THE RECOVERY OF POETIC FRAGMENTS FROM THE ...

/ £ ! 97/2 (1978) 161-73

THE RECOVERY OF POETIC FRAGMENTS FROM THE PENTATEUCHAL PRIESTLY SOURCE

JOHN S. KSELMAN

ST PATRICK'S SEMINARY, MENLO PARK, CA 94025

THERE is a growing conviction among a number of OT scholars that behind the narrative JE sources of the pentateuch lies the national epic of

ancient Israel. In this hypothesis, the epic was originally in the form of poetry, possibly orally composed by Israelite bards' for performance at covenant-renewal feasts in pre-monarchic Israel. In his Canaanite Myth and Hebrew Epic, Cross states this position neatly:

In the case of the Epic materials, however, we are inclined to reconstruct a long and rich poetic epic of the era of the league, underlying JE, and to take the prose epic variants (with their surviving poetic fragments)... as truncated and secondary derivatives. In any case, we possess early poetic epics from old Canaan, from ancient Mesopotamia, and Homeric Greece, and to find the same phenomenon in Israel would not be surprising.2

Evidence for this hypothesis includes the "surviving poetic fragments" to which Cross alludes, poetry lying close enough to the surface of the prose epic to be visible and, at least in part, recoverable.

An example of such epic poetry will make this point clear. Gen 28:17, probably to be assigned to E, seems to contain a tricolon of poetry:

mh nwr° hmqwm hzh Dyn ζ h ky °m byt °lhym wzh scr Mmym

The parallelism of the verse, and specifically the presence of the pair

'The seminal work here is A. B. Lord, The Singer of Tales (New York: Atheneum, 1965). 2F. M. Cross, Canaanite Myth and Hebrew Epic: Essays in the History of the Religion of

Israel (Cambridge: Harvard University, 1973) 124. Cf. also his remarks on 293: "We prefer to speak of J and E as variant forms in prose of an older, largely poetic Epic cycle of the era of the Judges. . . . No doubt the Epic cycle was originally composed orally and was utilized in the cult of covenant-renewal festivals of the league." On the source criticism that would distinguish the J, E, and Ρ strands of the pentateuch, cf. note 4 below.

©1978, by the Society of Biblical Literature

162 JOURNAL OF BIBLICAL LITERATURE

byt / I scr, formulaic in Hebrew and Ugaritic,3 point to the poetic nature of this tricolon.

If the presence of poetry in JE occasions no surprise, the same cannot be said of the pentateuchal tradition that comes from the priestly tradents,4 a tradition which has been described as "the literary product of a writer who was quite without 'Sinn für Poesie'."5 New data may require a change in this estimation, however: (a) in Yahweh and the Gods of Canaan, Albright pointed out the poetic bicolon in Exod 14:3, from P;6 (b) Porten and Rappaport have demonstrated in a recent study the presence of poetic structure in Gen 1:22 and 9:7;7 (c) the present author previously discussed the bicolon of Ρ poetry in Gen 7:11 ;8 and (d) in Canaanite Myth and Hebrew Epic, Cross has shown that Exod 40:34 is a fragment of poetry recoverable from the Ρ source.9 It is the purpose of this paper to continue this line of research and to offer for consideration other instances of poetry recoverable from the priestly tradition.

I

To test the hypothesis of possible poetic fragments in P, the first obvious source is the Ρ creation account. On the basis of such evidence as the refrain in Gen 1:4, 10, 12, 18, 21, 25 (and notice its occurrence in poetry in Gen 49:15: wyrD mnhh ky twb / / wDt hDrs ky ncmh) and such archaisms as the waw

3For the pair byt / / scr cf. M. Dahood and T. Penar, "Ugaritic-Hebrew Parallel Pairs," Ras Shamra Parallels /(ed. L. R. Fisher; AnOr 49; Rome: Pontifical Biblical Institute, 1972) 158 (No. 137). On the basis of the etiology in Gen 28:19 (wyqrD °t sm hmqwm hhP byt °l), a possible emendation would be to read byt Jl in ν 17 as well, in place of byt Dlhym. This would produce a meter of 6:6 in the two lines that employ the formulaic pair. If byt °lhym is secondary, could this be an instance of the levelling through of a more common form for a rare one, something one might expect in poetry transmitted as prose?

4The author is aware that the literary source criticism of the pentateuch is presently being challenged, and that many of the traditional assumptions connected with source criticism are under reconsideration, as in the work of John Van Seters, Abraham in History and Tradition (New Haven/ London: Yale University, 1975). On the particular issue of concern in this paper (the existence of Ρ material), however, Van Seters can say: "With respect to literary analysis of the Pentateuch perhaps the greatest agreement has been reached in the case of P, the 'Priestly' writer's contribution to the final corpus. . . " (p. 125). So too another critic of source criticism, Rolf Rendtorff, in his Das überlieferungsgeschichtliche Problem des Pentateuch (BZAW 147; Berlin/New York: Walter de Gruyter, 1977) gives at least restrained approval for the idea of a priestly stratum. The reader might consult the symposium on Rendtorff s work in Journal for the Study of the Old Testament 3 (1977). On the positive side, the Ρ material is rigorously and ably studied by R. Polzin, Late Biblical Hebrew: Toward a Typology of Biblical Hebrew Prose (HS M 12; Missoula: Scholars Press, 1976), esp. chap. 3 ("Significant Linguistic Features of the Priestly Document"). On the basis of the above review, I believe that the existence of a priestly stratum can be safely accepted.

