Proverbial poetry: its settings and syntax

a transitive to an intransitive clause.

This concludes a very incomplete discussion of

isomorphic and homomorphic features. What has been

proffered is: (1) that observing of bi-colonic syntactic

matching (Line type II) should be complemented by the

scrutiny of sub-lineal elements of syntactic and even

morphological equivalence and variation; (2) that the six

box tagmeme provides an adequate tool for monitoring such

sub-lineal, syntactic and morphological equivalences and

variations, and also points one to surface and deep

structure equivalences and contrasts--moving the analysis

one step toward a scientific, semantic analysis of deep

meaning relationships; and (3) that the writer has

attempted to manifest his method of how one should

syntactically read poetry, based on the data reflected in

the corpus. The major goal of this study is to sensitize

readers to the syntactic equivalences and variations of

syntactic parallelism which are artistically crafted by

the sages and also to provide specific methodology as to

how such features may be scientifically isolated and

monitored. The goal is not the understanding of the

tagmeme per se but of the text--using the tagmeme as a

tool allowing the reader to pry open the door to an

appreciation of the poetic text. Hopefully, this allows

us to move one step closer to the recreation of the actual

thought processes of the inspired sages and, having moved

into their shoes, to better understand their sayings. The

above results and analyses were gleaned from the corpus,

almost at random. This should suggest how much potential

resides in such linguistic descriptions; they should not

be viewed simply as compilations of data which are mere

mountains of syntactic minutia. Perhaps, had space and

time allowed, it would have been of interest to provide a

syntactic commentary verse-by-verse in order to show

further how to read Hebrew poetry. But this will be left

for the reader to reconstruct from the above isolated

examples in conjunction with the discussions on the

literary cohesion of Proverbs 10. One final suggestion

for future study would be the integration of a

linguistically satisfying semantic approach to be embedded

in, and complementary to, the syntactic methodology

developed in this study. Though the quagmires of semantic

description make such analysis extremely tenuous, it is

hoped, nevertheless, that attempts (even facile ones) will

be made in that direction. Such semantic analyses will

provide for further, more accurate mappings between the

syntactic surface structures and the semantic deep

structures, which are well beyond the capacities of the

case grammar employed here. Semantic analyses may also

prove to be more palatable and relevant to those who

merely desire theological conclusions--to those who view

the recreation of the poetic moment as an irrelevant and

fruitless endeavor in the proclamation of divine truth.


An Examination of the Patterns of

Proverbial Noun Phrases
Noun Phrase Frequencies
One of the syntactic characteristics of Proverbs

10-15 seems to be the prominence of the two-member noun

phrase form. This becomes apparent either from a

sensitized familiarity with the syntactic texture of

Proverbs or from a contrastive comparison with the results

of O'Connor's non-proverbial poetic corpus. O'Connor

found that out of 633 nominal formations 550 (87%) were

single noun units and only 83 (13%) were two unit noun

phrases.¹ The nominal phrase structure conspectus for

nominal sentences is considerably higher (out of 154, 82

are single nouns [53%]; 62 are two-member NP's [40%], and

10 are three-element NP's [7%]).² In "normal" Hebrew

poetry one immediately perceives that there is a majority

of the single-unit nominals over the two-unit noun phrase.

In Proverbs, on the contrary, the two-unit NP dominates

(approximately 333 [59%] two-unit NP's; 233 [41%] single-

unit nominals).³ Moreover, one may discover that while

45% of the two-unit NP's fall in the subject slot, as do

43% of the single unit nominals, there is quite a contrast

____________________

¹O'Connor, Hebrew Verse Structure, p. 336.

²Ibid., p. 333.

³Appendix VI presents 329 NP's for analysis (96 are

in isomorphic settings, 73 in homorphisms, and 160 are lone

NP's).
in terms of the distribution in the object slot. Only 10%

(35) of the two-unit NP's fall in the object slot, while

31% (72) of the single unit nominals fall in the object

slot. Thus, the subjects tend toward either single or

double unit nominals (100 single, 150 doubles), while the

object shows a definite favoring of the shorter single-

unit. The 31% of the single unit nominals being found as

objects is further heightened when it is realized that

only 33% of the lines contain a sentence pattern which

allows for an object, while virtually all contain a

subject tagmeme (SVO=101, SVM=53, SV=39, SVOP=13,

Nom.=134). The distributions in the prepositional phrases

(39 [16.7%] single units, 69 [21%] two member NP's) and

subject complements (Psc: 45 [19%] single units, 52 [16%]

two member NP's) are fairly close proportionately. One

wonders whether the contrast between O'Connor's corpus

dominated by single nominal units and the clear margin of

majority in Proverbs 10-15 favoring the two-membered noun

phrase could be suggested as another grammatical feature

which may reflect genre differences?

