Did Moses Sing? Perspectives on Deuteronomy 32


Based on current knowledge of how scribes worked in the ancient Near East…[Nilsen] proposes that there were groups of scribes in Jerusalem that produced literature in the late sixth-early fifth century or so, amongst them one Deuteronomistic group and one Isaianic group. Texts could cross boundaries between the groups (cf. e.g. Isa 36-39 and 2 Kings 19-20). The poem in Deut 32 is so similar to Isaiah 1; 56-66 that it is probable that the Isaianic group was its author. The poem then crossed the boundary of scribal groups, and was inserted into what we now call Deuteronomy 32, where it took on the role of “the song of Moses.”

See Also: The Origins of Deuteronomy 32: Intertextuality, Memory, Identity (Peter Lang Inc., International Academic Publishers, 2018).

By Tina Dykesteen Nilsen
VID Specialized University, Stavanger, Norway
January 2019



Many refer to the poem in Deuteronomy 32:1-43 as the “song of Moses” since the narrative framework states that Moses (and Joshua, compare 31:30 with 32:44) spoke this song to the Israelites. The poem is remarkable for many reasons. First, Moses very rarely performs a song (see also Exod 15:1-18). Second, the poem is unusually rich in literary techniques, making it of the highest poetic quality. Third, the poem was written on a separate scroll in Qumran, which testifies to its special significance in that community. Fourth, as far as we know, it was the very first Hebrew poem ever that was written colographically (i.e. with the layout of a poem, rather than a continuous text). Fifth, there are many riddles attached to this text. Here are some of them: We do not really know what the text of the poem is. The poem defies any attempt to be classified by genre. It is hard to understand what the poem is really about; it is both clear and vague at the same time. Dating the poem is inherently difficult, as is the task of identifying its author(s), and making sense of what its function is. Ask ten different experts, and you will get at least ten different answers to any question you may pose. In what follows, we will look at some questions and some answers.



We have no way of knowing whether the poem existed orally before it was written down. Whether it first existed orally or was a written composition from the beginning, one fact remains: we do not have any “original” manuscript on which the poem was written. We only have copies of copies of copies. In some places, the text on these copies differ from each other. Sometimes, the differences, called variants, are very small and do not really matter, but in some places, the differences are huge, because they alter the meaning of the text. The method of textual criticism tries to establish which variant reading is the best one, and while scholars sometimes agree on the result, sometimes they do not. Deuteronomy 32 contains many variant readings, and verses 8 and 43 offer the most interesting examples.

Verse 8 speaks of primeval times when Elyon, the highest god, created the peoples of the earth. It says that Elyon established boundaries of peoples, “according to the number of,” but the number of what? Some manuscripts say it is “the children of God” or “the children of gods” (the Hebrew can be translated both ways); others say “the children of Israel,” or “the angels of God,” or “the children of the mighty ones,” or “judges” or “rulers.” There are many possible explanations to what has happened here, but most scholars agree that “the children of God/gods” probably is the best reading, and that copyists changed the words to make the verse less polytheistic.

The most important variant readings for v. 43 are in three different manuscript traditions: the Masoretic text, Qumran and the Septuagint. Both the Masoretic text and Qumran are quite short, while they differ from each other in some other aspects. The Septuagint is longer, as it contains bits from both of the other two. Is the longer reading better, or the shorter one? In addition to the question of length, there is also another problem at the beginning of v. 43. The Masoretic text commands the nations to rejoice, but the Qumran text commands the heavens to rejoice and “all gods” to prostrate themselves before God. The Septuagint combines both, but instead of “all gods,” it reads “all children of God.” Perhaps another example of fear of polytheism, leading to scribal changes of the text? The matter is complicated, and while we do not have the space to explore it here, it at least serves as an example of a textual difficulty in the poem. It is also an example of a difficulty that scholars disagree on how to solve.



What kind of poem is Deut 32:1-43? This is the question of genre, and scholars disagree on this issue, too. Before the 1950ies, it was quite common to regard the poem as Moses’ farewell-speech, belonging to the genre of the dying patriarch. Undoubtedly, as the poem stands now, it is one of the last speeches of Moses. However, the poem might have existed independently of its present narrative framework, and there is nothing in the poem itself to suggest that it is a farewell-speech. A popular opinion since Wright’s article in 1962, is that the poem is a rib, that is, it belongs to a prophetic genre that imitates court proceedings. However, is Yahweh prosecutor and Israel defendant, or vice versa, or to the roles change throughout the poem? What about those parts of the poem which do not fit court proceedings? If a rib is embedded in the poem, it has been creatively reshaped. Parts of the poem belong to other prophetic genres, such as rhetorical questions (vv. 6.30.34-35), oracles of salvation (vv. 36.43), taunts (vv. 37-38), etc. There are also didactic elements, such as the invocation at the beginning of the poem (vv. 1-3) and the emphasis on understanding or lack thereof throughout the poem. However, the poem also resembles hymnic-prophetic-didactical psalms (e.g. Pss 78; 105). So perhaps it is a real song after all, but drawing on elements from many different genres?



