Responsive Voices in the Song of the Sea

Over the past century, the Song of the Sea has been scrutinized by noted biblical scholars. Nevertheless, this extensive investigation has not led to clear-cut conclusions about its structure and content. As mentioned by M.L. Brenner, “Still there exists a very great diversity of opinions concerning very basic questions.”¹ This situation results mainly from the inability to resolve three basic problems concerning this poem: the relationship between structure and meaning, the status of verse 19, and the so-called “Song of Miriam.”

The miracle of the Sea is evoked both in Exodus 14 as a prose narrative and in Exodus 15 as a song. In the former, the events are related in their order of occurrence, while no chronological sequence is observed in the latter. The Song of the Sea is obviously not subject to the same literary constraints as a prose narrative, and so its anachronisms and repetitions may be simply considered as poetical devices. Nevertheless, the chronological order of events is respected in other poetical works relating historical events of Ancient Israel (including the Exodus), such as Psalms 78 and 105.

The Status of Verse 19

Most scholars end the Song of the Sea at verse 18.² This opinion is supported by two claims. The first is the short length (four words in Hebrew) and formulaic nature of verse 18 (the Lord will reign for eternity and forever), which recalls the ending of a liturgical ceremony. The second is the extensive length (19 words) of verse 19: For Pharaoh’s horses, chariots and charioteers went into the sea, and the Lord brought back on them the waters of the sea; but the sons of Israel walked on dry land in the midst of the sea. This verse has been interpreted by almost all traditional exegetes, as well as modern scholars, as a prose recapitulation of the song, added during the writing of the book of Exodus.³

That opinion is challenged by the blank interspaces encountered in the Masoretic text of this hymn. Blanks are inserted immediately after the heading of the song (second half of verse 1: Ashirah la-Adonai ki ga’oh ga’ah), signaling the beginning of the poetical section. Blanks are also inserted in verse 19, indicating that it was considered by the Masoretes as an integral part of the Song of the Sea, and not simply a prose recapitulation of its content.⁴

To consider the blank interspaces in v. 19 as an editing error implies that the ending of the song at verse 18, so apparent today, had been dismissed by a scribe and by all his followers. Alternatively, in case verse 19 is part of the Song of the Sea, it remains to be explained why the poet ended this hymn in such a strange fashion. The problem of the status of verse 19 remains unresolved.

The “Song of Miriam”

Verses 20 and 21 are generally considered an independent poetical work called the Song of Miriam. It contains a long prose presentation (v. 20 and beginning of v. 21) and a short song (end of verse 21) parallel to v. 1. Some authors explain this unusually brief song by assuming that the Song of Miriam was a spontaneous and vibrant song of praise performed immediately after the rescue of the Israelites, while the Song of Moses was a well-crafted piece of work composed later on the basis of the former.⁵ However, if the Song of Miriam is really the original source of the Song of Moses, it should have been recorded first in the Bible. Furthermore, it is explicitly mentioned in the heading part of verse 21 that the choir of women conducted by Miriam answers a choir of men (Va-ta’an lahem [masculine form] Miriam) that is missing from the text.⁶

Accounting for such a problematic situation, it is not surprising that some authors have attempted to reintegrate verse 21 in the Song of the Sea.⁷ If so, it remains unclear why the Song of Miriam is preceded by performance instructions (v. 20 and beginning of v. 21). The status of vv. 20-21 remains unresolved.

The New Theoretical Basis

It has been known for a long time that the Song of the Sea was performed antiphonally. This fact is attested both by Philo of Alexandria⁸ and by ancient rabbinic sources.⁹ Four modes of antiphonal singing of this hymn may be suggested concerning the various possibilities of antiphony of the Song of the Sea:

Antiphonal opening: The couple of verses (1, 21) opens the song antiphonally, which is then performed only by the choir of men, from verse 2 to verse 19.