5S. McEvenue, The Narrative Style of the Priestly Writer (AnBib 50; Rome: Pontifical Biblical Institute, 1971) 1.

6W. F. Albright, Yahweh and the Gods of Canaan (London: Athlone Press, 1968) 33. 7B. Porten and U. Rappaport, "Poetic Structure in Genesis IX 7," VT2Ì (1971) 363-69. 8J. S. Kselman, "A Note on Gen 7:11," CBQ 35 (1973) 491-93. 9Cross, Canaanite Myth, 167-68.

KSELMAN: POETIC FRAGMENTS IN THE PRIESTLY SOURCE 163

compaginis of hytw Drs (1:24) and the enclitic mem of mqwm (miqwë-m for MT mäqom: 1:9), and the adverbial sense of ky (exclamatory or interrogative "how") in the refrain, Albright described Genesis 1 as "a prose paraphrase, possibly a condensation of an older poetic text."10

Further study has provided another productive tool for isolating instances of poetry in a prose source: the phenomenon of parallel pairs, many common to both Ugaritic and Hebrew poetic diction. The application of this tool to some narrative material from Ρ can reveal a number of cases of possible bicola and tricóla of poetry. I realize, of course, that these pairs could appear in a prose source. But I believe the case is cumulative; that is, the number of such formulaic pairs (some twenty-five in the material analyzed below) along with other poetic devices (breakup of stereotype phrases, chiasmus, inclusion, repetition of the same word or root in parallel cola, etc.) at least makes the possibility of poetry in Ρ worthy of further study. This paper, then, is of the nature of a probe, the collection of data, rather than a larger hypothesis about the nature of the Ρ material if this analysis has merit. Finally, on the vexed question of meter, we shall follow those recent studies that have suggested syllable count, rather than stress, as a way of bringing more objectivity to prosody.11 On the basis, then, of the presence of formulaic parallel pairs, seven possible examples of poetry emerge from Genesis 1.

Gen 1:2 wh^rs hyth thw wbhw whsk cl pny thwm wrwh °lhym mrhpt cl pny hmym

The noun series °rs / / thwm / / mym, built out of the formulaic pairs Drs 11 thwm and mym / / thwm, identifies this verse as a tricolon of poetry. Ugaritic poetry provides us with a superb example of the pair ars / / thmt, arranged chiastically as in Gen 1:2: tant smm cm ars / / thmt cmn kbkbm.12

The parallel expressions H pny thwm and clpny hmym provide an instance of the second parallel pair (thwm // mym in Ezek 26:19 and Ps 104:6; mym // iAwm in Ezek 31:4, Hab 3:10, Jon 2:6, Pss 33:7,77:17, Job38:30;and note the parallelism of mym [ / / nzlym] / / thmt collocated with rwh in Exod 15:8).

10W. F. Albright, "The Refrain 'And God Saw Ki Tob' in Genesis," Mélanges bibliques rédigés en l'honneur de André Robert (Paris: Bloud & Gay, 1955) 22-26. Albright alludes in this study to Cassuto's work on the original poetic prototype of Genesis 1. Cf. also Albright's remarks in From the Stone Age to Christianity (Garden City: Doubleday, 1957) 66-67.

nCf. D. K. Stuart, Studies in Early Hebrew Meter (HSM 13; Missoula: Scholars Press, 1976) for an important study of prosody. On pp. 24-28 Stuart gives a convenient list of the features of the vocalization of Hebrew poetry. An earlier important study is that of R. C. Culley, "Metrical Analysis of Classical Hebrew Poetry," in Essays on the Ancient Semitic World (ed. J. W. Wevers and D. B. Redford; Toronto: University of Toronto, 1970) 12-28. With the exception of the monosyllabic character of the segholate, we have followed Culley's more cautious procedure, given the tentative nature of prosody.

12Dahood and Penar, "Parallel Pairs," 127 (No. 72); R. Whitaker, A Concordance of the Ugaritic Literature (Cambridge: Harvard University, 1972) 613.


Other possible indications of poetry here might include the assonance of the rhyming pair thw wbhwn and of rwh. . . mrhpt, and the dislegomenonc/ pny thwm occurring here and in Prov 8:27, part of an archaic poetic cosmogony patterned after Canaanite-Phoenician models.14 Finally, the recognition of the pair Drs / / thwm reveals an instance of inclusion framing the six days of creation, where cl pny thwm before the first day (1:2) is matched by cl pny kl hDrs after the sixth (1:29).15 The syllable count for the three lines is 11:7:13.

