ẓa‘arinan”—it is forbidden to cause Ammon sorrow. In contrast to that, in line 124 the text permits causing Moab sorrow––ẓa‘aurei ẓa‘arinan. The parallel in B to line 117 employs the word ’angarya, as in the Bava Kamma passage; I and P adopt N's style. The explanatory addition in line 120 does not appear in B, I, or P.

2. 2.

   2. At the end of the passage in B. Bava Kamma, an explanation appears about the provenance of Ammon's entrance into the Israelite nation: “‘Oved, Yishai, David, ve-Shelomo ve-’ilu ẓe‘eirah ‘ad Raḥava‘am she-ne'emar ve-shem ’imo Na‘amah ha-’Amonit.” Part of this explanatory addition appears in I and P, but it is absent from N and B.

None of Horayot's textual witnesses precisely matches the versions in Nazir or B. Bava Kamma: B is similar to B. Bava Kamma in line 117, but in line 125 it is similar to Nazir, while the opposite is true of I and P; all three (B, I, P) do not contain line 120, in contrast to the versions in Nazir and B. Bava Kamma. Nevertheless the versions of this passage in B, I, P do not seem to be another distinct variant on this passage because of their inconsistency. This inconsistency, which occurs neither in Nazir's textual witnesses nor in B. Bava Kamma's, suggests that this addition to Horayot is a result of copyists who copied from the passages in Nazir and B. Bava Kamma, respectively. The differences between the versions of this passage in B, I, and P and the versions in Nazir and B. Bava Kamma can be attributed to sloppy copying.

39. Epstein and Weiss argue about which of the parallel sugyot in Nazir and Horayot is the original and which the copy: see Jacob Nahum Epstein, Mevo'ot le-sifrut ha-'amora'im, 77; Weiss, Abraham, Le-korot hithavut ha-Bavli (Jerusalem: Makor, 1970), 50–51Google Scholar. Epstein states that Nazir was copied from Horayot. He does not make an explicit argument but his assumption that the Nazir sugya was copied from Horayot is based on his theory that tractate Nazir (like the other “exceptional tractates”) was edited later than the rest of the BT. Weiss's diametrically opposed conclusion is based on his analysis of the differences between the textual witnesses he had at his disposal: the Bomberg Talmud and MS Munich 95 for both tractates. He explains that the differences between B and M were the result of copyists' errors. It seems, however, that the very notion of treating the two texts as original and copy is inappropriate in this case. The similarity between the two sugyot indicates that they had a common ancestor—the earlier version of the sugya. The supposition that M preserves the latter sugya's structure better than any other textual witness fits in neatly with Epstein's opinion that the Horayot sugya predates the Nazir one. This notwithstanding, we cannot dismiss Weiss's assumption that the additions present in some of Horayot's textual witnesses were included because of the influence of the parallel Nazir sugya.

40. Friedman, Shamma Yehuda, Talmud ‘arukh: Perek ha-sokher ’et ha-'umanin (New York and Jerusalem: Jewish Theological Seminary, 1997), 66Google Scholar.

41. Indeed a long list of rabbis and scholars read those three stories as exemplars of ‘averah li-shmah; see ha-Meiri, R. Menaḥem, Beit ha-beḥirah, Horayot, ed. Sofer, Avraham (Jerusalem: Makor, 1969), 272Google Scholar; Kara-Ivanov, Ruth Kaniel, ‘Gedolah averah lishmah’, Nashim 24 (2013): 30Google Scholar.

42. Interestingly, Tosafot B. Nazir 23a, s.v. ve-dilma ’aihu nami le-shem shamayim mekhaven, cites a variant that replaces “intended to fulfill a commandment” (le-shem miẓvah), in line 31, with “for the sake of God” (le-shem shamayim). The possibility that this change has implications for the other temptation stories in the sugya is supported by Tosafot B. Nazir 23b, s.v. Tamar zintah. There the Tosafot comment that Tamar acted “for the sake of God” (le-shem shamayim). The phrase “intended to fulfill a commandment” (le-shem miẓvah) also appears in line 27, but we have no way of determining Tosafot's version there. This variation is also found in other indirect textual witnesses of the Nazir sugya: MSS Parma 3010 and London 406, which are manuscripts containing Talmudic aggadot; and in the first printed edition of Haggadot ha-talmud (Constantinople, 1511), which is another corpus of talmudic aggadot attributed to an unknown Spanish scholar. In line 27, these indirect textual witnesses replace “intended to fulfill a commandment” (le-shem miẓvah) with “for the sake of God” (le-shem shamayim). Probably these textual witnesses reflect a later evolution of the text which aims to solve the difficulty of the original version: it is not clear what is the commandment that Lot's daughter intended to fulfill.

43. The relocation and novel use of terms, expressions, and statements in new contexts endows them with new meanings. This phenomenon is typical of rabbinic literature. The phenomenon can be viewed from the perspective of Wittgenstein's language philosophy as a “language game.” For example, we can examine the sugya's novel use of the phrase kavvanah le-shem miẓvah in the context of biblical female figures' sexual sins. Likewise, we can explore the use of the term li-shmah as it is used, not only in the context of fulfilling commandments but also, as RNBI innovates, in the context of committing transgressions. By examining a word or phrase's prior usage, we can better apprehend the meaning or the possible meanings of the new usage. About the implication of this approach for the interpretation of the talmudic literature see: Yuval Blankovsky, “Ma'amar ‘al parshanut ha-Talmud,” (Jerusalem: Orientation, 2013), http://www.daat.ac.il/daat/vl/belan-talmud/belan-talmud01.pdf.

44. Tamar's marriage is explicitly a levirate one, as Genesis 38:7 attests; and, as mentioned above, the sugya uses the unique terminology applied to levirate marriage in discussing Lot's daughters’ story.

45. On Christian Doctrine, in The Nicene and Post-Nicene Fathers, ed. Schaff, Philip (Grand Rapids: Eerdmans, 1978–1979), 2: 560–561Google Scholar (III.12.18).

46. Future research along similar lines can be undertaken concerning Rava's dictum “even for a matter of transgression” (B. Berakhot 63a). This dictum appears in the context of a discussion of the verse: “In all your ways know the Lord” (Proverbs 3:6). On the connection between Rava's statement and RNBI's statement, see ha-Meiri, R. Menaḥem, Beit ha-beḥirah, Nazir, ed. Lis, Avraham (Jerusalem: The Institution of the Complete Palestinian Talmud, 1967), 81Google Scholar; Rakover, Matarah, 55–59; Urbach, Ephraim Elimelech, Ḥazal: Pirkei ’emunot ve-de‘ot (Jerusalem: Magnes, 1983), 300Google Scholar.