Four Major Noun Phrase Tagmemes
There are four, two-member noun phrase tagmemes

which are conspicuously dominant in Proverbs 10-15. There

are 73 (22%) tagmemes of the following type:
Hd : N Mod : N[Adj]

-------- + --------------

It : Pos :

[Qual]:
Examples of this structure may be found in Proverbs 10:4,

16, 20, 24, 28, 32 et al. The tabulation of the verse

locations where such tagmemes may be found is in Appendix

VI, which gives a compilation of the two member NP's from

the corpus. It will be noticed immediately that this

tagmeme is found principally in the subject slot (58 times

[79%]) and only rarely as an object (9 times [12%]) or in

a prepositional phrase (5 times [7%]). It occurs

predominantly in isomorphic constructions (54 times

[75%]). This will provide a subject-dominating tagmeme,

the semantic fillers of which will be examined

subsequently. This tagmeme parallels the development of

two other tagmeme groups.

First, there is the

Hd : N Mod : N/Adj/Ptc

-------- + -----------------

It : Qual :

group, which occurs 69 times (21%) and is found 14 times

in isomorphic constructions, 12 times in homomorphisms and

23 times in non-homomorphic patterns (vid. 11:1, 18, 30;

12:19; 14:5, 27 et al.). Thus, while It + Pos[Qual] types

are characteristic of isomorphic constructions, It + Qual

is found predominantly in non-homomorphic mappings. The

grammatical slots which It + Qual take are largely

subjects (40 times [58%], 7 times as objects [10%], 6

times in prepositional phrases [9%], and 16 times as Psc's

[23%]).

The second variation of the It + Pos[Qual] type is

the It + Pos tagmeme:

Hd : N Mod : Ps/N/PN

------- + -------------

It : Pos :
While it is less frequent than the previous tagmemes (30

times [9%]) it is found mostly in non-homomorphic settings

(19 times) and rarely in isomorphic constructions (4

times; vid. 12:11, 15; 14:10, 12, 21, 24, 26 et al.).

Though the two previous tagmemes were Subject fillers,

this one tends toward objects (9 times), prepositional

phrases (8 times), and Psc's (5 times), although it occurs

in subjects as well (8 times).

The fourth major NP tagmeme is of the type:
Hd : N Mod : PS/PN/N

------- + ---------------

It : Sp :
It occurs 55 times, 35 of which are in non-homomorphic

constructions (vid. 11:9, 12, 19, 28, 29; 12:10, 16, 26).

It is used heavily to fill Psc slots (17 times) with the

subject slots (14 times), object slots (12 times), and

prepositional phrases (11 times) all closely behind. The

high percentage of the number of occurrences in the Psc

slot is multiplied when one adjusts for the greater

frequency of the subject slot. An interesting phenomenon

occurred with the non-homomorphic proper noun (PN) tagmeme
of this type. It was found only in first line

constructions, which may represent a tendency to put the

PN (usually the divine name) in the first line while

sometimes pronominally referencing back to it in the

second line (vid. 10:27a, 29a, 15:9a, 16a, 33a et al).

These, then, are the four major NP tagmemes. The

first, It + Pos(Qual), is characteristically used in

isomorphic constructions in the subject slot. The second,

It + Qual, is found largely in the subject slot and in

non-homomorphic mappings. The It + Pos is often

discovered in non-homomorphic mappings and is not as

subject-bound, more frequently filling object slot, Psc,

and prepositional phrase usages. The fourth, It + Sp, is

utilized strongly in Psc and object positions, although it

also occurs in the subject slot. When the specifier is a

proper noun (PN), this tagmeme is always in the first line

of the bi-colon.
Matching Noun Phrase Morphological Patterns
The morphological variations of the NP should not

be ignored. There are cases where, for example, there is

a perfect syntactical isomorphism manifesting a total

syntactic equivalence on the surface and deep levels yet

traces of variation are frequently found embedded in the

morphology. Thus, it may be suggested that morphology and

syntax are played off against one another, since
syntactical equivalence is not allowed to stifle

morphological variation. The shifts of gender are more a

product of the word choices themselves than of a poetic

use of gender shifts although that may be the case in rare

instances.¹

A more definite creative manipulating of

morphology may be seen in the sages' use of number

variations. The morphological number variations have been

examined in all two-member isomorphic noun phrases. It

was observed that out of 33 isomorphic mappings-only 11

times (33%) was there equivalence of number (7x plural; 4x

singular). What was more significant was that 22 times

there was what appears to be a purposeful variation in

number and that 18 of those were from first colon singular

to second colon plural with only 4 examples in the reverse

direction (vid. Appendix VI). Thus, in the NP

isomorphisms examined there seems to be a clear preference

for number variation--possibly to off-set the syntactical

repetition--and the order preferred is singular nouns in

the first colon and plural ones in the second. The

singular-to-plural movement almost always takes place on

the noun which modifies the item (i.e., the second noun in

the phrase) which tells of the character quality of the

possessor. It should also be noted that the evil

____________________

¹Berlin, "Grammatical Aspects of Parallelism," pp.