Gen 1:5 wyqr° Dlhym Pwr ywm wlhslc qr° lylh

This bicolon of poetry employs two common formulaic pairs: Dwr / / hsk (Isa 9:1, 45:7, Ezek 32:8, Job 38:19) and ywm / / lylh (Isa 4:5, Jer 31:35, Pss 19:3, 22:3, 42:9, 88:2, 121:6, Job 3:3, 5:14). And although the sequence wyqrD / / qrD follows the normal rules of Hebrew prose syntax, might one see here the poetic device of paralleling the same verbal root in its yqtl and qtl forms? Apparently first observed by Cassuto, this stylistic feature has been thoroughly discussed by Held and Gevirtz.16 The chiastic ABC / / BAC pattern is also found in Ugaritic poetry, as the following example (2 Aqht [CT A 17] VI:28-29) shows: assprk cm bcl snt / / cm bn il tspr yrljm. By syllable count, the meter is 9:7, an acceptable range for poetry transmitted in prose.

Gen 1:10 wyqr° °lhym lybSh °rs wlmqwh hmym qr° ymym

13The chiastic thw wbhw / / hik is echoed in Jer 4:23 (thw wbhw / / °yn °wrm), a poem modelled on Genesis 1; cf. M. Fishbane, "Jer IV 23-26 and Job III 3-13: A Recovered Use of the Creation Pattern," VT2Ì (1971) 151-67.

J4W. F. Albright, "Some Canaanite-Phoenician Sources of Hebrew Wisdom," Wisdom in Israel and in the Ancient Near East (éd. M. Noth and D. Winton Thomas; VTSup 3; Leiden: Brill, 1960) 1 -15; G. Landes, "Creation Tradition in Proverbs 8:22-31 and Genesis l,nA Light unto My Path: Old Testament Studies in Honor of Jacob M. Myers (ed. Η. Ν. Bream, R. D. Heim, C. Α. Moore; Philadelphia: Temple University, 1974) 279-93.

,5For this type of inclusion or envelope construction, cf. D. N. Freedman's "Prolegomenon" to G. B. Gray, The Forms of Hebrew Poetry (New York: KTAV, 1972) xxxvi-xxxvii. This inclusion is set within another inclusion that frames the Ρ creation account, the sequence of òrD, ¿mym, °rs in Gen 1:1 and its inversion in ¿mym, Drs, br° of 2:4a. Talmon has described this stylistic feature as "inclusion by distant inverted parallelism" ("The Textual Study of the Bible—A New Outlook," Qumran and the History of the Biblical Text [ed. F. M. Cross and S. Talmon; Cambridge: Harvard University, 1975] 364).

,6M. Held, "The yqtl-qtl (qtl-yqtl) Sequence of Identical Verbs in Biblical Hebrew and Ugaritic," Studies and Essays in Honor of Abraham Neumann (éd. M. Ben-Horim; Leiden: Brill, 1961 ) 281 -90; S. Gevirtz, "Evidence of Conjugational Variation in the Parallelization of Selfsame Verbs in the Amarna Letters," JNES 32 (1972) 99-104.


This bicolon has the same ABC / / BAC chiasmus and the same yqtl 11 qtl verbal sequence as in Gen 1:5, as well as the Ugaritic-Hebrew pair Drs 11 ymym.11 The syllabic meter is 11:10.

Gen 1:16 wycs °lhym °t iny hmDrt hgdlym °t hm°wr hgdl Immslt hywm w°t hm°wr hqtn Immslt hlylh

Besides the pair ywm / / lylh discussed above, this tricolon includes a second pair, gdwl / / qtn, found also in Amos 6:11 and 8:5 (Ihqtyn / / Ihgdyl). Besides the pair, the repeated Immslt may be an indication of poetry. The phrase w^t hkwkbym in the third colon disturbs the clear parallelism and looks like the kind of addition one would expect in prose transmission. The syllable count for the verse is 16:12:14.

Gen 1:20 yfrsw hmym frs npihyh wcwp ycwpp cl h°rs cl pny rqyc hSmym

The three members of the tricolon recall the tripartite division of the cosmos, a common motif in Hebrew poetry.18 As to formulaic language, the common word-pair Drs / / smym19 can be discerned in the second and third lines, while mym in the first line and imym in the third recall Smy hsmym / / hmym in Ps 148:4. The closest parallel is Amos 9:6, where the full sequence mym / / Drs / / ¿mym is found in reverse in a poetic source: hbwnh bsmym mclwtw / / wDgdtw cl Drs ysdh / / hqrD Imy hym / / wySpkm clpny hDrs.20 Ps 136:5-6 contains a similar sequence (hfmym / / hDrs cl mym). Finally, the collocations of the pair Drs / / ¿mym with mym in Judg 5:4, Pss 33:5-7 and 73:9-10, and Job 28:3-4, and the similar sequence in Ps 96:11 (smym / / Drs / / ym) and in Pss 69:35,135:6, and 146:6 (imym w°rs \ / ym) all support the conclusion that the series of three nouns in Gen 1:20 reflects Hebrew poetic diction. The syllable count is 9:8:8.

Gen 1:27 wybr° °lhym °t hDdm bslmw bslm °lhym br° °tw zkr wnqbh br° °tm

,7Dahood and Penar, "Parallel Pairs," 122-23 (No. 64). The parallelism is not exact, since the Ugaritic-Hebrew pair is commonly °rs / / ym (singular). This has led H. J. van Dijk, Ezekiel's Prophecy on Tyre (BibOr 20; Rome: Pontifical Biblical Institute, 1968) 58-59, to suggest reading yam-mi (singular with enclitic mem) rather than plural yammim in Gen 1:10 and 1:22. However, such variations may point rather to a certain fluidity and freedom in the poet's use of these parallel pairs, perhaps for metrical reasons.

