THE USE OF POETRY IN EXPOSITORY PREACHING

The modern era of the study of Biblical poetry commenced in

A.D. 1753 with Robert Lowth’s magisterial work.[i] Yet the analysis of
Biblical poetry still languishes. Part of the problem, concludes Stephen
A. Geller, is our failure to distinguish between different aspects of
parallelism such as the two types of parallelism, namely, semantic
parallelism of meaning and grammatical parallelism of form.[2.] Being
aware of this distinction is of critical importance because most scholars
regard semantic parallelism as the chief feature of Biblical poetry.
Another problem has been the almost fruitless search for some
workable formula for meter similar to that of classical literatures. But,
alas, no one can convince anyone else as to what that formula should be.
Furthermore, even if we could work out a formula, it would no doubt be
of marginal significance. Hence we will not even trouble the exegete with
the impressive speculations which have been made on the subject.
The greatest roadblock to poetical analyses is the vague category
which Bishop Lowth
called “synthetic parallelism.” Unlike the firmly secure categories of
“synonymous parallelism” and “antithetic parallelism,” synthetic
parallelism has always been a problem since Lowth introduced this
nomenclature. Exegetes experience such frustration in this one category,
which covers more couplets than do the other two categories together, that
they begin to distrust the whole approach. This difficulty can be overcome.
But before we get into that, let us set out an orderly method of treating
poetical texts.

The Distinctive Features of Hebrew Poetry

Despite the wealth of classical poetic form in Greek and Latin sources,
most translators and exegetes of the Hebrew Bible continued to be
oblivious to the poetic form in the Old Testament. A few medieval
commentators such as Ibn Ezra and David Kimchi recognized that a few
passages are in a parallel form. But this observation was not applied very
widely at all. For this advance, we had to wait until 1753, when Lowth
noted that parallelism is the chief characteristic of Biblical Hebrew poetic
style.
The definition Lowth supplied has never been superseded: “The
correspondence of one verse or line with another, I call parallelism.

When a proposition is delivered, and a second is subjoined to it, or drawn

under it, equivalent, or contrasted with it in sense, or similar to it in the
form of grammatical construction, these I call parallel lines; and the words
or phrases, answering one to another in the corresponding lines, parallel
terms”[3] From Lowth’s day to this, it has never been seriously questioned
that parallelism, as he defined it, is the dominant stylistic feature of poetry
in the Old Testament.[4]
The only major development in this field since the 1930s has come with
the recent decipherment of the texts from Ugarit. These poetic texts, which
date from the fourteenth and thirteenth centuries B.C., were written in a
language and a style much like the poetry of Biblical Hebrew. Features of
the type of parallelism that they exhibit are virtually identical with many of
the features of Old Testament poetry.
The find at Ugarit also had the effect of helping us to isolate and
identify another distinctive feature of Hebrew poetry. We discovered that
both literatures have several hundred “pairs” of words that appear in a fixed
parallel relationship to each other. These conventionally linked pairs of
words are balanced off against each other in such a way that the “A” word
appears in the “A” line and the parallel “B” word appears in the “B” line.
The recognition of "parallel pairs" immeasurably aids our ability to
interpret Hebrew poetry.[_5] The latest analysis of Ugaritic and Hebrew
texts indicates there are some seven hundred pairs of these words. While
the order in which these words appear is not always fixed in a rigid se-
quence, that literally hundreds of words do occur in parallel pairs is now
very clear.[£]
Besides the presence of parallelism and parallel pairs of words, Hebrew
poetry may also be identified by the absence, or very sparse use, of several
grammatical features. Generally speaking, Hebrew poetry avoids: (1) the
use of the definite article; (2) the sign of the accusative case (Hebrew nx or
nx—the nota accusativi)', (3) the conjunction 1 (ordinarily translated as
“and”); (4) the so- called relative pronoun (Hebrew lux, “which, who,
that”) and (5) the consecutive or conversive forms of the verb (such as the
waw-conversive with the imperfect which gives the narrative past tense;
e.g., Hebrew inx'1, “And he said”).
This is not to say that these forms will never be found in poetry, but they
will appear only rarely, whereas they can be found almost anywhere in a
Hebrew prose text. Thus Hebrew poetry does have some very distinctive
marks. The exegete may not label a text as poetical merely for reasons of
convenience, aesthetics, or apologetics.
The first step that the exegete must make when he suspects a passage
may be poetical is to determine if it is indeed poetry. The Revised Standard
Version of 1952 was the first English translation to cast all the poetic
passages in the Old Testament in the distinctive sticho- metric format. Prior
to that time, only the poetic books of the Old Testament had been given this
distinctive form. Now nearly all of the editions and translations of the
Hebrew text attempt to indicate in some manner the presence of a poetical
format. This does not mean that these decisions must be accepted as final
and canonized by the exegete. The tests suggested above should still be
run.

