Continue to Sing, Miriam! The Song of Miriam in 4Q365

In my contribution to our panel I will explore the Song of Miriam preserved in the Qumran manuscript 4Q365 – a text that has been transmitted to us under the titles 4QReworked Pentateuchc and 4QPentateuchc. By focusing on the material evidence, I will discuss how this manuscript provides important insight into how the Song of Miriam was understood by its ancient audiences.

1.     Song of Miriam in 4Q365

Manuscript 4Q365 frag. 6 preserves parts of Exod 14-15: the children of Israel escaping from Egypt, crossing of the Sea of Reeds and the songs performed by Moses and Miriam. Significantly, 4Q365 6a I preserves Exod 14:12-21: Israel’s escape from Egypt and Moses’ parting the Sea of Reeds. The manuscript then breaks, and the following fragment 6b continues the text of Exod 15:16, beginning part way through the song of Moses. Emanuel Tov and Sidnie White Crawford, the editors of the manuscript, assume that the gap between the two fragments could have preserved around 28 lines of text.[1] Supposing that the text followed the Exodus narrative, it could have contained vv. 14:22-15:15. Fragment 6b, placed now in the bottom of the column finishes with Exod 15:20, but it is possible that v. 21 also belonged to the text. In the next column, 6a II, the text picks up from Exod 15:22, which narrates Moses leading the people from the Red Sea in line 8, and continues the text until Exod 15:26. Of significance for this discussion, the text 6a II +6c adds at least 7 lines of previously unknown material.[2] As this expansion belongs to a passage prior to the narrative, Tov and White Crawford attributed it to the song of Miriam.

As we turn to this “new material,” it should be emphasized that while the scholars refer to “the previously unknown material” the manuscript itself does not demonstrate any signs of it being an addition. Importantly, the 7 lines of new material are integrally interwoven into the surrounding text. The text is written by the same hand as the rest of the fragment, paleographically dated ca. 75 BCE.[3] Thus, we should assume that one scribe wrote the whole passage. This observation suggests that this song of Miriam belongs essentially to the manuscript 4Q365. Therefore, while we refer to these 7 lines as “addition”, the ancient audience may not have perceived it in the same way. Moreover, this language (i.e., “expansion” / “addition”) reveals the tendency to assume the Masoretic Text (hereafter the MT) is the “more original” text that was later expanded.  

Tov and White Crawford claim that the style of the scribe who copies 4Q365 reflects the system of orthography and morphology of the so –called “Qumran system”.[4] They therefore propose that the scribe who copied 4Q365 could have been a representative of the Qumran scribal school and that the text was possibly copied at Qumran.[5] The book of Exodus features prominently within the context of the Dead Sea Scrolls. Significantly, at least 14 manuscripts preserve copies of Exodus.[6] In addition, the DSS contain numerous texts that depend on Exodus traditions and rework them. Despite this general familiarity with Exodus, the collection preserves only one copy of the Song of Miriam - the version preserved in 4Q365. Given the lack of any other material evidence to the song of Miriam one should be careful to not make assumptions about other copies that possibly circulated around the same time.

I have examined the vocabulary of this expansion previously so let me just highlight some key findings. Tov and White Crawford suggested that the Song of Miriam in 4Q365 employs some distinct terms that are well known from the Song of Moses.[7] Both I and Ariel Feldman, who has recently reconstructed this song, agree with them. In particular terms that appear in lines 1-2, where 4Q365 reads כי גאה גאה, (cf. Exod 15:1) and 5-6, where 4Q365 preserves במים אדירים (cf. Exod 15:10) draw from the Song of Moses. [8] However, as I have demonstrated elsewhere, the Song of Miriam also contains terminology independent from the Song of Moses, including some shared vocabulary with the Aramaic Targumin.[9] Therefore, the Song of Miriam in 4Q365 should not be viewed as a simple remake of the Song of Moses.