18Cf. the entry "tripartite division of cosmos"in the Index of Motifs in M. Dahood, Psalms HI (AB 17A; Garden City: Doubleday, 1970) 491.

i'Dahood and Penar, "Parallel Pairs," 126-27 (No. 71) and 356 (No. 554). 20J. Crenshaw, Hymnic Affirmation of Divine Justice (SBLDS 24; Missoula: Scholars Press,

1975) 143, argues that the doxologies in Amos are from the exilic or early post-exilic period, the same era as that to which the literary work of Ρ is ordinarily assigned.


That Gen 1:27 is a fragment of poetry has long been recognized. Its clear parallelism, chiastic structure, and the threefold repetition of the verb (in the yqtl 11 qtl sequence) all support this conclusion. The syllable count is 13:9:10. The meter is improved in the following rearrangement:

Gen 1:27 wybrD °lhym °t tfdm bslmw ..„ , _ -

• °lhym brD °tw

zkr wnqbh br° °tm

The structure proposed here differs from the MT in finding a conflate reading in the second colon. One variant (bslmw) would produce the translation "In his image God created him"; the other (with bslm in construct to the following Dlhym) would mean "In the image of God he created him," agreeing with Gen 9:6 (ky bslm Dlhym csh Dt hDdm). We have then the not uncommon situation of two ancient variants in the tradition.21

In defense of this analysis, the following observations are pertinent. First, the LXX includes only one of these variants (kai epoièsen ho theos ton anthröpon katf eikona theou [ = bslm Dlhym] epoièsen autori).22 Second, we have in Ps 33:5 = Ps 119:64 a syntactical situation almost exactly like that which we propose for Gen 1:27. In Ps 33:5 we have hsd yhwh mPh hDrs, where hsd stands in construct relationship to the following yhwh (as in bslm Dlhym): "Of the kindness of Yahweh the earth is full." Ps 119:64 has the variant hsdk yhwh mPh hDrs, where hsd is construed with a pronominal suffix and the following yhwh is a vocative (as in bslmw Dlhym: noun with suffix followed by a nominative): "Of your kindness, O Yahweh, the earth is full." Finally, one should note the stylistic devices of climactic parallelism23 and inclusion in the reconstructed tricolon, which has a syllabic meter of 10:10/9:10.

Because of its similarity to Gen 1:27, Gen 5:1 can be considered here.

Gen 5:1 bywm brD Dlhym °dm bdmwt °lhym cs'h °tw zkr wnqbh br°m

The pair brD / / csh of the tricolon can be found in Isa 41:20, 45:7, 12 in reverse order. In Isa 45:18 and Amos 4:13 csh and brD appear as part of a longer sequence (Isaiah: ysr / / csh / / kwn (polal) / / brD; Amos: ysr / / 6rD

/ / csh).

21 For the phenomenon of conflate readings cf. S. Talmon, "Double Readings in the Massoretic Text," Textus 1 (1960) 144-85.

22Msgr. P. Skehan has suggested to me that the LXX could also be the result of homoioarkton (kai epoièsen ho theos ton anthröpon rkat' eikona autou1 kat' eikona theou epoièsen autori). He asks further: "Would a Greek writer write the full form? LXX Isa, which is early, usually translates only three out of any four hemistichs, where there is parallelism" (private communication).

23On climactic parallelism, cf. Albright, Yahweh and the Gods of Canaan, 4-7, and S. E. Loewenstamm, "The Expanded Colon in Ugaritic and Hebrew Verse," JSS 14 (1969) 176-96.


The examples from Genesis 1 analyzed above were chosen to illustrate the thesis of poetic parallelism effected by the use of formulaic language. We shall conclude this section by adding a few more pieces of evidence from Genesis 1 that support the thesis of poetry less directly. First, one could point to the following examples of other possible formulaic pairs in the Ρ creation account: crb / / bqr,2* tht / / Ί (\:1),25 dP / / csb (1:11-12).26 In Gen 1:22 the ABCB sequence of four verbs is a device attested elsewhere in Hebrew poetry; it will be discussed further in considering Gen 9:7 below.27

II

Having presented the evidence for poetry in Genesis 1, we shall now extend this hypothesis to other parts of the Ρ tradition where, contrary to Genesis 1, the presence of pieces of poetry is somewhat unexpected.

Gen 9:7 w°tm prw wrbw srsw b°rs wrbw bh

The poetic character of Gen 9:7 was demonstrated by Porten and Rappaport, taking up a suggestion of Gevirtz. Gevirtz was apparently the first to notice that the word sequence ABCB was a regular feature of Hebrew poetic diction. For instance, in Isa 28:23 (hDzynw wsmcw qwly / / hqiybw wsmcw Dmrty) we find the verbal sequence hDzynw (A), Smcw (B), hqsybw (C), smcw (B). Gevirtz noticed this same pattern in Gen 27:29 and Isa 54:7.28

Porten and Rappaport increased the number of examples of this device,29

including two passages from Ρ (Gen 1:22 and 9:7). Meter supports their position in 9:7, where the ABCB verbal sequence of prw, rbw, srsw, rbw occurs in an 8:8 bicolon. The recovery of this pattern demonstrates the soundness of MT and prohibits the emendation of rbw to rdw.