The Strophe in Hebrew Poetry

The question of whether it is possible to group lines of Hebrew poetry

into stanzas or strophes was stimulated by the work of Friedrich B. Koster
in 1831.[7] While some scholars remained skeptical, it is plain from the
existence of acrostic poems such as Psalm 119 that such an arrangement is
not only possible, but actually required. Hence, what the paragraph is to the
exegete of prose, the strophe is to the exegete of poetry.
One of the most prominent devices used for marking off strophes is the
presence of refrains. Just as Ugaritic poetry has exhibited the strophic
structure by the use of refrains, so have some eighteen Psalms (Pss. 39, 42-
43, 44, 46, 49, 56, 57, 59, 62, 67, 78, 80, 99, 107, 114, 136, 144, 145). And
there are examples from the prophets, such as Isaiah 5, 9-10, and Amos 1,
2, 4. Thus in Psalm 46 verses 7 and 11 have this refrain:

The Lord of hosts is with us,

 The God of Jacob is our refuge.

An even more elaborate example is the thrice- repeated refrain in Psalm

42:5, 11; 43:5:

Why are you cast down, O my soul, and why are you troubled
within me?
Hope in God; for I shall again praise him, my Help and my God.

Isaiah 9:8-10:4 is divided off into strophes by four appearances of the

refrain (9:12, 17, 21; 10:4):

For all this his anger is not turned away

And his hand is stretched out still.

In addition to the recurring refrain, the Hebrew word nb)? is frequently

used to mark the end of a strophe. Selah occurs seventy-one times in thirty-
nine Psalms as well as in Habakkuk 3:3, 9, 13. The problem with sug-
gesting this unusual word as a strophe divider is that we are still uncertain
as to what the word means. Furthermore, the word Selah occurs in some of
the titles of the Psalms, which many have understood to be musical instruc-
tions. However, if the meaning of Selah is “to raise," as some have
conjectured, then the idea that it is a musical notation and a general strophe
marker may not be incorrect, since a “lifting up” of voices or a crescendo
of instruments might well come at the end of some line of thought in the
poetry, namely, at the end of a strophe. It would be unfair, therefore, to rule
it out completely as a strophe divider, for the appearance of this term, at
what would otherwise be awkward places, must function in some of these
cases to mark the end of a strophe. Nevertheless, Selah must be used with
very great caution as an indication of a strophe division.
A third, but much surer aid is the alphabetic acrostic scheme as found
in Psalms 9-10, 25, 34, 37, m, 112, 119, 145, and Lamentations 1-4. In this
arrangement the first word of the initial stanza commences with the first
letter of the Hebrew alphabet, the first word of the second stanza with the
second letter of the alphabet, and so on. In a few cases every line in the
same strophe begins with the same letter. That these poetic lines were
intentionally grouped into one strophe cannot be denied.
Charles Franklin Kraft has a number of additional criteria for
determining strophes: (1) notable changes in rhythm or length of lines
(shortening or lengthening the final line of a strophe); (2) repeated
catchwords (such as the reiterated call “Yahweh” and the use of
introductory or closing formulae such as “Thus says the Lord” or “Says the
Lord”); and (3) chiasmus or “introverted parallelism” (e.g., a four-line
strophe is so arranged that the first and fourth line correspond to each other
as do the second and third—an ABBA pattern).[8]
Two other devices which will help to identify a few examples of
strophic structure are anacrusis and “distant parallelism." Anacrusis is a
technical poetical device in which a single word (it may be an interrogative
or an exclamation such as “How” in Lam. 1:1) stands outside of the basic
pattern of balance and parallelism in the couplets or strophe. The
presence of these words has an effect on all that follows them. Frequently
this feature appears in strophes containing especially expressive materials.
[3.] Mitchell Dahood labeled as “distant parallelism” a rhetorical device in
which parallel paired words are separated from each other. He illustrated
this phenomenon with the parallel pair of (“to
smite”) and Jin^ (“to annihilate”) in
Psalm 18. “I smote them" occurs in verse 39 (Heb.) while “I annihilated
them” comes in verse 41 (Heb.) with four lines intervening. The ancient
listener or reader, in Dahood’s view, would have recognized the parallel
brace and instinctively linked the two. It would seem that distant
parallelism holds some promise as an eighth way in which we might detect
the presence of a strophe.[io]
The point in all this, of course, is that the identification of the strophe is
not an optional matter for the exegete. As the paragraph stated one central
idea and then developed or organized itself around that one theme propo-
sition, so we would contend that the strophe exhibits a central rallying point
around which it organizes its content.
Because parallelism occurs basically in two or occasionally in three
lines, the most frequent strophic structure will be the simple couplet or
triad. Kraft estimates that for 70 to 75 percent of the poetry in the Psalms
the strophic unit is one simple couplet.[n] Less frequently it is the triad
(three lines) and in extremely rare cases it involves four lines (or a
quatrain) or perhaps six lines (e.g., Ps. 19:7- 9)-
There is one further question as to whether some poems which are
composed of couplets (e.g., Pss. 18, 28, 39, and 40; perhaps Pss. 2, 16, 17,
23, and 26) might not occasionally combine two couplets into a quatrain to
form stanzas. The answer to this probe is not yet clear, but existence of the
stanza as such seems secure. Whether couplets may intermingle with triads
or occasionally even an isolated line (which must not be therefore treated
as spurious, i.e., as a later redaction) to make up stanzas must be tested in
each individual passage while we await definitive answers based on
cognate studies and more exact analyses of all Hebrew poetic texts.