27-29.
characters are not exclusively the ones designated by the

plural (vid. 10:3, 4, 6; 14:8; 15:19). This proverbial

poetic propensity should be documented further, but is

substantial in the corpus of Proverbs 10-15.
Four Noun Phrase Examples
To facilitate an appreciation for the sage's use,

both in terms of equivalence and variation, of the two

membered noun phrase, four examples shall be observed from

the corpus. First, a two-membered isomorphism from the

subject slots of Proverbs 10:4 will be examined. כַף־רְמִיָּה

[lazy hand] and יַד חָרוּצִים [hand of the diligent] clearly

provide a match. They are two nouns in a construct

relationship--both composed of an item being described

(hand [ כף]; hand [ יַדa ]), followed by the one who possesses

the hand in terms of the character quality of the

possessor ( רְמִיָּה [lazy]; חָרוּצִים [diligent]). The

"synonymous" semantic parallel between כַף / יַד is well

established, being antithetically turned by the presence

of the antonymic contrast of the character of the

possessor רְמִיָּה / חָרוּצִים. Thus, the sage uses syntax as well

as semantics to draw these two noun phrases together for

contrast. Note, too, that both noun phrases fill subject

slots, both of which are causers (cf. 10:6 for the same

type of example, although the isomorphic noun phrases are

embedded in different non-syntactically parallel positions

[PP,O]). The syntactic equivalence, however, is varied
via the morphological shift from the singular רְמִיָּה

(sluggard) in the first colon to the plural חָרוּצִים

(diligent) in the second colon. Other similar examples

are abundant (vid. 10:8, 11, 17, 20, 24, 28 et al.).

Proverbs 10:16 provides a good example of a

perfect matching isomorphism where the subjects contain

the two membered isomorphic noun phrase ( פְּעֻלַּת צַדִּיק [wages

of the righteous]; תְּבֻאַת רָשָע [income of the wicked]) and

the subject complements contain a single noun in

prepositional phrases ( לְחַיִים [to life]; לְחַטָאת [to

punishment]). This illustrates not only the poetic

mapping of equivalent syntactic structures from the first

colon onto the second, but also the tendency, as noted

above, to have the subject filled by a two-membered noun

phrase while the object or subject complement is a lone

noun. Notice, too, that the isomorphic lone nouns also

exhibit morphological number variation beginning with a

plural and going to a singular (isomorphic lone noun

morphology has not been examined in this study).

Perhaps a more interesting example may be seen in

Proverbs 10:27. Here the first colon has a normal two-

membered noun phrase (item + specifier type), with the

specifier being a proper noun ( יְהוָה ), which (as was noted

above) is always in the first colon. The object noun is a

lone noun specifying the time ( יָמִים ) which the Lord adds.

Thus, the quality specified by the subject results in the

extension of days (object). The tendenz for a two-

membered subject is observed in the second colon ( שְנוֹת

רְשָׁעִים [years of the wicked]). There is, hence, a feeling

of surface grammar sameness as both subjects are filled by

two member noun phrases. The beauty of this proverb

unfolds when one uncovers the deep structure of the noun

phrase subject of the second line. One immediately sees

that the two noun phrase tagmemes are different. There is

a collapsing effect which combines the first colon

character-designating subject quality ( יִרְאַת יְהוָה [fear of

the LORD]) and the time-specifying object element ( יָמִים

[days]), the patient upon which the verb acts ( תוֹסִיף

[lenghtens]), collapses into the noun phrase subject of

the second clause which specifies character ( רְשָׁעִים

[wicked]) and the time frame ( שְנוֹת [years]) which are the

patients receiving the action of the verb.

10:27a S O

יִרְאַת וְהוָה יָמִים


שְׁנוֹת רְשָעִים

10:27b S
The deep structure as recorded in the third box indicates

that there is a deep structure link between the object of

the first colon (days, which are lengthened) and the

subject of the second (years). This example is important

because it demonstrates the benefits of the tagmemic

approach, which meticuously maps surface structure

similarities (S:NP), but does not neglect deep structure

relationships (O:Pat; S:Pat). Thus, the syntactic

interweaving between the surface and deep structure has

been described; but this would have been missed if a mere

surface grammatical analysis or a sole case deep grammar

approach would have been taken. There is a collapsing of

three O'Connorian units in the first colon to two in the

second--thereby generating the common 134/123 constituent

count. This analysis, then, allows the reader exactly to

account for how this syntactic reduction takes place.

