1. Heineman, Yizhak, Darkhei ha-aggadah (Jerusalem: Magnes, 1970), 137–38Google Scholar; 156–57; Kamin, Sarah, Rashi: Peshuto shel mikra u-midrasho shel mikra (Jerusalem: Magnes, 1986), 82–83Google Scholar; Japhet, Sara, Dor dor u-parshanav (Jerusalem: Bialik, 2008), 135–37Google Scholar. For a comprehensive review of the history of allegorical exegesis on Song of Songs, see: Pope, Marvin H., Song of Songs—A New Translation with Introduction and Commentary (AB) (New York: Doubleday, 1977), 89–112CrossRefGoogle Scholar. Scholars have identified a few solitary instances of a literal commentary on a single verse from Songs; for a review of these see Kadari, Tamar, “Rabbinic and Christian Models of Interaction on the Song of Songs,” in Interaction between Judaism and Christianity in History, Religion, Art and Literature, ed. Poorthuis, Marcel, Schwartz, Joshua and Turner, Joseph, Jewish and Christian Perspectives 17 (Leiden: Brill, 2009), 74–77Google Scholar; Kadari, Tamar, “Li-she'elat kehal ha-yaʽad shel sifrut ha-midrash: ‘al rovdei parshanut u-she'elat ha-ʽarikhah,” in Doresh tov le-'amo—ha-darshan, ha-derashah ve-sifrut ha-derush be-tarbut ha-yehudit, ed. Kaplan, Kimmy, Horowits, Carmi, and Ilan, Nahem (Jerusalem: Zalman Shazar Center, 2012), 37–41Google Scholar. It should be noted that Origen interprets the book systematically on three levels, with the first level being a literal description of a bride longing for her betrothed. See: Torjesen, Karen, Hermeneutical Procedure and Theological Method in Origen's Exegesis (Berlin: W. de Gruyter, 1985), 54–57Google Scholar, 156–60; Kadari, “Li-she'elat,” 42–44.

2. For a discussion of R. Tuviah, his location, and his period, see Midrash lekaḥ tov—Bereshit-Shemot, ed. Buber, Shlomo (Lvov: n. p., 1878), 20–21Google Scholar; Inbal Touitou, “Darko ha-parshanit shel rabbi Tuviah ben ’Eli‘ezer be-perusho Lekaḥ tov la-torah—ʽiyyun be-masekhet sipurei ha-'avot (Bereshit 12–36)” (Master's thesis, Bar-Ilan University, 2005), 1–6; Ta-Shma, Israel M., Kenesset meḥkarim— ʽiyyunim ba-sifrut ha-rabanit bi-yemei ha-benayim, vol. 3, ’italiah u-bizantion (Jerusalem: Bialik, 2005), 259–60Google Scholar.

3. I have described elsewhere R. Tuviah's approach in his commentary to the literal level of the text, as well as the surprising similarity between his commentaries and those of Rashi. See Jacobs, Jonathan, “Perush Lekaḥ tov ‘al shir ha-shirim: mekomo be-toldot parshanut shir ha-shirim ve-yaḥaso le-perush Rashi la-megilah,” Shenaton 23 (2014): 225–241Google Scholar.