Iterative responsa: Verse 21 is a responsive refrain sung by the choir of women echoing each verse of the Song of the Sea (vv. 1-19) performed by a choir of men.¹⁰

Steady responsa: The song is divided into a strophe and antistrophe, each sung by a half-choir. A dialogue then emerges by the twinning of parallel verses from strophe and antistrophe.¹¹

Cross-responsa: The insertion of performance instructions at the beginning and end of the song points to an antiphonal dialogue between two voices, the first (choir of men) singing the song in the ascendant order of verses and the second (choir of women) answering, verse to verse, by singing the same text in a descendant order of verses. In this mode of performance, all the verses are paired in a chiastic fashion around a central axis. Each couple of verses becomes therefore twice performed, and the order of verse precedence is inverted before and after the central axis.

The first and second hypotheses cannot be justified without ignoring the abnormal position of the text sung by the choir of women and the strange ending of the whole song with verse 19. The third hypothesis cannot account for the singular position of the performance instruction of the second choir at verses 20-21. Cross-responsa is not explicitly attested in biblical poetry, though its occurrence has already been identified.¹² It is, however, the single hypothesis accounting for all the editorial singularities of the Song of the Sea, once we assume that the song encompasses the whole text (vv. 1-21), except for the performance instructions (first half of v. 1, v. 20, and first half of v. 21).

The Cross-Responsa Setting of the Song of the Sea

To set the Song of the Sea in cross-responsa fashion, the nature of the fragments (antiphonal units) to be paired should be determined first. No indication is provided in the Song of the Sea concerning the size of the antiphonal units. Thus, according to the criterion of parsimony, we have first assumed that all the verses contain the same number of antiphonal units. Verses 12 and 18, the shortest verses of the Song of the Sea, contain only four words each. We therefore deduced that the verses have a maximum of four antiphonal units. The three verses containing repetitions (vv. 6, 11, and 16) may help to determine their number, assuming that repeated phrases of praise of God do not belong to the same antiphonal unit.

A high level of symmetry is observed in verse 6 (Your right hand, Lord, is majestic in power; Your right hand, Lord, has shattered the enemy), providing a natural division of this verse into two antiphonal units of equal length, each one starting with yeminekha (“Your right hand”), or into four antiphonal units by subdividing each section starting with yeminekha into two segments. The same reasoning applied to verses 11 and 16 leads to the conclusion that each verse of the Song of the Sea should be divided into four antiphonal units, and the pausal indications of the Masoretic cantillation provide a reliable, literarily independent criterion for identifying exactly where those divisions fall.^{13 14 15}

Once the antiphonal units are individualized, they may be mixed in cross-responsa fashion, pairing verse 1 with 21, verse 2 with 19, verse 3 with 18, and so on inward to the central pairing of verses 10 and 11 — after which the same couples recombine in inverse order of precedence (11/10, 12/9, 13/8 … 19/2, 21/1) to bring the performance to its close. The cross-responsa form does not show any dissonance between the pairs of antiphonal units; if anything, its reading is no more abrupt, awkward, or incongruous than the linear reading of the poem, once literally translated. Rather, echo patterns are observed in this composite setting, not only in the four couples issued from the pairing of verses 1 and 21 (as expected), but also in the pairs (3, 18a), (8, 13a), and (8, 13d). The emergence of a coherent text from mixing distant fragments of verses is not a trivial feature; it is indicative of the accuracy of the cross-responsa hypothesis, a conclusion confirmed by investigation of the structural properties of this pattern of verse pairing.

Structural Properties of the Cross-Responsa Setting

The Turn to the Lord. The Song of the Sea is a hymn of praise, so special attention should be devoted to the way the poet turns to the Lord. God is referred to in the second person in nine verses (vv. 6-8, 10-13, 16, and 17) and in the third person in six verses (vv. 1-3 and 18, 19, 21). In the five other verses (vv. 4, 5, 9, 14, and 15), the poet does not explicitly turn to the Lord. Such a distribution remains meaningless in the linear context of reading. In the cross-responsa pattern, however, the verses not turning to the Lord are never paired together, and in every instance where the two verses paired by cross-responsa turn to the Lord, they always do so in the same grammatical person. It appears, therefore, that the poet turns to the Lord in every couple of verses paired in cross-responsa fashion, and that a grammatical dissonance is never engendered by such a pairing.