As manuscript 4Q365 preserves and presents the songs of Moses and Miriam one after another, the audience of the text would have been familiar with both of them and perhaps associated one with the other. Therefore, it is important to ask how they possibly relate to one another. In light of the preserved material, we can say that 4Q365 presents the two songs in a more balanced way than the MT where Miriam only repeats one line from the song of Moses.[10] Meanwhile, 4Q365 portrays Miriam as a character who performs next to Moses independently and whose song is more extensive than what the MT preserves. The more balanced portrayal in 4Q365 could have led the audience to see the two as equally important figures in the context of Exod 15.

Ancient Jewish narratives composed in Greek offer additional support for the theory that for many Jews of antiquity, the song of Miriam was understood to be more than repetition of one line of the song of Moses. The LXX translation of the passage employs two terms that deserve a mention because it is my view that they present the singing and dancing in a more organized and structured way than the MT. First, Exodus 15:21 of the MT reads: “Miriam sang to them” (NRSV). The verb the LXX employs here for the same verse is ἐξάρχω, which literally means “to begin, to act as a leader, to initiate.”[11] The verb was used in classical Greek, where it was understood in particular Greek tragedies to designate alternating choruses.[12] Further, the LXX translation of Exodus uses the term χορός which means a dance or a band of dancers when it talks about the two choruses led by Moses and Miriam. The use of the term χορός[13] suggests that the women’s singing on the shore of the Red Sea was not interpreted as a random event or a spontaneous act of celebration. Rather, the women were thought to form a group of dancers, similar to those that performed at other festivities.[14] This perspective on the songs reveals that Moses and Miriam were imagined to lead the two choruses on their turns. Also in this portrayal the two choruses appear to be more balanced than what is preserved in the MT.

The idea of the two choruses is also present in the writings of Philo of Alexandria, who utilized the LXX. He discusses Exod 15 and the songs in two of his treatises: in De Vita Contemplativa and De Agricultura. In De Vita Contemplativa, Philo portrays men and women of the Therapeutic community performing imitations of the two choruses of Exodus 15. While the groups of men and women performed the celebration, Philo relates that, as in Exod 15:20, Miriam acted as the leader for the women’s chorus.  

(87) When the Israelites saw and experienced this great miracle, which was an event beyond all description, beyond all imagination, and beyond all hope, both men and women together, under the influence of divine inspiration, becoming all one chorus, sang hymns of thanksgiving to God the Saviour, Moses the prophet leading the men, and Miriam the prophetess leading the women. (88) Now the chorus of male and female worshippers being formed, as far as possible on this model, makes a most humorous concert, and a truly musical symphony, the shrill voices of the women mingling with the deep-toned voices of the men. The ideas were beautiful, the expressions beautiful, and the chorus-singers were beautiful. (Contempl. 87-88).[15]

Significantly, also in this literary context the Song of Miriam is not presented as a one verse repetition from the Song of Moses, but as a distinct independent singing. Philo provides as much attention to both female and male choruses and emphasizes that women’s performance that makes the celebration more complete.[16] This analysis reveals how problematic it is to take the MT as the starting point for the analysis of the Song of Miriam in 4Q365 without further analysis. 

2.     Pentateuch/Reworked Pentateuch or neither of the two?

When Emanuel Tov and Sidnie White Crawford published 4Q365, they acknowledged that the text varied from the MT. Its fragments appear to preserve an edition of the Pentateuch that was closer to the Samaritan Pentateuch rather than the MT. Further, the text contains some omissions and additions in comparison with the MT (as we have just seen). Therefore, Tov and White Crawford did not call the text 4QPentateuch but instead 4QReworked Pentateuchc.[17]