4. Urbach, Ephraim E., “Derashot ḥazal u-perushei ’Origines le-shir ha-shirim ve-ha-vikuaḥ ha-yehudi-noẓri,” Tarbiẓ 30 (1960): 148–49Google Scholar. Concerning rabbinic allegorical exegesis, see further: Lachs, Samuel Tobias, “Prolegomena to Canticles Rabba,” JQR 55 (1965): 235–55CrossRefGoogle Scholar; Stern, David, “Ancient Jewish Interpretation of the Song of Songs in a Comparative Context,” in Jewish Biblical Interpretation and Cultural Exchange: Comparative Exegesis in Context, ed. Dohrmann, Natalie B. and Stern, David (Pennsylvania: University of Pennsylvania Press, 2008), 89–92Google Scholar; Goshen-Gottstein, Alon, “Did the Tannaim Interpret the Song of Songs Systematically? Lieberman Reconsidered,” in Vixens Disturbing Vineyards: Embarrassment and Embracement of Scriptures, in Honor of Harry Fox, ed. Yoreh, Tzemah, Glazer, Aubrey, and Segal, Miryam (Boston: Academic Studies Press, 2010), 260–71Google Scholar. Recently, Boyarin and Goshen-Gottstein have proposed that the rabbinic commentary on Songs does not in fact meet the criteria of allegorical exegesis; see: Boyarin, Daniel, Intertextuality and the Reading of Midrash (Bloomington: Indiana University Press, 1990), 105–110Google Scholar; Goshen-Gottstein, Alon, “Pulmusomania—hirhurim metodiyim ʽal ḥeker ha-vikuaḥ ha-yehudi-noẓri be-ʽikvot perushei ḥazal ve-’Origines le-shir ha-shirim,” Jewish Studies 42 (2003–2004): 153–61Google Scholar.

5. Lieberman, Saul,“Mishnat Shir Ha-Shirim,” in Jewish Gnosticism, Merkabah Mysticism and Talmudic Tradition, ed. Scholem, Gershom (New York: JTSA, 1965), 118–26Google Scholar. In his view, the tannaitic sages interpreted the text in a consistent manner in each of the four historical allegorical modes mentioned above, each according to his own approach. The amoraic sages interwove these different approaches, creating a rabbinic midrash that comprises all of the historical allegories together.

6. Heineman, Darkhei ha-aggadah, 137–38; Kamin, Rashi, 82. This fact does not necessarily mean that the midrash itself follows no organized progression; see Tamar Kadari, “Li-melekhet ha-ʽarikhah be-midrash shir ha-shirim” (PhD diss., Hebrew University, 2004), 365–83. For a presentation of interpretations of several consecutive verses in Songs, see: Kadari, Tamar, “Le-tofaʽat ha-petirot ha-ḥozrot be-midrash shir ha-shirim rabbah,” Tarbiẓ 80 (2012): 363 and n. 3Google Scholar. Kadari notes rabbinic interpretations of units comprising four, five, and six verses.

7. Churgin, Pinkhos, The Targum to Hagiographa (New York: Horev, 1945), 129Google Scholar; Alexander, Philip S., The Targum of Canticles, The Aramaic Bible 17a (Collegeville, MN: Liturgical Press, 2003), 13–19Google Scholar.

8. Alexander, The Targum, 13–19; Baruch Alster, “’Ahavah ’enoshit ve-zikatah le-’ahavah ruḥanit ba-parshanut ha-yehudit le-shir ha-shirim” (PhD diss., Bar-Ilan University, 2006), 30. It should be noted that the exegesis of the Targum was preceded by that of the Christian scholar Apponius who proposed that Songs describes the relationship between Jesus and the Church as a historically evolving plot, but it would seem that the author of the Targum was not familiar with Apponius. See Alexander, Philip S., “The Song of Songs as Historical Allegory—Notes on the Development of an Exegetical Tradition,” in Targumic and Cognate Studies, JSOTSS 260, ed. Kathcart, Kevin and Maher, Michael (Sheffield: Sheffield Academic Press, 1996), 26–27Google Scholar.

9. The excerpt is at the end of Rashi's commentary on Song of Songs, in MS New York - Jewish Theological Seminary Lutzki 778 (f 24010), and it reads as follows:

My thanks to Prof. Yosef Ofer for drawing my attention to this Masoretic fragment.