The Couplet in Hebrew Poetry

The variety of strophic pattern by this time should be clear. Psalm 2 has
a clear structure of four triads,[12] while Psalm 20 has two stanzas (each of
which is composed of a quatrain or two couplets) and an additional line.
Kraft sees Proverbs 8 as a long poem of four stanzas:

Stanza I (vv. 1-9): Three triads.

Stanza II (vv. 10-21): Six couplets, including one quatrain (vv. 14-17),
forming an "envelope parallelism”—the quatrain occurs in the
middle of the stanza with two couplets before it and two couplets
after it.
Stanza III (vv. 22-31): Two triads (vv. 24-26, 27-29) within an
envelope of two couplets (vv. 22-23) and (vv. 30-31).
Stanza IV (vv. 32-36): Two final couplets of exhortation.[13]

Now that we are certain that Canaanite and Hebrew poetry were
arranged into strophes and in some rarer cases into stanzas, we must ask
how we are to analyze the substance of each strophe and stanza. We cannot
depend here on the theme proposition or topic sentence as we did in
paragraph analysis—at least not in such prosaic forms.
Instead, the exegete must now deal with the couplet. The couplet is
indispensable in analyzing the strophe. The couplet is made up of an A and
B line. These lines may be in parallel form or they may have no parallel
units whatsoever. The tendency, however, in a large number of cases is for
these two lines to exhibit the phenomenon of parallelismus membroram;
that is, a balancing of units of thought, meaning, and form in the two (three
in the case of a triad, four in the case of a quatrain) parallel lines, but not a
balancing of sounds as in European poetry.
Recently, Geller has insisted that we clearly distinguish three aspects of
Hebrew parallelism:^.]
1. Grammatical parallelism exists where words in lines A and B are
fully parallel grammatically, that is, in form, but not in meaning. For
example, lines A and B may both take the form of subject-verb-direct
object, but none of these words are parallel in meaning.
2. Semantic parallelism, on the other hand, is a parallelism in
meaning or thought, not just in form.
3. Rhetorical parallelism designates features which are intended to
produce a certain literary effect.[15] In our usage here, the term rhetorical
parallelism refers to such features as the ballast variant, emblematic
symbolism, climactic parallelism, chiasm, merism, and paronomasia.
Semantic Parallelism

