Bibliotheca Sacra 103 (411) (July, 1946) 283-92.
Copyright İ 1946 by Dallas Theological Seminary. Cited with permission.
Semitics and Old Testament
THE POETIC STRUCTURE OF THE BOOK OF
JOB AND THE UGARITIC LITERATURE
BY CHARLES LEE FEINBERG, TH.D., PH.D.
Within the short period of less than half a century (1887-
1929) the scholarly world was placed under heavy debt to
two peasants. Through a peasant woman at Tell El-Amarna
in Egypt the valuable Amarna Tablets were brought to
light (1887, and through the plowing of an Alaouite peasant
at Ugarit in Syria the even more important Ras Shamra
texts were later unearthed by the French archaeologist
Schaeffer (1929)." The texts resulting from these discoveries
date from a period about the middle of the second millennium
B.C. The findings at Ras Shamra have opened to us the vast
extent of the Canaanite civilization: its society, commerce,
political institutions, and religion.1 These had formerly been
only imperfectly known through allusions in the Hebrew Bi-
ble and from Greek sources. As study progresses much light
is being thrown not only upon Hebrew lexicography, gram-
mar, and poetry, but also upon the cultural milieu in which
Israel came to live in Canaan.
The task of comparing the Biblical literature with the
Ras Shamra alphabetic texts is an exacting one and has
many ramifications. The purpose of this article is to com-
pare the poetic structure of both literatures. The matters
1 W. F. Albright, CBQ, Vol. VII, 1945, pp. 5-9, and the fuller discussion
in Studies of the History of Culture, pp. 11-50. Note the abbreviations
used in this article: AJSL, American Journal of Semitic Languages and
Literatures; BASOR, Bulletin of the American Schools of Oriental Re-
search; CBQ, The Catholic Biblical Quarterly; JPOS, Journal of the
Palestine Oriental Society; RB, Revue Biblique; RP, Revue de Paris;
ZA, Zeitschrift fur Assyriologie.
284 Bibliotheca Sacra
of similarities and differences in grammar, vocabulary, and
concepts will occupy us in future studies.
Though unanimity has not been achieved on all points
and much remains yet to be done, the study of Hebrew meter
has made definite advance. Some of the early deliverances
on the subject were those of Josephus and Philo, who held
that Hebrew poetry had meter.2 Whether they were judg-
ing by Greek models or not, as some affirm, it is impossible
to determine. Toward the end of the eighteenth century
Lowth made his contribution to the study in his lectures at
Oxford.3 To him we are indebted for characterizing the
basic relationship in Hebrew verse as parallelismus membro-
rum.4 This phenomenon had been noticed before him by Ibn
Ezra (twelfth century) and Kimchi (thirteenth century), but
the latter had not designated it in the clear fashion which
Lowth did. Lowth also maintained that the utterances of
the prophets especially, as well as other parts of the Hebrew
Bible, were originally in metrical form. Subsequent study
has borne out the validity of this position. His shortcomings
were that he drew his examples from Greek and Latin
sources, since he was not conversant with Oriental literature
as such, and that, though he recognized the Hebrew poets
must have had metrical rules, he felt it was impossible to
ascertain them now.
Because of the rich discoveries of the past century through
archaeological campaigns in the Near East, comparisons were
made possible with Babylonian and Assyrian, as well as
Egyptian, poetry.5 Assyrian poems, like the Epic of Crea-
tion and the Descent of Ishtar, reveal that the Accadians had
a regular metric system and that the meter was accentual.
2 The statements of Josephus are not pertinent to Job, because Ant. II 16.1
refers to the song of Exodus 15; Ant. IV 8.44 to Deuteronomy 32; and
Ant. VI 12.3 to hymns composed by David.