10. Rashi makes no mention of the Targum as the source for his commentary, and in the past it has been claimed that Rashi's commentary is not dependent on the Targum—see Rosenthal, Rashi, 7; Kamin, Rashi, 259 n. 112. Today, the consensus rests with the opposite view. According to Alexander, what Rashi proposes is not just an interpretation of Songs, but also, in a certain sense, an interpretation of the Targum on Songs. See Alexander, The Targum, 45. Concerning the relationship between Rashi and the Targum see also Marcus, Ivan G., “The ‘Song of Songs’ in German Hasidism and the School of Rashi, a Preliminary Comparison,” in Rashi 1040–1990, Hommage á Ephraïm E. Urbach, ed. Sed–Rajna, Gabrielle (Paris: Cerf, 1993), 265–72Google Scholar; Alexander, “Historical Allegory,” 23. In Alster's view, the difference between the Targum and Rashi is that the former suggests a prospective, developing plot, while the latter molds a retrospective plot that starts mid-story. See Alster, Baruch, “‘Ha- ’ishah ha-ʽazuvah’ ve-parshanut shir ha-shirim,” JSIJ 5 (2006): 101–103Google Scholar. For a discussion of the relationship between the midrash, on one hand, and the Targum and Rashi, on the other, see Gottlieb, Isaac B., “The Jewish Allegory of Love: Change and Constancy,” Journal of Jewish Thought & Philosophy 2 (1992): 1–17CrossRefGoogle Scholar.

11. See the introduction to his commentary. For a discussion of the divine inspiration that the sages attribute to Solomon, see Kadari, Tamar, “Li-she'elat mekoriyutan shel ha-petiḥta'ot she-be-rosh midrash shir ha-shirim rabbah,” Tarbiẓ 75 (2005–2006): 167–69Google Scholar. She posits that Midrash Shir ha-Shirim Rabbah diminishes significantly the role of Solomon as author of Songs; see ibid., 172.

12. Kamin, Rashi, 78–79; 249. The Karaite commentators Salmon ben Yeruḥam and Yefet ben ʽEli similarly understand Solomon's view of history as reflecting a prophetic perspective of their own period—the tenth century. See Frank, Daniel, “‘Ve-kol ha-tor nishmaʽ be-’arzenu’: perushei ha-kara'im Salmon ben Yeruḥam ve-Yefet ben ʽEli le-shir ha-shirim,” International Rennert Guest Lecture Series 7 (Ramat Gan: Bar-Ilan University, 2001), 12–13Google Scholar and n. 47.

13. The occasional verses that disturb the chronological progression are explained by Rashi as flashbacks or flashes of foresight. There are a number of different possibilities for analyzing the stages of development of the plot according to Rashi's interpretation; see, for example, Alster, “’Ahavah ’enoshit,” 43–44.

14. This assertion runs contrary to the opinion of Grossman, who draws a comparison between Rashi and R. Tuviah in this regard. See Grossman, Avraham, “R. Shemaʽya ha-shoshani u-perusho le-shir ha-shirim,” in Sefer ha-yovel la-rav Mordekhai Breuer, ed. Bar-Asher, Moshe (Jerusalem: Academon, 1992), 49Google Scholar.

15. In CGW the word שירים here is omitted.

16. Concerning Solomon and his composition of the Songs with divine inspiration, see R. Tuviah, pp. 6, 7, 10, 11, and elsewhere.

17. R. Tuviah uses these same terms in his commentary on the Pentateuch, but less frequently. He sometimes refers to the present as ציווי עומד—see Genesis 6:9, and to the future as עתידות—see Exodus 10:26. His comment on Exodus 7:9 is worthy of note:

18. See, for example, Shir ha-Shirim Rabbah, 66; 77; Midrash Shir ha-Shirim, 55.

19. M: יתן רשות לישר' ליפרע מן האומות.

20. CGNW: מתוכה.

21. The parallel between Moses and Elijah is found in ancient midrashic sources, too, but as a local allusion pertaining to just one verse. See, for example, Midrash Shir ha-Shirim, 61.

22. His commentary includes many more dualistic allegorical interpretations. In some instances he refers explicitly to “past-future”—see, for example, 36–37; 40; 44; 46–47; 53–54; 85; 89–90. In other instances he offers a dualistic allegorical interpretation without explicit reference to the two time frames.

23. According to Kamin, this is intentional, so as not to undermine the intrinsic value of the exile (Kamin, Rashi, 256). More simply, however, it is a structural function of the chronological allegorical reading, which necessarily reaches the future stage only at the end.