Only some examples of literary techniques in the poem may be given here. One of these is parallelisms, such as e.g. when two consecutive lines say the same thing but in different ways (e.g. vv. 2.7). Such parallelism are ubiquitous in the poem.

Another technique is the use of imagery, whether by metaphors, similes or other. To mention but a few: speech is like precipitation (v. 2); Yahweh is like a vulture (v. 11); Yahweh is a rock, as are the gods (vv.; Yahweh has sold his people (v. 30). Different variations on the imagery of Yahweh as parent (both mother and father) and the people as children are also found (vv. 5-6.10-11.13.18-20).

A large array of rhetorical techniques are used; above I mentioned, for example, rhetorical questions. We also find the rhetoric of entrapment, whereby the audience may think the text speaks of the punishment of their enemies, but then they realise the punishment will befall themselves. The poet conveys this through the blurring of the identity of the speaker (vv. 28-33).

A final example is the poem’s linguistic characteristics: if the poem is dated late (e.g. sixth-fifth century BCE), the poem shows archaising features, i.e. elements that imitate an earlier period. Examples are mythical elements in the content (e.g. polytheistic tendencies in vv. 8.43, mythical beings in vv. 17.23-24.33) and in the language itself (vocabulary, unusual suffixes, spellings, etc.).



The poem is a historical resume focussing on the relationship between Yahweh and Jacob (i.e., Israel). It begins with an overture (vv. 1-7), and then moves chronologically, from Yahweh’s creation of and care for Jacob in the desert and his way into an abundant land (vv. 8-15a), via Jacob’s apostasy and Yahweh’s decision to punish him (vv. 15a-25), through Yahweh’s change of mind (vv. 26-27), to future punishment of Jacob’s enemies and his restoration (vv. 28-43). Despite this clear structure, much remains vague and open to different interpretations. For example, who are the enemies? What does the punishment and the restoration consist of? Are any specific historical events intended?

            Through this larger plot, several subthemes emerge. One concerns divinities and apostasy: the poem seems to presuppose the existence of several deities (vv. 16-17.31.37-38.43), while upholding that Yahweh is the only god with any real power, the only one caring for Jacob, and the only one Jacob should venerate (vv. 12.30.37-39.43). Another example is that of food and drink; abundance testifies to Yahweh’s care (vv. 13-14), while Jacob’s gluttony symbolises apostasy (v. 15; perhaps also vv. 32-33). Yahweh’s potential punishment of Jacob involves the destruction of food production (v. 22; perhaps also vv. 32-33), and the enemies are “eaten” by Yahweh’s weapons (this latter being a common expression in biblical Hebrew; vv. 24.42). A third example may be that of remembering and forgetting, understanding and foolishness which permeates the poem and are used in many different contexts. Jacob’s foolishness and lack of remembrance causes it not to understand the course of the past and Yahweh’s role in it (vv. 6-7.28-30). Jacob has forgotten Yahweh and follows other gods previously not known (vv. 17-18). Yahweh’s potential judgment on Jacob includes the threat that Jacob will not be remembered, whether by people or by Yahweh self (v. 26). The enemies, too, are foolish, not understanding Yahweh’s control of history (vv. 21.27).



With the rise of historical-criticism, scholars raised questions concerning mosaic authorship of the Pentateuch, including the authorship of Deut 32. Scholars today agree that Moses did not write the poem in Deut 32, but they disagree who wrote it and when. In trying to date the poem, scholars use criteria such as linguistic features, historical allusions, form, redaction and similarities to other texts. Frequently, scholars using the same criteria conclude differently. Suggestions range from the eleventh to the fourth century BCE, or any period in between. Perhaps the most influential study since 1996 has been Sanders, who concluded that the poem dates from before the exilic period (i.e., before the sixth century BCE), and that it may even be pre-monarchical (before the eleventh century). He bases his conclusions on an analysis of linguistic features, literary context and similarities to other literature. However, in the most recent contribution to the debate, Nilsen (2018) concludes differently. She, too, compares the poem with other ancient, particularly biblical, literature, but focusses the comparison primarily on imagery, lexemes and phrases, while discussing also linguistic features, forms, themes and biblical parallels. She also elaborates on questions of social memory and social identity formation through texts. Nilsen concludes that the text closest to Deut 32 is Isaiah 1; 34-35; 40-66, a result finding support also in the work of other scholars (e.g. Kim 2004; Keiser 2005). Based on current knowledge of how scribes worked in the ancient Near East, she proposes that there were groups of scribes in Jerusalem that produced literature in the late sixth-early fifth century or so, amongst them one Deuteronomistic group and one Isaianic group. Texts could cross boundaries between the groups (cf. e.g. Isa 36-39 and 2 Kings 19-20). The poem in Deut 32 is so similar to Isaiah 1; 56-66 that it is probable that the Isaianic group was its author. The poem then crossed the boundary of scribal groups, and was inserted into what we now call Deuteronomy 32, where it took on the role of “the song of Moses.”