3 Robert Lowth, Lectures on the Sacred Poetry of the Hebrews, 1829.
4 For his definition of this phrase cf. R. Lowth, op. cit., pp. 35, 43, and 157.
5 W. F. Albright, JPOS, II, 1922, pp. 69-71.
Job and the Ugaritic Literature 285
Usually the couplets were of two bicola (four hemistichs),
each with a caesura. Delitzsch and Zimmern showed that the
bicolon was 2 plus 2. Some of the poems manifest a com-
plex strophic arrangement as well as a refrain, as in the
Ishtar and Saltu poem.6 The strophes are quatrains with
four bicola. When dealing with the Assyrian poems, we
must keep in mind that much of the Accadian poetry has
been translated from a Sumerian original.7 Not only is the
meter of Assyrian poetry accentual, but, as Erman has
shown, that of Egyptian poetry was also. Generally the me-
ter was 3 plus 3 or 2 plus 2. The period of greatest develop-
ment in prosody in Egypt and Babylonia was during the
Twelfth Dynasty (c. 1989-1776 according to Edgerton's re-
vised low chronology).
The work of Ley and Sievers, along with Budde, Duhm,
and others, was destined to lay the foundation for later
strides in the study of Hebrew metrics. Over a period of
some twenty years Ley occupied himself with the subject and
published three basic works.8 Sievers set out to find the
rhythm of Hebrew poetry and to judge the Hebrew meter
from it.9 The conclusion was that Hebrew did not count syl-
lables, that is, it was not quantitative in the strict sense of
the term, but depended upon the number of accents. Lyric
meter was found to be 2 plus 2 (Canticles), dirge (qinah) is
3 plus 2 (Lamentations), and epic or didactic is 3 plus 3
(Job and Proverbs).
6 E. Sievers, ZA, N.F., 4, 1929, pp. 22-29. Note refrain in strophes XIII
and XXVIII, pp. 23, 24, and 26.
7 W. F. Albright, BASOR, 91, 1943, p. 44.
8 J. Ley, Die metrischen Formen der hebraischen Poesie, 1886 (here much
emphasis was placed on alliteration as a metric form of Hebrew
(poetry); Grundzuge des Rhythmus, des Vers- und Strophenbaues in
der hebraischen Poesie, 1875. (esp. pp. 8-15 on accent as. the principle
of Hebrew meter); and Leitfaden der Metrik der hebraischen Poesie,
9 E. Sievers, Studien zur hebraischen Metrik, 1901. The two basic laws of
his system may be summarized thus: (1) no more than four unaccented
syllables may accompany an accented syllable, so that a word with five
syllables would have two stresses; (2) the accented syllable follows the
unaccented ones and may not in turn be followed by more than a single
unstressed syllable. Cf. G. B. Gray, Forms of Hebrew Poetry, pp. 143-
286 Bibliotheca Sacra
THE POETRY OF JOB
Before entering into a more detailed treatment of the
poetry of Job, we note the view of Bickell and, more recently,
Holscher, because it differs from the position just stated that
the meter of Job is 3 plus 3.10 These scholars, judging the
Biblical material from Syriac patterns where the law of ac-
centuation places the tone on the penult, seek to construct a
system of quatrains for the Book of Job. Bickell holds that
the strophe of the book is "durchgangig je zwei siebensilbige,
rhythmischjambische, inhaltlich parallele Verszeilen zu einem
Doppelverse, und zwei von diesen zu einer Strophe verbin-
det."11 The arrangements resulting from these attempts are
not only quite subjective, but require much emendation of the
text. Rigid conformity to one pattern is not possible through-
out the whole poem, as we shall see.
What type of poetry is Job? Is it drama, Greek tragedy,
a didactic poem, or an epic poem? No one will deny that
the book has dramatic action, but the action in the pro-
logue and epilogue is subordinate to the main purpose of the
work. Nor can we call Job a Greek tragedy for, among other
distinctions, there is nothing in it to answer to the inter-
spersed choral odes. Though its subject matter is of a didac-
tic nature, it is not a didactic poem, for its differences from
the poetry of the Book of Proverbs are clear. It is definitely
an epic poem, treating of a lofty theme with unity and some
progress in the action.12
This poem, the longest in the Old Testament, is unique
in that it combines prose and poetry and utilizes the dialogue,
the narrative being in prose and the dialogue in poetry. In
the historical books of the Hebrew Bible, as well as in the
prophetic books, we have the combination of prose and
poetry, but not in the same manner as Job. Dialogue may
10 G.Holscher, Syrische Verskunst, 1932, esp. pp. 49-123, and Das Buch
Hiob, 1937, esp. pp. 3, 4, 8. His p~ition, as far as Job is concerned, is
that the poem follows the same metric system as the Syriac.