24. In M the phrase is more elaborate: מנגדי מאתי שימי לבך לכל העניינים שעשיתי. The same formula appears as a note in N.

25. ben-Shammai, Haggai, “Meẓi'a ’aḥat she-hi shetayim: perush ha'azinu le-R. Shmu'el ben Ḥofni u-perush va-yosha' le-rav Saʽadiah Ga'on be-ketav yad nishkaḥ,” Kiryat Sefer 61 (1986–1987): 323Google Scholar. For R. Sa‘adiah Gaon's position regarding tenses in the biblical text, see Dotan, Aron, ’Or rishon be-ḥokhmat ha-lashon—sefer ẓaḥut lashon ha-ʽivrim le-rav Saʽadiah Gaon (Jerusalem: World Union of Jewish Studies, 1997), 149–51Google Scholar.

26. Other midrashim indicate the uniqueness of Songs in relation to other songs, but they adopt different exegetical positions. See Shir ha-Shirim Zuta, 4; Midrash Shir ha-Shirim, 4.

27. Buber makes brief mention of the issue, see: Lekaḥ tov, 43.

28. For narrower and broader definitions of the boundaries of the Judeo-Christian polemic (in the context of rabbinic exegesis and the exegesis of Origen on Songs) see Goshen-Gottstein, “Pulmusomania,” 134–48. Goshen-Gottstein adopts an extremely narrow technical definition of the polemic, and he concludes that there are no traces of it in either the rabbinic or the Christian exegesis on Songs. I subscribe to the more widely accepted definition, maintaining that it includes also “the clarification of self-identity … on the basis of the positions held by others, with the aim of distinguishing between us and them” (see ibid., 139 n. 67). Ta-Shma comments that in R. Tuviah's Commentary on the Pentateuch there is no hint of any anti-Christian polemic, perhaps owing to self-censorship; see Ta-Shma, ’Italiah, 269 n. 32. It should be noted that some scholars view parts of the Aramaic Targum as anti-Christian polemic: see Loewe, Raphael, “Apologetic Motifs in the Targum to the Song of Songs,” Biblical Motifs: Origins and Transformations, ed. Altman, Alexander (Cambridge: Harvard University Press, 1966), 173–84Google Scholar.

29. Kamin, Rashi, 251–62; Kamin, Sarah, “Perush Rashi ʽal shir ha-shirim ve-ha-vikuaḥ ha-yehudi-noẓri,” in Bein yehudim le-noẓrim be-parshanut ha-mikra (Jerusalem: Magnes, 2008), 24–35Google Scholar. In Alster's view, Rashi's polemic is not covert; see Alster, “Ha-’ishah,” 101–22.

30. Kamin, Rashi, 261.

31. Grossman makes very brief mention of R. Tuviah's attention to the polemic—see Grossman, “R. Shmaʽya,” 493.

32. Kamin, Rashi, 261.

33. ANP: ופשטו.

34. R. Tuviah also mentions the First Crusade in his commentary on Leviticus 22:33.

35. AMP: כיון שרבו הצרות והתחילו ימי.

36. M inserts the word התחילו; likewise a note in N.

37. See pp. 21, 25, 69, 72, 96, 99, and 100. In his commentary on the Pentateuch, R. Tuviah engages extensively in polemic against the Karaites: see Buber, Lekaḥ tov, 34–35; Ankori, Zvi, Karaites in Byzantium—The Formative Years 970–1100 (New York: Columbia University Press, 1959), 261–91CrossRefGoogle Scholar; Ta-Shma, ’Italiah, 269–77. So far I have not found evidence of anti-Karaite polemic in his commentary on Songs—perhaps because the nature of this text does not lend itself to it, but the issue needs to be examined, through careful comparison to the Karaite commentary on Songs by Salmon ben Yeruḥam and Yefet ben ‘Eli.