Meanwhile, scholars including Eugene Ulrich, Michael Segal, and James VanderKam suggested that rather than focusing on the differences 4Q365 exhibits vis-à-vis the MT text, the text does not present itself as anything else but a copy of a manuscript that contains Pentateuchal passages. Therefore, they have argued that 4Q365 should not be called 4QRPc but rather 4QPentateuchc.[18] Since the initial publication of 4Q365, Tov has changed his mind regarding the status of the text, and is now ready to call 4Q365 (and other RP texts) “Pentateuch.”[19]  While a growing number of scholars agree with this position, others are not ready to call the text a copy of the “Pentateuch.”[20]  Therefore, the debate over the original status the text once held continues despite the evident problems that juxtaposition “Reworked Pentateuch” and “Pentateuch” imply.[21]   

The research history shows how the scholarly interest around this text has largely focused on the scribes and their intentions and inquiries over the possible status of 4Q365.[22] Further, it is evident how scholars have granted the MT a prime position vis-à-vis all other text forms. This understanding and hierarchy between the texts dramatically influences our understanding of the past. In the case of the Song of Miriam preserved in 4Q365, the song is often referred to as “an expansion,” implicitly assuming that the scribe was familiar with the MT.

3. Shifting the Focus from Scribes to Ancient Audiences

In this short paper I have focused on the materiality of 4Q365 and the version of the Song of Miriam it preserves: the only surviving material copy of the Song of Miriam that dates back to the Second Temple era. I observed that the song is an integral part of the text and presents nothing that would suggest it was an addition to the text. Therefore, rather than analysing it as elaboration of the Song of Miriam from the MT, it should be seen as an early witness to the song tradition i.e., a snapshot to the history of the song.[23]

Scholars have often treated the single verse song of Miriam in Exod 15:21 as an anomaly. Its function has been questioned.[24] The song preserved in 4Q365 encourages us to re-think the significance of different versions of this song. In light of the 4Q365, and in addition the Targumim, the LXX and De Vita Contemplativa it appears that many ancient witnesses to Exodus 15 present the songs in a more balanced way.  For various ancient audiences, the songs of Moses and Miriam were actually not as disproportionate as they seem in light of Exod 15.

When analysing the coexistence of different versions of the Song of Miriam in antiquity and the problematic task of finding the “original” version of the song, it may be helpful to turn to a contemporary example. Leonard Cohen originally wrote the song “Hallelujah” in 1984, but he kept on changing the lyrics during his live performances on tours. Some parts of the song were similar to the 1984 recording, others not. Cohen has famously stated that “many Hallelujahs exist.” This is clearly the case: since Cohen’s own recording, Hallelujah has been covered over 300 times. Significantly, the cover versions do not all contain the same lyrics. Rather, interpreters of the song have made use of numerous different versions. All in all, in the history of Cohen’s Hallelujah it is difficult to pinpoint which version of the song is the “original” version. Is it the time Cohen first sang the song? Is it the first time he performed it in front of an audience? Is it the first recording? The best recording? The version that contains the lyrics Cohen himself preferred? Moreover, for people who have gotten to know the song as a cover version and are unaware of Cohen’s legacy, a cover version may represent the “original version.”[25] At least in the case of Hallelujah, it seems that the only way to understand the question of originality is to see it from the perspective of the audience. This observation finds interesting parallels in in the transmission history of the song of Miriam. Rather than trying to find an original version and compare it with the “corrupted” ones, it is possible that each version of the song was “original” to some audience. 

Hanna Tervanotko is an assistant professor in the department of Religious Studies at McMaster University in Hamilton, Ontario.

[1] Emanuel Tov and Sidnie White Crawford, “Reworked Pentateuchc  (4Q365),” in Qumran Cave 4, VIII: Parabiblical Texts, Part 1, Ed. Harold Attridge et al. Discoveries in the Judean Desert 13. Oxford: Clarendon Press, 1994, 255-318, 269.

[2] Tov and White Crawford, “Reworked Pentateuchc  (4Q365),” 269-271.

[3] Tov and White Crawford, “Reworked Pentateuchc  (4Q365),” 255-261.