11 G. Bickell, Das Buch Hiob, p. 11.
12 R. Dussaud, RP, 1937, p. 216, thinks Ras Shamra has what Hebrew and
Arabic poetry lack; namely, epic poetry. Surely Job can be placed in
the category of the epics.
Job and the Ugaritic Literature 287
be found in the Song of Solomon (for example, 2 :1-3), but it
is not employed in the same type of discussion.
Attempts have been made to find parallels to the Book of
Job in Semitic literature. The Babylonian poem on the right-
eous sufferer, the so-called Babyloman Job, has been com-
pared to the Biblical Job.13 Even a cursory reading of the
Babylonian selection reveals that the resemblances are slight,
while the differences are considerable. The cuneiform poem
is, moreover, monologue and not dialogue. Among the As-
syrian texts published by Ebeling he entitles one Ein baby-
lonischer Kohelet, but Dhorme thinks the relationship to Job
is closer, although he is not dogmatic on the point.14 The se-
lection contains a discussion of the problem of evil and bears
some striking parallels to Job. It is composed in twenty-seven
strophes and employs the dialogue. Our Judgment would be
that a closer parallel to the subject matter of Job must still
be sought. As to the use of the dialogue in epic poetry, both
a Babylonian and an Egyptian source have been posited.
The "Descent of Ishtar" has been compared with Job, be-
cause in both dialogue is introduced into epic.15 "The Say-
ings of Amenemope" has been suggested as the Egyptian
source of the dialogue." These maxims are arranged
in thirty chapters, and are counsels directed to Amenemope's
youngest son, who was priest in the Temple of Min in Panop-
olis. In form they scarcely parallel Job. Comparisons with
the philosophical dialogue of the Greeks are not relevant.
The 3 plus 3 meter in the Book of Job is unmistakable.
Whether it be in the cycles of addresses of Job and his
friends or in the Elihu monologue or in the Jehovah speeches,
the predominant epic meter is clear. Jerome had spoken of
"the hexameters" of Job 3:2 to 42:6 in distinguishing the
prose from the poetry. There is no serious disagreement
13 H. Zimmern, Der Alte Orient, 7, 3, pp. 28fF.
14 E. Ebeling, Berliner Beitrage zu Keilschriftforschung, 1922. P. Dhorme,
RB, 32, 1923, pp. 5-27.
15 E. Konig, Einleitung in das Alte Testament, pp. 410 f.
16 G. Holscher, Das Buch Hiob, p. 4 refers to H. Gressmann, Altorienta-
lische Texte zum Alten Testament (2nd ed.), 1926, pp. 38-46."
288 Bibliotheca Sacra
with this view, apart from the position of Bickell and Hol-
scher discussed above. The three basic parallelisms--syn-
onymous, antithetic, and synthetic or constructive-appear
in the text, with the great majority of the last type and few
of the second type. An example of each will suffice.
Synonymous parallelism, Job 8:3:
Will God pervert justice,
Or will Shaddai pervert righteousness?
Antithetic parallelism, Job 8:7:
Though your beginning was small,
Yet your latter end will be very great.
Synthetic parallelism, Job 5:19:
In six troubles he will deliver you,
And in seven no evil will touch you.
Though the prevailing rhythm of Job is that of the bal-
anced bicolon with three accents to each colon, rigid uni-
formity is not maintained throughout the poem. Attempts
to impose such uniformity have been unsatisfactory. On the
other hand, variations are comparatively few and must be
dealt with cautiously.17 Ley, according to Budde, claimed to
be able to find 800 bicola out of 1,000 verses.18 The presence
of tricola can be explained as resulting from the poetic
freedom and skill of the writer. Most of the alleged ex-
amples, however, are doubtful or open to suspicion. Those
in Job 3 :4, 5, 6, and 9 probably arise from disturbance in
the text. Possible examples are 7:11; 8:6; 19:12; 38:41;
and 39:25. What appears to be a tricolon of 2 plus 2 plus 2
in 9:21 disappears when we see the probability that the first
two words are vertical dittography from line 20. Few cases
of 3 plus 2 and 4 plus 3 rhythm are original, while 3 plus 4;
4 plus 4; and 2 plus 2 are very rare. However, there are too
many variations from the dominant rhythm to allow the
conclusion that none of them is original.