38. Jacobs, “Lekaḥ tov.”

39. Kamin claims that in the sources that Rashi used, the word dugma did not appear as a reference to the allegorical level. See Kamin, Sarah, “‘Dugma’ be-perush Rashi le-shir ha-shirim,” in Bein yehudim le-noẓrim be-parshanut ha-mikra (Jerusalem: Magnes, 2008), 70Google Scholar. Ta-Shma proposes that the term appears already in the commentaries on piyyutim (liturgical poems) that were composed in Mainz; see Ta-Shma, Israel M., Kenesset meḥkarim: ʽiyyunim ba-sifrut ha-rabanit bi-yemei ha-benayim, vol. 1, Ashkenaz (Jerusalem: Bialik, 2004), 85–86Google Scholar. See Kamin's response—Rashi, 83 n. 77. Apparently, neither scholar was aware of the use of the term דוגמא by R. Tuviah.

40. Of course, it is possible that the commentators based themselves on a midrashic source that has not reached us. There is a certain resemblance here to the Aramaic Targum,

Likewise, there is some similarity to the yoẓer recited on the first day of Passover, "שִׁיר הַשִּׁירִים אוֹרָה וְשִׂמְחָה לַמְטִיבִים", composed by Rabbi Shelomo Suleiman Al-Sanjary, and containing the liturgical phrases, "שְׁנֵי שָׁדַיִךְ צֶמַח וְתִשְׁבִּי אוֹתָךְ כְּהִבְחִין \ יְגַל שְׁנֵיהֶם מַלְכֵי עַמִּים לְהַגְחִין". This poem is to be found in MS Strasbourg, the Israel National Library, 4077/81–85 (145/21) (f 4384). My thanks to Dr. Eden HaCohen who shared this finding with me. For more on this tenth-century liturgical poet of the Land of Israel, see Eden HaCohen, “Kedushta'otav shel Rabbi Shelomo Suleiman Al-Sanjari le-moʽadei ha-shanah” (PhD diss., Hebrew University, 2003), 222–30. The linguistic similarity to both the Targum and to the piyyut is weak, while the linguistic similarity between Rashi and R. Tuviah is clearly evident; for this reason I believe that there is no basis for the assumption that the piyyut is a common source that Rashi and R. Tuviah each saw, separately. Nevertheless, it is entirely possible that what we have here is an echo of a tradition of the Land of Israel that connects Pinḥas, Elijah, and the future righteous priest—a tradition that reached R. Tuviah in Byzantium, from whence it reached Rashi in France.

41. In his commentary on 4:5; he writes: "אלו מלך המשיח וכהן צדק העתידים להושיע את ישראל כדרך שעשו משה ואהרן"—“These are the King Messiah and the righteous priest who are destined to deliver Israel just as Moses and Aaron did,” and he then goes on to create a complex analogy between Moses and Aaron, on one hand, and the Messiah and the righteous priest, on the other. The fairly unusual title kohen ẓedek—righteous priest—is found in Avot de-Rabbi Natan (ed. Schechter, p. 100), as well as in B. Sukkah 52b, and in some other midrashic sources. The phrase is strongly reminiscent of the moreh ẓedek (righteous leader) familiar to us mostly from the Dead Sea Scrolls, although it also appears elsewhere.

42. For a list of criteria for comparisons of medieval commentators, see Jacobs, Jonathan, “Ha-’im hekir Ramban ’et perush Rashbam la-torah?” Mada‘ei Ha-yahadut 46 (2009): 86–90Google Scholar.