[4] Tov and White Crawford, “4QReworked Pentateuch b-e and 365a 4QTemple?” in Qumran Cave 4, VIII: Parabiblical Texts, Part 1, Ed. Harold Attridge et al. Discoveries in the Judean Desert 13. Oxford: Clarendon Press, 1994, 187-196, 198. For the Qumran system, see Emanuel Tov, “The Orthography and Language of the Hebrew Scrolls Found at Quran and the Origin of These Scrolls,” Textus 13 (1986): 31-57; idem, “Hebrew Biblical Manuscripts from the Judean Desert: Their Contribution to Textual Criticism,” JJS 39 (1988): 1-37. 

[5] Tov and White Crawford, “4QReworked Pentateuch b-e and 365a 4QTemple?” 189.

[6] Exodus is preserved in the following manuscripts:1Q2; 2Q2; 2Q3;4Q13; 4Q14; 4Q15; 4Q16; 4Q17; 4Q18; 4Q19; 4Q20; 4Q21; 4Q22 (Paleo Exodus); 7Q1 (LXX Exode).

[7] Tov and White Crawford,” Reworked Pentateuchc,” 269-271; Hanna Tervanotko, Denying Her Voice: The Figure of Miriam in Ancient Jewish Literature, JAJSup 23, Göttingen: Vandenhoeck & Ruprecht, 2016, 150-151.

[8] Tervanotko, Denying Her Voice 150-151; Ariel Feldman, “The Song of Miriam (4Q365 6a ii + 6c 1-7) Revisited,” JBL 132 (2013): 905-911.

[9] Tervanotko, Denying Her Voice, 156-159.

[10] Meanwhile, several scholars argue that the short Song of Miriam is more archaic than the Song of Sea attributed to Moses. See Tervanotko, Denying Her Voice, 44-46.

[11] See, Liddell-Scott Jones Lexicon of Classical Greek. Cf. Takamitsu Muraoka, A Greek-English Lexicon of the Septuagint (Leuven: Peeters, 2009), 248.  For the English translation of the LXX, see New English Translation of the Septuagint available at ccat.sas.upenn.edu/nets/edition.

[12] E.g., Iliad 18.51; 18.606; Odyssey 4.19. Also in the middle voice in Hymni Homerici 27:18. In the usage of the LXX, Exod 15:21 is not the only passage where this verb appears. This verb in the active is also used in the following passages: Exod 32:18; Num 21:17; 1 Sam 18:7; 21:11 (12); 29:5; Jdt 15:14; 16:2; Ps 146 (147):7; Isa 27:2; 1 Macc 9:67; 3 Macc 4:6. As most of these passages refer directly to a context that presents a song or some other poetic activity, it is evident that the Greek verb ἐξάρχω is often used in the context of the opening of a song or poem.

[13] Similarly in Classical Greek. See, LSJ; Muraoka, A Greek – English Lexicon of the Septuagint, 734.

[14] For further references, see e.g., Exod 32:19; Judg 9:27; 1 Sam 29:5; 2 Sam 6:13; 1 Kgs 1:40; Ps 149:3; 150:4; Cant 7:1; Lam 5:15; Isa 5:12; Jdt 3:7; 15:12; 3 Macc: 6:32; 35; 4 Macc 8:4; 13:8; 18:23. For women’s performative roles as musicians see e.g., Carol Meyres, “Of Drums and Damsels: Women’s Performance in Ancient Israel,” BA 54 (1991): 16-27, eadem, “Mother to Muse: An Archaeomusical Study of Women’s Performance in Ancient Israel,” in Recycling Biblical Figures: Papers read at a NOSTER colloquium in Amsterdam, 12-13 May 1997. Ed. Athalya Brenner and Jan Willem van Henten. Studies in Theology and Religion 1. Leiden: Deo Publishing 1999, 50-77.

[15] Translation by C.D. Yonge, The Works of Philo: Compete and Unabridged. Peabody, Mass.: Hendrickson Publishers, 1993.