17 B. Gray, AJSL, 36, 1919-20, pp. 95-102. His emendations are not con-
l8 K. Budde, Das Buch Hiob, p. VII, n. 2.
Job and the Ugaritic Literature 289
When we examine the bicolon more closely, we find a
number of variations in the sentence structure. While the
literary form a b c--a b d occurs in the Hebrew Bible,
there are no examples in the Book of Job. The common
harmonic sequence in Job is a b c--a' b' c'. Variations
from this pattern occur, but we shall occupy ourselves with
the bicolon most frequent in the poem. Following Gordon's
arrangement,19 we allow s, v, o, p, and x to represent sub-
ject, verb, object, prepositional phrase, and adverb or any
miscellaneous particle. Analysis shows that these harmonic
Ibalances are present: pv pv, 4:9,
"By the breath of God they perish,
And by the blast of his anger they are destroyed";
vsp vsp, 6:5 (also 8:11),
"Brays the wild ass upon (when he has) the grass,
Or lows the ox over his fodder?"
pvo pvo, 7:2 (also 26:12),
"As a servant that desires the shade,
And as a hireling awaits his wages";
ovo ovo, 10:11,
"(With) skin and flesh thou dost clothe me,
And with bones and sinews thou dost knit me together";
and vpo vpo, 15:33,
"He shall shake off as the vine his unripe grape,
And he shall cast off as the olive-tree his flower."
Instances could be multiplied, but variety, even within cer-
tain types of bicola, is clear. We are coming to realize in-
creasingly that Hebrew prosody was much more complex
than formerly recognized.20 Early in this century Arnold
held that "The rhythmopoiia of Hebrew is, as we should
expect, of the simplest and crudest description."21 His pro-
nouncement is not borne out by subsequent studies.
In concluding our discussion of the poetry of Job, we
19 C. H. Gordon, Ugaritic Grammar (Rome, 1940), p. 85 if.
20 W. F. Albright, CBQ, 7, 1945, p. 19.
21 W. R. Arnold, Old Testament and Semitic Studies in Memory of W. R.
Harper, Vol. I, p. 202.
290 Bibliotheca Sacra
may note that the poem employs alliteration and assonance
(6:14, 16, 25), rhyme (10:8-18; 39:3), and paronomasia.22
Rhyme, like strophe (see 31:5-10; 37:9-10), is only an occa-
sional form of Hebrew poetry. Efforts have been made to
divide large portions of Job strophically, as in Bickell's
system, but the results are subjective and arbitrary.
With the finding of the Ras Shamra texts we have poetry
which comes from a cultural and literary setting more
closely related to Hebrew poetry than either the Babylonian
or Egyptian. We do well to remember also that the cunei-
form tablets have not undergone the copyings which the
Hebrew poetic books have. In the short period in which the
mythological poems of Ugarit have been studied, certain
distinctive features of the prosody have been noted. Like
Hebrew poetry, Ugaritic poetry is accentual. It is charac-
terized by parallelism with the common rhythm of three
accents to a colon. Examples are numerous so we confine
ourselves to one case. 49 (I AB) III 6, 7:
NrFmt Nmw Mmw
Mtbn jlt MlHn
The heavens rain oil;
The wadies run with honey.
Not only is the bicolon frequent, but the tricolon is common
as well. A case in point is 49 VI 27.