Gen 17:6 wnttyk Igwym wmlkym mmk ys°w

Gen 35:11 °ny Ί sdy prh wrbh gwy wqhl gwym yhyh mmk wmlkym mhlsyk ys?w

24Cf. Gen 49:27, Isa 17:14, Zeph 3:3, Pss 30:6,90:5,6, Qoh 11:6 for occurrences of crb / / bqr in poetry.

25Dahood and Penar, "Parallel Pairs," 373 (No. 589). Perhaps one should add the prepositions b / / k in Gen 1:26 and 5:3, since they also seem to form such a pair in Hebrew poetry, often without appreciable difference in meaning. Van Dijk, Ezekiel's Prophecy, 43-45, lists the following examples of b / / k: Isa 5:18, 66:15, Ezek 26:20, Pss 37:20,78:72, 102:4,143:3, 150:2, Qoh 11:5.

26The concordance shows that of all the occurrences of the noun dp, only the two in Gen 1:11 — 12 are supposedly prose uses. Note also the four occurrences of the pair dP / / csb in Deut 32:2, Jer 14:5-6, Isa 37:27 = 2 Kgs 19:26 (where the order is reversed), and Prov 27:25 (dP / / csbwt).

27Cf. note 7 above. 2 8 S. Gevirtz, Patterns in the Early Poetry of Israel (Chicago: University of Chicago, 1963)43-

44. 29In their study Porten and Rappaport add the following: Num 23:7-8, Isa 5:5,42:15, Jer 5:3,

11:11-12, 12:7-8.


The similarities that these two passages share may indicate that they are variants of the kind frequently encountered in orally composed poetry.30

Besides mmk and ysDw, both passages contain the formulaic pair gwym /I m/^m (Isa 41:2,45:1, 52:15,60:3, 11,16,62:2,Pss 72:11,135:1ο)·*1

The expression mhlsyk may also be part of Hebrew poetic diction, since of the ten biblical occurrences of this noun, only one (1 Kgs 8:19 = 2 Chr 6:9) could be counted as prose usage, if Gen 35:11 is counted as poetry.32 Another poetic feature has been recognized by Dahood, who parses the waw of wqhl as emphatic, translating "A nation, nay, a host of nations shall be your descendants."33 Note the chiasmus in 35:11 oiyhyh mmk and mhlsyk ysDw; and it may also be worth noting the assonance of wmlkym mhlsyk y fw, and the syllable counts of 8:9 for 17:6 and 10:10:11 for 35:11.

Gen 17:17 hlbn m>h snh °wlyd w°m srh bt tPym snh tld

Two textual changes are proposed here; the change of ywld to Dwld, following the Samaritan Pentateuch, is proposed by Lohfink and accepted by McEvenue.34 This reading reveals another poetic feature, the presence of an emphatic lamed with Ibn. The parallelism thus revealed would be exactly synonymous, as McEvenue's translation shows: "Will a centenarian beget? And will Sarah who is ninety years of age give birth?" The second change would be the deletion of h with hbt in the second colon. Given the h / / wDm structure of rhetorical questions in both prose and poetry,35 the interrogative particle in the second colon is otiose and may be explained as a dittography from the preceding srh.36

Instances of formulaic language that may point in the direction of poetry include the pair bn / / bt,31 and the repetition of the same word or root in

30Cf. Lord, The Singer of Tales, passim, on the degree of variation possible in orally composed poetry.

31 For the related pair mmlkh / / gwy cf. W. Moran, "A Kingdom of Priests," The Bible in Current Catholic Thought (ed. J. L. McKenzie; New York: Herder and Herder, 1962) 7-20.

32And on the evidence of other fragments of poetry in the prayer of Solomon (1 Kgs 8:12-13), even this usage may be poetic: rq °th Ρ tbnh hbyt / / ky °m bnh hyp mhlsyk / / hwD ybnh hbyt Ismy, an 8:11:8 tricolon with wordplay on bn and bnh, as in Gen 16:2.

"Dahood, "Hebrew-Ugaritic Lexicography XII," Bib 55 (1974) 389; for other instances of emphatic waw, cf. Dahood, Psalms HI, 400-02.

3*N. Lohfink, "Textkritisches zu Gen 17, 5.13.15.17," Bib 48 (1967) 439^2; McEvenue, Narrative Style, 172.

35M. Held, "Rhetorical Questions in Ugaritic and Biblical Hebrew," Eretz~Israel9 (1969)71-79.

36A. Ehrlich, Randglossen zur Hebräischen Bibel (Leipzig: J. C. Hinrichs, 1908) 1.69. Another possibility might be to see the present text as a conflation of two original variants wDm bt and whbt.