43. According to Touitou, the places in Rashi's commentary where traces of Lekaḥ tov are to be found are not original to Rashi, but rather later additions. See Touitou, Elazar, “ʽIkvot shel Lekaḥ tov be-perush Rashi la-torah,” ʽAlei Sefer 15 (1989): 37–44Google Scholar. Ta-Shma maintains that Rashi and R. Tuviah saw each other's commentaries after they had each completed the writing of their own works, and they introduced fragments of each other's work in later addenda. See Ta-Shma, ’Italiah, 179–80, 266–67 and n. 25. R. Tuviah appears to have completed his work in the mid-1090s. According to Buber, he publicized his commentary on the Pentateuch in 1097, and after editing and amendments it was reissued in 1107. See Buber, Lekaḥ tov, 25. See also Ta-Shma, ’Italiah, 263. According to Kahana, Midrash Lekaḥ tov on the book of Numbers was composed already in 1094; see Menaḥem Kahana, “’Akdamut le-hoẓa'ah ḥadashah shel Sifri Bamidbar” (PhD diss., Hebrew University, 1982), 98 and n. 13. The dissemination of this work was relatively limited in scope; see Buber, Lekaḥ tov, 58–59; Jacobs, Jonathan, “Le-birur heikef heikeruto shel Rashbam ʽim midrash Lekaḥ tov,” in Ta-Shma: Meḥkarim be-madaʽei ha-yahadut Le-ziḥro shel Yisra'el M. Ta-Shma, ed. Reiner, Avraham (Allon Shevut: Tevunot, 2011), 475–99Google Scholar; Jacobs, Jonathan, “Sefarim ḥadashim she-hitgalu la-Ramban be-hagiʽo le-’ereẓ yisra'el,” JSIJ 11 (2012): 1–14Google Scholar. According to Gelles, Rashi completed his commentaries on the Bible in the mid-1090s; see Gelles, Binyamin, Peshat u-derash be-parshanuto shel Rashi (Jerusalem: Elinar, 1992), 146–48Google Scholar. Therefore it seems that we must accept Ta-Shma's assumption that if the two scholars were indeed exposed to each other's commentary, this would only have been after they had each completed their respective commentaries on the Pentateuch.

44. The prevailing view is that Rashi wrote first his commentary on the Pentateuch, then on the historical works—the books of the Prophets—and finally, on the Hagiographa. See Poznanski, Samuel, Mavo ʽal ḥakhmei ẓorfat mefarshei ha-mikra (Warsaw: Mekitzei Nirdamim, 1913), XIVGoogle Scholar; Gelles, Peshat u-derash, 144–45; Florsheim, Yoel, “Kelalei lashon ha-mikra be-perush Rashi la-Talmud,” Lešonénu 38 (1974): 254Google Scholar. Gelles claims that his exegesis on the entire Bible was completed by the mid-1090s. However, his proofs refer to the commentaries on the Pentateuch and on the Prophets. To my mind, it is reasonable to assume that Rashi wrote his commentary on the Hagiographa, including the Five Scrolls, in his final years, after he had completed his commentaries on the Pentateuch and the Prophets. As we know, he never reached the end of his commentary on Job, and never wrote a commentary on Chronicles, Ezra, or Nehemiah. See also Ahrend, Aaron, Perush Rashi le-masekhet megilah (Jerusalem: Mekitzei Nirdamim, 2008), 24–25Google Scholar.

45. Ta-Shma, Ashkenaz, 96–114; Ta-Shma, ’Italiah, 177–87, 202–17.

46. Steiner, Richard C., “A Jewish Theory of Biblical Reduction from Byzantium: Its Rabbinic Roots, Its Diffusion and Its Encounter with the Muslim Doctrine of Falsification,” JSIJ 2 (2003): 145–46Google Scholar; Steiner, Richard C., “The ‘Lemma Complement’ in Hebrew Commentaries from Byzantium and Its Diffusion to Northern France and Germany,” JSQ 18 (2011): 377–79CrossRefGoogle Scholar. For a discussion of R. Shema'ya's possible Byzantine origins, see Grossman, Avraham, Ḥakhmei ẓarfat ha-rishonim (Jerusalem: Magnes 1995), 350–52Google Scholar.

47. Elbaum, Jacob, “Yalkut sekhel tov: derash, peshat ve-sugiyat ha-sadran,” in Davar davur ‘al ’ofanav—meḥkarim be-parshanut ha-mikra ve-ha-koran bi-yemei ha-benayim, ed. Bar-Asher, Moshe, Hopkins, Simon, Stroumsa, Sarah, and Chiesa, Bruno (Jerusalem: Ben-Zvi Institute, 2007), 81Google Scholar.