Though the poetry was not quantitative in the strict sense,
as we understand it from Indo-European models, there ap-
pears to have been an attempt at counting syllables. Words
vary from two to four, and even five, syllables. Cases with
more than four are rare. Verbs with double energic nun
appear to have five syllables: Nnprwt and Nnyrdt in 49 II 32 and
33. The number in each colon varied from eight to ten sylla-
bles, with the commonest at nine. If the second member of a
bicolon omitted a word found in the first, there was added in
22 I. M. Casanowicz (Paronomasia in the Old Testament) cites 52 examples
(pp. 91-92) of this literary device in Job.
Job and the Ugaritic Literature 291
the former one or more words to counterbalance the latter, a
"ballast variant" as Gordon calls it.23 A list of such devices
shows how largely it entered into Ugaritic versification. Al-
bright explains the fact thus, "The regularity in the number
of syllables must be connected with the fact that these poems
were chanted with simple melodies adapted to regular poetic
syllabification, not as psalms and liturgies are chanted today
in ecclesiastical music, where almost any number of syllables
can be accommodated to the melody.24
As in Job, the Ugaritic poetry manifests variations from
the parallel cola with three beats. Dussaud, after referring
to the dominant rhythm in Phoenician poetry, holds that
when a colon of two accents follows two cola with three
stresses each, it is always by the intention of the poet. The
uneven colon marks the pause or punctuation.25 Besides the
tricola, Ginsberg marks other divergences from the bicolon:
single (extra-metric) words, as fdxv in 49 III 8; single (ex-
tra-metric) lines,26 as the oft-repeated Hcyv hg xwy; run-on
lines; apocopated end-lines; and rhyme, as ydy and ydhy
in 67 VI 17-21.27
Ugaritic poetry enjoys a wide variety of harmonic bal-
Ances within verses. The poets of Ras Shamra endeavored
by artistic devices to avoid monotony, and the result is an
elaborate system of sentence structure. Gordon has listed
twenty-six different types of verses, and this number does
not exhaust the possibilities.
Before we summarize the similarities between the poetry
of Job and the Ugaritic texts, we call attention to some dif-
ferences. First, there is nothing in Job that answers to the
long sections in Ugaritic poetry which are repeated twice.
Second, the verse-form a b c-a b d common in the cunei-
form texts is completely lacking in Job. Third, Ugaritic poetry
23 C. H. Gordon, op. cit., pp. 83, 84.
24 W. F. Albright, BASOR, 91, 1943, pp 43-44.
25 R. Dussaud, Syria, 16, pp. 196, 198; RP, 1937, p. 212. De Vaux (RB,
1937, pp. 534-535) holds the same position.
26 Such extra-metric lines are found in Job 4:1; 6:1; 8:1; etc.
27 H. L. Ginsberg, Orientalia, N.S., 5, 1936, p. 171.
292 Bibliotheca Sacra
makes use of refrain (49 VI 16-22) as well as strophic ar-
rangements (51 IV 52-57). Job has no example of the for-
mer, and the occasional examples of the latter in the book
are not so extended as the Ugaritic patterns.
The similarities between the poetry of Job and the Ras
Shamra literature may now be summarized briefly. (1) Par-
allelism, with its repetition, marks both literatures. (2) The
3 plus 3 meter based on accented syllables is the dominant one
for both. (3) Lines vary as to the number of words, and
words differ in the number of syllables they contain. The
corollary to this fact is that neither Hebrew nor Ugaritic
poetry is quantitative in the strict sense. (4) There does
seem to be a conscious effort to keep lines approximately to
(the same quantity. (5) Rhythms vary in both literatures,
so that change in rhythm cannot be interpreted as "the blend-
ing of different poems."28 Rigid uniformity is not to be im-
posed on either the Hebrew or Ugaritic poems. (6) The sen-
tence structure within verses reveals great artistic skill.
Prose order does not apply; the elements of the verse may be
found in any order.
Definite points of contact, then, between Hebrew and
Ugaritic poetry cannot be denied. Indeed, the relationship is
closer than that which exists between Hebrew poetry and
that of Mesopotamia and Egypt.
28 C. H. Gordon, op. cit., p. 79, sec. 12.2.
This material is cited with gracious permission from:
Dallas Theological Seminary
3909 Swiss Ave.
Dallas, TX 75204
Please report any errors to Ted Hildebrandt at: firstname.lastname@example.org