37Dahood and Penar, "Parallel Pairs," 146 (No. 114).


parallel cola (snh / / snh; Dwld / / tld).u The numerical sequence X / / X + 10, found in Ugaritic poetry, may be present as well;39 however, the absence of this particular pattern from Hebrew and its occurrence here in reverse order (X + 10 / / X) make one hesitate to identify mDh / / tsQym as truly formulaic, although examples of similar reversals of fixed pairs can be adduced.40 The syllabic meter of the bicolon is 9:11.

Exod 14:3 nbkwm41 hm bDrs sgr clyhm hmdbr

Albright's identification of Exod 14:3 as a bicolon of poetry is confirmed by the presence of the formulaic pair Drs / / mdbr.42 Assonance is achieved by the repetition of the syllable hm in both cola, and by the almost rhyming words that end the cola (bazars, midbär). The syllable count of the bicolon is 6:8 (or 7:8 if hm in the first colon is understood as hèmmâ, without the final he as mater lectionis,*n instance of defective orthography that would fit in with Albright's comment on the antiquity of the fragment).43

Exod 16:12 byn hcrbym t°klw bsr wbbqr tsbcw Ihm

This bicolon is made up wholly of formulaic pairs: byn / / b, crbym / / bqr (noted above44), Dkl / / sbc, and bsr / / Ihm. The first prepositional pair is the equivalent of the Ugaritic-Hebrew b / / bn, but in reverse order.45 The pair Dkl / / sbc normally occurs in either a paratactic sequence (Deut 8:10, 14:9, Hos 4:10, Isa 9:19, Joel 2:26, Ps 78:29), or in the

38Dahood and Penar, "Parallel Pairs," 201 (No. 228), 365 (No. 574). On such repetition as characteristic of Ugaritic and Hebrew verse, cf. Dahood, "Ugaritic-Hebrew Syntax and Style," Ugarit-Forschungen 1 (1969) 30-31. On °wld / / tld, cf. the article of Gevirtz referred to in note 16.

39On the numerical sequence cf. Gevirtz, Patterns, 21-22; W. M. W. Roth, "The Numerical Sequence x/x + 1 in the Old Testament," VT12 (1962) 302-7 and Numerical Sayings in the Old Testament (VTSup 13; Leiden: Brill, 1965) 6.

40F. M. Cross, "Prose and Poetry in the Mythic and Epic Texts from Ugarit," HTR 67 (1974) 6-9.

41The root bwk, found in the difficult form nbkym here and only twice more in MT (Joel 1:18, Esth 3:15), probably bears the meaning "squeeze, distress"; in Albright's felicitous translation, "Trappedare they in the land; the desert has barred them in."Cf. J. Barr, Comparative Philology and the Text of the Old Testament (Oxford: Clarendon Press, 1968) 323 (No. 49) and C. Rabin, "Etymological Miscellanea," Studies in the Bible (ed. C. Rabin; Scripta Hierosolymitana, 8; Jerusalem: Magnes, 1961) 388. The presence of the root in three late sources may cast some doubt on the antiquity of Exod 14:3.

42On *rs 11 mdbr, cf. Dahood and Penar, "Parallel Pairs," 122 (No. 63). 43This understanding of hm was suggested to me by Dr. Robert Coote. Cf. Cross and

Freedman, Early Hebrew Orthography (AOS 36; New Haven: American Oriental Society, 1952) 56-57.

^See note 24 above. The pair byn hcrbym / / bbqr is found elsewhere in Ρ (Exod 29:9,30:7-8, Num 28:4).

45Dahood and Penar, "Parallel Pairs," 136 (No. 95).


common idiom Dkl lsbc as in Exod 16:3 below (also Lev 25:19,26:5, Isa 23:18, Ezek 39:19, Prov 13:25). However, there is one clear example of this pair in poetry: wyDkylhw mhlb hth / / wmswr dbsDsbyck (Ps 81:17). And to this datum can be added the following evidence for the synonymity of DÀ7and sbc: Prov 12:14 (mprypy Dysysbc twb) and Prov 13:2 (mprypy Dys'yDkltwb). On this basis we can probably identify Dkl / / sbc as one of the pairs that form the repertoire of Hebrew poetry,46 possibly the result of the breakup of a stereotype phrase. Evidence for the last pair is equally scanty. Dahood's analysis of Isa 58:7 reveals a good example of Ihm / / bsr.41 And we have a prose text (1 Kgs 17:6) where, along with the pair bbqr / / bcrb, bsr and Ihm are linked. Their appearance in Exod 16:12 (and in 16:8 below) in parallel cola may be a second instance of the breakup of a stereotype phrase.48 The meter is 9:7.

We may note two other fragments of poetry in Ρ that employ this pair:

Exod 16:3 bsbtnw Ql syr hbsr bDklnw Ihm lsbQ

Exod 16:8 bQrb bsr Pkl wlhm bbqr lsbc

The chiastic structure of 16:8 (ABC / / BAC) we have met elsewhere in priestly poetry (Gen 1:5, 10). The syllable count for 16:3 is 9:7; for 16:8, 7:7.