The Composite Opening/Closure. The three pairs of verses performed at the beginning (1/21 to 3/18) function as a “composite aperture,” and thereafter as a “composite closure” once performed in inverted order of precedence (18/3 to 21/1). The composite opening starts with the strict echo-pattern pair (1/21), a prologue announcing the intent of the two choirs to sing antiphonally about the miracle performed by the Lord. The second pair (2/19) associates the two most singular verses of the Song of the Sea — verse 19 is so unlike the rest of the song that it was excluded by most traditional exegetes and modern scholars alike, and verse 2 has likewise been isolated from the whole by some — generating a dialogue between a voice praising the Lord for personal reasons (my God, the God of my fathers, v. 2) and a voice praising Him for collective historical reasons (the miracle of the Sea, v. 19). Their order of precedence is not incidental: at the opening, the “subjective voice” (v. 2) leads, inviting the “objective voice” (v. 19) to celebrate the Lord; at the closure, after the events have been detailed, the objective voice becomes the leader, stimulating a new dimension of faith for the subjective voice. The third couple (3, 18) combines two very short verses of similar content whose brevity separates the opening/closing section from the composite development that follows.

The Composite Development. The composite development is built by the seven pairs of verses performed first from 4/17 to 10/11, and immediately after from 11/10 to 17/4, with the 10/11 pairing — repeated twice at the middle of the performance — functioning as the central symmetry axis. Six composite verses (from 4/17 to 9/12) surround this axis in a mirrored pattern: the first two (4/17 and 5/16) recall the destruction of the Egyptians in the sea, the two central ones (6/15 and 7/14) describe the terror of the Canaanites, and the last two (8/13 and 9/12) relate, once again, the rescue of the Israelites in the sea — a symmetrical arrangement fully appropriate to the inversion of precedence of the composite verses in the second half of the performance.

The above considerations reveal the emergence of a highly regular structure in the cross-responsa setting: an opening in which the two choirs invite one another to sing and commemorate the miracle of the Exodus, framing a proclamation of faith (1/21, 2/19, 3/18); a composite development that moves from the destruction of the Egyptians through the terror of the Canaanites to the rescue of the Israelites and the Lord’s acclamation, and back again (4/17 through 17/4, with 10/11 at the center); and a closure that mirrors the opening in reverse (18/3, 19/2, 21/1). This recapitulative scheme reveals that all the composite verses of the Song of the Sea are symmetrically organized around axes once the song is set in cross-responsa fashion — a level of regularity considerably higher than any strophic structure inherent in the linear reading of the poem, confirming the validity of the cross-responsa hypothesis.

Conclusion

The cross-responsa hypothesis enables us to resolve the open questions about the Song of the Sea without introducing any modification of the original Masoretic source. It generates a coherent composite text through the use of pausal indications from the Masoretic cantillation and the blank interspaces inserted in the Masoretic Text. Furthermore, the organization of the composite text according to a central and two secondary axes cannot be considered a feature emerging incidentally from a poem conceived to be performed in a linear fashion; rather, such a highly organized composite structure should be regarded as the strongest argument indicating that the Song of the Sea was originally conceived to be performed antiphonally.

Antiphony — a mode of performance in which two voices respond to one another — is identified in ancient Hebrew poetry by echo-patterns between the dialoguing voices. In complex antiphony, where two choirs sing distinct texts to be mixed during the performance, two extensions of the basic echo-pattern are observed: the antithetic pattern, and a complementary pattern in which the two statements are intimately related, one overextending the meaning of the other. Both cases are clearly illustrated here in the cross-responsa form of the Song of the Sea. Until now, complex antiphony has garnered little attention, with the result that biblical poetry has been systematically approached in a linear fashion. However, complex antiphony seems to be a well-accomplished genre in biblical poetry, revealed both by the mastery of poetical craft required to create the composite structure observed in the Song of the Sea and by the new meanings issued from the mixing of statements from the two choirs. These findings reveal that the problems encountered in any attempt to understand the meaning of a biblical psalm read linearly do not necessarily reflect its mutilation, erosion, or misleading re-editing, as is so readily assumed by modern scholars; they may rather reflect the complex antiphonal dimension of an original and undamaged poetical work.