For the purposes of exegesis, grammatical parallelism is of very little

help. Geller would take Lowth’s somewhat dubious category called
synthetic parallelism (George Buchanan Gray’s formal parallelism) and
speak instead of grammatical parallelism.[16] But we are very interested as
exegetes in the meaning package, semantic parallelism. It has two basic
types.
The first type is synonymous parallelism. Here the second line repeats
the idea of the opening line without significant addition or subtraction.
When every grammatical element in line A has a parallel synonym in line
B, we also have grammatical parallelism, or what some have called
complete matching of grammatical units. Some examples of complete
synonymous parallelism are:[i7_]

abb c
Israel does-not know

a* b* do-not c'
My-people consider (Isa. 1:3)

a b c
A-wickcd-doer gives-heed to-false-lips
a* b' c*
And-a-!iar gives-ear to-a-naughty-tongue (Prw. 17:4)

Oftentimes the parallelism is synonymous, yet one key element

(perhaps the subject, the verb, or the object of a transitive verb) will be
deleted. In this case there is an incomplete (i.e., grammatically incomplete)
synonymous parallelism.

a b c
The-earth is-the-Lord’s and-everything-in-it
a’ (bl C
The-world [ ] and-everyonc-who-lives-in-it (Ps. 24:1)

But there is a rhetorical device for just such circumstances. This is what
Cyrus H. Cordon called a "ballast variant.”[i£] In those couplets or triads
where there was no matching unit for one grammatical element (as seen in
Ps. 24:1—"is the Lord’s”), both Ugaritic and Hebrew often compensated
for the omission by lengthening that line. This may be illustrated as
follows:

a b C
A. When-came-out Israel from-Egypt
lai b* C' d
B. ( ] The-house-of-Jacob from-a-people of-strange -
(Ps. 114:1; cf. Jer. 17:10b) language

Notice that the “a” unit is left unanswered in the B line, but bulk is added
almost as if it were compensating “ballast” for the earlier omission in the
line. Thus we would label with Gordon the phrase "of-strange-language”
the ballast variant. Another illustration of this same rhetorical device may
be seen in Psalm 103:7.

a b c
He-made-known his-ways to-Moses

The second type of semantic parallelism is antithetic parallelism. In

b’ c' d' (ballast variant) of-
this type, the second line contrasts with or negates the thought or meaning
his-decds to-the-sons Israel
of the first line. The best place to search for examples of antithetic
parallelism is in the Wisdom or Gnomic Literature of the Old Testament,
especially in Proverbs 10-22. Here are some examples of complete anti-
thetical parallelism:

a be
A-soft-answcr turns-away wrath a’b’ c’

B u t-grievous-word s stir-up anger (Prov. 15:1; cf. 15:2. 20)

a be
Righteousness exalts a-nation
a’ b‘ c*
But-sin is-a-reproach (o-any-people (Prov. 14:34)

Sometimes the antithetical thought is not found internally within the

couplet, but externally between two couplets. A good example can be
found in the prophet Isaiah:

ab c
Thc-ox knows its-owncr

a1 [b'J c' d' (ballast variant)

and-the-ass [ ] its-master's crib

a be
But-Israel does-not know

a' b' c*
My-people do-not consider (Isa. 1:3)

The adversative “but” in the middle of the verse helps us to realize that
there is a clear external antithesis here between the two couplets—one with
a ballast variant.

Rhetorical Parallelism

Besides the balance of form, thought, and meaning, Hebrew poetry

uses a number of rhetorical devices to increase both the beauty and the
simplicity of the meaning. Already we have observed the operation of the
ballast variant at work in both types of parallelism, though it would appear
that it is more at home in synonymous parallelism.
One interesting feature that marks Ugaritic (Canaanite) and Hebrew
poetry off from other Semitic and classical forms of poetry is the feature of
deleting the verb.[19] Thus in Ugaritic we have:
 a be
You-will-take your-eternal kingdom

fa’l b' c'

[ ] your-everlasting dominion (Text 68:10 or
III: AB, A: 10)

a b c d
For-a-son is-born to-me like-my-brothers
a* lb'1 [c*J d'
a-scion [ ][ I ] likc-my-kinfolks (II D; 2:14—15)

Likewise in Hebrew we have:

ab c
Saul has-slain his-thousands

a’ (b’J c1

and-David [ ] his-ten-thousands (I Sam. 18:7b)

a' (b’l c*

The mountains skipped like-rams