The function of the poem in isolation from its textual context, i.e. its function before its insertion into Deuteronomy, depends on its dating and/or genre. Suggestions include use of the poem as a war-song at the time of the judges (Cassuto, 1973), or a song used in an annual covenant renewal festival at the time of Samuel (Wiebe, 1989), or as a liturgical composition a choir performed with different voices (Thiessen, 2004). Nilsen (2018) looks at how texts create rivalling social identities of “Israel” as Yahweh’s people through the encoding of social memories. Nilsen argues that the poem in Deut 32, just like Isa 56-66, is purposefully vague in order to negotiate between different positions in the early fifth century BCE.

            Once the poem was inserted into Deuteronomy, it may have kept this original function: the poem’s dense historical resume created a certain identity for the people. However, the insertion into Deuteronomy also gave the poem new functions. Being placed in Moses’ mouth, it gained the function of becoming his last words, giving Moses’ life a fitting end. When people remembered the song, they would also remember Moses and the events connected to his character. In addition, the poem would remind the audience of certain main-points in Deuteronomy’s teachings (an extra-territorial, monarchy-less nation with warnings against apostasy). As well as rounding off Deuteronomy, it would stand as an introduction to the rest of the Deuteronomistic History, from Joshua to 2 Kings.



As a literary character, Moses did sing, and one of his songs was Deut 32:1-43. In contrast, we have no reliable data that can inform us whether the historical Moses ever sang anything, or whether he even existed. However, we can be sure that if he existed, he did not sing the poem in Deut 32. That poem was probably composed several centuries after the period in which Moses is supposed to have lived, and it only became “the song of Moses” when it was inserted into its present context in Deuteronomy. However, that does not diminish the poem’s qualities, whether literary or theological.


A vast amount of literature has been written on Deut 32:1-43. A bibliography may be found in Nilsen (2018). Literature directly referred to in this article is as follows:

Cassuto, Umberto (1973). The Song of Moses (Deuteronomy Chapter xxxii 1–43) (Israel Abrahams, Trans.). In Umberto Cassuto (Ed.), Biblical and Oriental Studies (Vol. 1, pp. 41–46). Jerusalem: The Magnes Press. (Reprinted from: Atti del XIX Congresso Internazionale degli Orientalisti, Roma: 1938, pp. 480–484).

Keiser, Thomas A. (2005). The Song of Moses as a Basis for Isaiah's Prophecy. Vetus Testamentum, 55, 486–500.

Kim, Hyun Chul Paul. (2004). The Song of Moses (Deuteronomy 32:1–43) in Isaiah 40–55. In Ellens J.H., D.L: Ellens, R.P. Knierim, & I. Kalimi (Eds.), God's Word for Our World, vol. I: Biblical Studies in Honor of Simon John De Vries (pp. 147–171). London: T&T Clark.

Nilsen, Tina Dykesteen (2018). The Origins of Deuteronomy 32: Intertextuality, Memory, Identity. New York: Peter Lang.

Sanders, Paul (1996). The Provenance of Deuteronomy 32. Leiden: Brill.

Thiessen, Matthew (2004). The Form and Function of the Song of Moses (Deuteronomy 32:1–43). Journal of Biblical Literature, 123, 401–424.

Wiebe, John M. (1989). The Form, Setting and Meaning of the Song of Moses. Studia biblica et theologica, 17, 119–163.

Wright, Ernest G. (1962). The Lawsuit of God: A Form-Critical Study of Deuteronomy 32. In Bernhard W. Anderson & Walter Harrelson (Eds.), Israel's Prophetic Heritage: Essays in Honour of James Muilenberg (pp. 26–67). New York: Harper & Brothers.

Article Comments

Submitted by Robert Rezetko on Wed, 01/16/2019 - 11:51


Nilsen’s dating of Deuteronomy 32 to the late sixth-early fifth century or so provides an interesting contrast to the work of Hendel and Joosten (cf. preceding article) who believe the “archaic” language of the chapter would date it to a long time before the sixth century (pp. 45-46, 128, 130 in their book). I wonder how, in the framework of their "consilience" model that incorporates language, text, and culture/history, they would handle the different criteria that Nilsen mentions?

Thanks for your comment! In this short online article, there is not place to discuss the linguistic criteria, but I do so in my book (also referring to your work). The "archaic" elements in Deut 32 are also found in younger texts in the Book of Isaiah, where they are definitely archaizing features. In short: linguistic features cannot be used to date this text.

Submitted by Martin Hughes on Mon, 01/28/2019 - 14:10


May I speculate irresponsibly? Could the argument of the poem be that now, after serious and horrible disturbances, all national gods are back in their proper place but that Yahweh has performed such an exceptional task both in reconciling formerly apostate worshippers and in inflicting vengeance on their enemies that all other gods must celebrate and defer to him? This would make him the supreme ruler for practical purposes, Elyon having no more to do. The role of assigning lesser deities to different nations has passed to Yahoo in Deuteronomy 4.
The idea of ‘every god in his place’ seems to share the ideological of the Cyrus Cylinder. The idea of the supreme God being above and beyond everything and being represented in the universe by a more involved and active force looks forward to the ideology of Timaeus.

Add new comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.