Exod 40:34 wyks hcnn °t °hl mwcd wkbwd yhwh mP °t hmskn

As reconstructed by F. M. Cross, the bicolon would read:

yksh cnn °hl mwcd kbwd yhwh mP mskn

This reconstruction removes such prosaizing elements as the conjunction, article, and the nota accusativi to reveal what Cross considers to be something approximating the original form of the verse. Even without such attempted reconstruction, the character of the passage as poetry is evident from the pairs with which it is composed. Cross comments: "We know that Dhl and mskn constituted a formulaic pair already in Ugaritic verse. One might argue that cnn and kbwd similarly form a poetic pair."49 In addition to the prose parallel

46R. G. Boling, "'Synonymous' Parallelism in the Psalms," JSS 5 (1960) 228. 4 7M. Dahood, "The Chiastic Breakup in Isaiah 58, 7," Bib 57 (1976) 105.

48E. Z. Melamed, "Break-up of Stereotype Phrases as an Artistic Device in Biblical Poetry," Studies in the Bible, 115-53; J. S. Kselman, "A Note on Ps 51:6," CBQ 39 (1977) 251-53.

49Cross, Canaanite Myth, 168. For °hl / / mskn, cf. Dahood and Penar, "Parallel Pairs," 102-3 (No. 15); W. H. Schmidt, "î?t̂ o als Ausdruck Jerusalemer Kultursprache," ZAW75( 1963) 91-92.


cited by Cross (Exod 24:15-16), further evidence for just this pair is provided by another bicolon of poetry from the Ρ tradition, with a syllabic meter of 9:7:

Num 17:7 whnh kshw hcnn wyr° kbwd yhwh

In addition to this evidence for the formulaic character of Dhl / / mskn and cnn 11 kbwd, the third pair of Exod 40:34 (ksh / / mP) also seems to be formulaic. It occurs in Hab 2:14 (ky tmP hDrs ldct Dt kbwd yhwh / / kmym yksw clym), Hab 3:3 (ksh smym hwdw / / wthltw mPh hDrs), and Isa 11:9 (a variant of Hab 2:14: ky mPh hDrs dch Dtyhwh / / kmym lym mksym). Note the collocation of mP with kbwd yhwh in Exod 40:34, Hab 3:3 and Isa 11:9. So in Exod 40:34, as in Exod 16:12, the bicolon is composed wholly of formulaic pairs, a fact which strongly argues for the presence of poetry. The syllabic meter of Exod 40:34 is 10:11 (or 8:8 in Cross' reconstruction).

Exod 40:38 ky cnn yhwh cl hmsTcn ywmm w^s thyh lylh bw

As to formulaic language, ywmm / / lylh need no further discussion beyond that in Gen 1:5. The prepositions c/ / / b occur as a parallel pair in both Ugaritic and Hebrew verse.50 And the epic motif of the pillars of cloud and fire has generated the Hebrew pair cnn / / Ds here and in Pss 78:14,97:2-3, and 105:39. In Ps 78:14 and in Isa 4:5 the pair is collocated with ywmm and lylh. Structurally, note the chiastic order ofc/ hmsknywmm and lylh bw. The meter by syllable count is 11:7 (or a better 9:7 if yhwh in the first colon is omitted, as in LXX).

Num 14:2 Iw mtnw b°rs msrym Dw bmdbr hzh Iw mtnw

The word pair Drs / / mdbr was encountered above in Exod 14:3. In that bicolon, Egypt was apparently identified with the desert, an identification that is possible on the basis of such texts as Ezek 20:36 and (on the basis of a new reading) Ps 105:27.51 The situation is slightly different in this text from Numbers, where Egypt and the desert are not two synonymous elements, but rather the two stages of Israel's trek to Canaan. This understanding of the pair Drs 11 mdbr is reflected in a similar use of the pair in such poetic texts as Jer 2:6 and Amos 2:10. Note the chiasmus and repetition52 in this 7:9 bicolon of poetry.

Deut 34:7 Ρ khth cynw wP ns Ihh

50Dahood and Penar, "Parallel Pairs," 292 (No. 417). 5 1H. J. van Dijk, "A Neglected Connotation of Three Hebrew Verbs," VT 18 (1968) 28,

proposed this reading for Ps 105:27: sâmû bammidbär (MT barn dibrê) °ôtôtâyw ûmôpètîm beberes ham ("They brought over the desert his signs, and his wonders over the land of Ham").

«Dahood and Penar, "Parallel Pairs," 271 (No. 374).


Although there is no formulaic pair in Deut 34:7, the clear parallelism suggests the presence of poetry. The apparent hapax legomenon Ih has been discovered elsewhere in Hebrew and Ugaritic poetry.53 The 6:5 meter can be improved to 5:5 by the minor emendation of cynw to cyn, since no suffix is required with the names of parts of the body. An alternate explanation would see the suffix in the second colon as double-duty, with the waw of cynw as a dittograph from the following wP.54

We shall conclude this section by calling attention to four other parallel pairs in Ρ material that may point, if only distantly and indirectly, to poetry. First, three other occurrences in Ρ oí byn hcrbym / / bbqr, met in Exod 16:12, were mentioned in note 44. Gen 23:2 contains the Ugaritic-Hebrew formula bkh I ¡ spd (wybD Dbrhm Ispd Isrh wlbkth);55 and gr wtwsb in Gen 23:4 and Lev 25:23 occurs as a parallel pair in Ps 39:13 and 1 Chr 29:15. The phrase wyrbw wycsmw in Exod 1:7 reflects the pair rb / / cswm> again common to Ugaritic and Hebrew verse.56 Continued study will doubtless uncover other such remnants of poetic repertoire and style in P.57