[16] For Philo’s portrayal of the figure of Miriam and the singing after crossing of the Sea of Reeds in De Agricultura, see, Tervanotko, Denying Her Voice, 241-244.

[17] I have earlier discussed the status of 4Q365 more in detail in Tervanotko, Denying Her Voice, 147-149.

[18] Eugene Ulrich, “Dead Sea Scrolls and the Biblical Text,” in The Dead Sea Scrolls after Fifty Years: A Comprehensive Assessment, Ed. Peter W. Flint and James C. VanderKam, 2 vols, Leiden: Brill, 1998, 1:79-100; James C. VanderKam, “Questions of Canon Viewed through the Dead Sea Scrolls,” in The Canon Debate, Ed. Lee Martin McDonald and James A. Sanders; Peabody, Mass.: Hendrickson, 2002, 91-109; Michael Segal, “4QReworked Pentateuch or 4QPentateuch?” in The Dead Sea Scrolls: Fifty Years After Their Discovery. Ed. Lawrence H. Schiffman et al., Jerusalem: Israel Museum, 2000, 391- 99.

[19] Emanuel Tov, “The Many Forms of Hebrew Scripture: Reflections in Light of the LXX and 4QReworked Pentateuch,” in From Qumran to Aleppo: A Discussion with Emanuel Tov about the Textual history of Jewish Scriptures in Honor of his 65th Birthday, Ed. Armin Lange, Matthias Weigold and József Zsengellér, FRLANT 230, Vandenhoeck & Ruprecht, 2009, 11-28. Meanwhile, Tov’s co-editor of 4Q365, Sidnie White Crawford, Rewriting Scripture in Second Temple Times, The Dead Sea Scrolls and Related Literature, Grand Rapids, Mich.: Eerdmans, 2008, 56-57, calls 4Q365 “scriptural” but does not refer to it as a copy of the Pentateuch.

[20] Cf. Moshe Bernstein, “‘Rewritten Bible’: A Generic Category Which Has Outlived its Usefulness?” Textus 22 (2005): 169-96, who continues to argue that 4Q365 and other RP texts are not copies of “the Pentateuch.”

[21] The problems of the present scholarship that focuses on the status of 4Q365 have been singled out by Molly Zahn, “The Problem of Characterizing the 4QReworked Pentateuch Manuscripts: Bible, Rewritten Bible, None of the Above?” DSD 15 (2008):315-339; eadem, “4QReworked Pentateuch C and the Literary Sources of the Temple Scroll: A New Old Proposal,” DSD 19 (2012): 133-158, 146-149.

[22] I owe this observation to Hugo Lundhaug and Liv Ingeborg Lied, “Studying Snapshots: On Manuscript Culture, Textual Fluidity, and New Philology,” in Snapshots of Evolving Traditions: Jewish and Christian Manuscript Culture, Textual Fluidity, and New Philology, Ed. Liv Ingeborg Lied and Hugo Lundhaug, TU 175, Berlin: De Guyter, 2017, 1-19.

[23] See n. 22.

[24] Tervanotko, Denying Her Voice, 44.

[25] I am indebted to Adele Reinhartz who discussed Hallelujah during the Enoch Nangeroni meeting in Rome in June 2018. Her remarks inspired me to think about how its transmission history relates to the Song of Miriam. Further, Laura Barton, “Hail, Hail Rock’n’ Roll” https://www.theguardian.com/music/2008/dec/19/leonard-cohen-hallelujah-christmas; Ashley Fetters, “How Leonard Cohen’s Hallelujah became everyone’s Hallelujah.” https://www.theatlantic.com/entertainment/archive/2012/12/how-leonard-cohens-hallelujah-became-everybodys-hallelujah/265900/; Laura Stanley, “Hallelujah’” https://www.thecanadianencyclopedia.ca/en/article/hallelujah