While many exercises could be carried out on the

data of Appendix VI [Types of NP's], one that has proven

very profitable is to take one tagmemic kind of syntactic

noun phrase and to examine what kinds of semantic units

fill the respective tagmemes. As the It + Pos [Qual] was

the dominant noun phrase type, it provides a good starting

point for such studies. What was found fits well with

intuitive suspicions regarding this major proverbial noun

phrase type. The data could be classified from two

directions. Either the first item unit could be used to

classify or the second possessor (quality) unit could

provide the schema.

The first approach reveals that there are four

major divisions of items so referenced and a fifth

category of miscellaneous types. The first, and most

obvious, is when the items are body parts (20 out of 69).

It is rather common to find the following noun phrases in

Proverbs: the hand of the diligent ( יַד חָרוּצִים --10:4;

12:4); the heart of the righteous/wicked/fool ( לֵב רְשָׁעִים

--10:20; 15:28; 15:7); the tongue of the righteous/wise

( לְשׁוֹן צַדִּיק --10:20, 31; 12:18; 15:2); the lips of the

righteous/wise ( שִׁקתי צַדִּיק --10:21, 32; 15:7); and

especially common, the mouth of the wicked/

upright/righteous ( פִּי רְשָׁעִים --10:6, 11, 14, 31, 32; 11:11;

12:6; 15:2, 28). A second type, not as common as body

parts, is the characterization of mental phenomena: the

thoughts of the wicked/righteous (מַחְשְתוֹת צַדִיקִים --12:5;

15:26); the desires of the wicked/righteous ( חַוַּת רְשָׁעִים

--10:3, 24, 28; 11:23; 12:10; 13:4); and the words of the

wicked/pure ( דִבְרֵי רְשָׁעִים --12:6; 15:26). A third category

is the material possessions owned by the various

characters: the wages of the righteous/wicked ( פְּעֻלַּת צַדִּיק

--10:16; 15:6); the house/tent of the righteous/wicked/

upright ( בֵית צַדִּיקִים --12:7; 14:11, 19; 15:6, 25); or the

boundaries/gates of the widows/righteous ( שַׁעַרֵי צַדִּיק

--14:19; 15:25). The fourth category would be the way of

the righteous/wicked ( דֶרֶךְ שָׂעִים --12:26; 13:15; 15:9, 19).

The miscellaneous categories are again categorized as

being possessed by the righteous, wicked, upright, fools

and wisemen. The items possessed are prayers (15:8, 29);

sacrifices (15:8); root (12:3, 12); years (10:27); light

(13:9); crown (14:24); or sometimes even qualities

themselves such as the folly of the fool (14:8, 24) or the

righteousness of the man of integrity (11:5, cf. 14:8).

The words describing the characters possessing these items

represent the major word groups which occur with high

frequency in the wisdom tradition. Thus the righteous (22

times); the wicked (22 times); fools (4 times); wise (4

times); upright (6 times); diligent (3 times); and several

other with less frequency are used as the possessors of

the various items in the two-unit noun phrases. It seems

possible to take these items of discussion and the

character qualities and, in a manner akin to the earlier

discussions of wisdom literature in this study,

reconstruct the matters of concern to the sages

themselves.

Several conclusions may be drawn from the tagmemic

analysis of the noun phrases of Proverbs 10-15. The two-

unit noun phrase seems to favor a subject position, while

the single-unit noun is more common in the object slot.

The single-noun unit may be readily used in the subject

slot as well. There is an extensive use of the two

membered noun phrase in Proverbs 10-15, which is not

normative when compared to O'Connor's results from a more

standard Hebrew poetic corpus. Thus, one wonders whether

the predominance of the two-membered noun phrase rather

than the lone noun may be a syntactic means bearing on the

question of genre. The four major noun phrase tagmeme

types were: Hd : N Mod :N[Adj] Hd : N Mod :N/Adj/Ptc

------ + ------------, ------ + --------------,

It : Pos : It : Qual:

[Qual]
Hd: N Mod : Ps/N/PN Hd : N Mod : Ps/PN/N

----- + --------------------, and --------- + ------------- .

It: Pos : It : Sp :
The first is found largely a subject and rarely in object

or prepositional phrase slots. It also is frequent in

isomorphic constructions. The second is located most

often in non-homomorphic mappings in subject and subject

complement slots. The third occurs in non-homomorphic

settings in object, prepositional phrase, and subject

complement positions. The final noun phrase tagmeme group

occurs mostly in non-homomorphic settings in all slots,

but is especially common in subject complements. It was

also noticed that the proper name (PN) tagmeme type of the

fourth category was found only in first colon positions.

Morphology was only briefly touched where it was suggested