Ill

To sum up: through a study of some narrative material from P, some twenty examples (nineteen here and Gen 7:11) of poetic bicola and tricóla

53For Ih in Ugaritic, cf. W. F. Albright, "The 'Natural Force' of Moses in the Light of Ugaritic," BASOR 94 (1944) 32-35; A. van Selms, "A forgotten god: lah,n Studia Biblica et Semitica (ed. W. C. van Unnik and A. S. van der Woude; Wageningen: Veenman en Zonen, 1966) 318-26. For Ih in Job 5:26 and 30:2 cf. M. Pope, Job (AB 15, 3rd ed.; Garden City: Doubleday, 1973) 46-47, 219; in Jer 11:19, cf. Dahood, "Ugaritic-Hebrew Lexicography IV," Bib 47 (1966) 409.

54Cf. Dahood, Psalms III, 429-30, for examples of double-duty suffixes in the second colon. The suggestion that cyn without suffix may have originally stood in the text may gain some support from the use of h suffix to mark the ö-vowel of Ihh in the second colon. This mater lectionis predates the use of waw as ö-marker, and the appearance of both early {Ihh) and late (cynw) forms in parallelism may not be original. Another possibility has been suggested to me by M. Dahood: "I suspect that in Deut 34:7, and in other cases of 3rd m. s. h, Ihh should be read lehêhû; in other words Hebrew also preserved Ugaritic mase, suffix -hü in addition to -ó" (private communication).

55Dahood and Penar, "Parallel Pairs," 143 (No. 106). 56Dahood and Penar, "Parallel Pairs," 338 (No. 516). The last three examples of Gen 23:2,4

and Exod 1:7 may all illustrate the device of breakup of a stereotype phrase. Or more likely in this case, the literary device may be more accurately described with M. Held ("Philological Notes on the Mari Covenant Rituals," BASOR 200 [1970] 37) as "the practice of forming a hendiadys from pairs of words employed in synonymous parallelism." In other words, the poetic brace would be primary, and the prose parataxis, secondary.

57Since the submission of this paper, two other possible indicators of poetic repertoire have come to light. First, the pair r°h / / ydc in two passages from P, Exod 2:25 and 6:3, occurs also in Deut 33:9, in Job 11:11 and Qoh 8:16 in reverse order, and most frequently as a prose parataxis ( 1 Sam 12:17, 14:38, 23:22-23, 24:12, 25:17, 2 Sam 24:13, 1 Kgs 20:7, 22, 2 Kgs 5:7, Jer 2:19, 5:1). Second the phrase wbcsym wbDbnym in Exod 7:19 recalls the pair cs / / abn in Ugaritic. Cf. Dahood and Penar, "Parallel Pairs," 302 (No. 441); Z. Zevit, "The Priestly Redaction and Interpretation of the Plague Narrative in Exodus," JQR 66 (1976) 199-200.


have come to light. The basis for their identification as poetry has been the presence of indicators like parallelism, formulaic pairs, the breakup of stereotype phrases, chiasmus, inclusion, word and root repetition, emphatic lamed and waw. Of course the most objective of these is the formulaic pair occurring elsewhere in Ugaritic and/or Hebrew verse. The other rhetorical features are confirmatory rather than probative. Meter, on the basis of syllable count, was determined; and in only two cases (Gen 1:2 and 1:16) was the syllable count for the members of the bicola or tricóla beyond the two-syllable difference that seems acceptable for poetry transmitted in a prose source. If not all of the examples above can be sustained, the possibility of instances of poetry embedded in the Ρ material at least merits examination and further study.

^ s

Copyright and Use:

As an ATLAS user, you may print, download, or send articles for individual use according to fair use as defined by U.S. and international copyright law and as otherwise authorized under your respective ATLAS subscriber agreement.

No content may be copied or emailed to multiple sites or publicly posted without the copyright holder(s)' express written permission. Any use, decompiling, reproduction, or distribution of this journal in excess of fair use provisions may be a violation of copyright law.

This journal is made available to you through the ATLAS collection with permission from the copyright holder(s). The copyright holder for an entire issue of a journal typically is the journal owner, who also may own the copyright in each article. However, for certain articles, the author of the article may maintain the copyright in the article. Please contact the copyright holder(s) to request permission to use an article or specific work for any use not covered by the fair use provisions of the copyright laws or covered by your respective ATLAS subscriber agreement. For information regarding the copyright holder(s), please refer to the copyright information in the journal, if available, or contact ATLA to request contact information for the copyright holder(s).

About ATLAS:

The ATLA Serials (ATLAS®) collection contains electronic versions of previously published religion and theology journals reproduced with permission. The ATLAS collection is owned and managed by the American Theological Library Association (ATLA) and received initial funding from Lilly Endowment Inc.

The design and final form of this electronic document is the property of the American Theological Library Association.

THE RECOVERY OF POETIC FRAGMENTS FROM THE ...

Documents

Transcript of THE RECOVERY OF POETIC FRAGMENTS FROM THE ...