A Dissertation Proposal Submitted to The Department of Near Eastern Languages and Civilizations

By Eric Reymond

© 1998 All Rights Reserved
Department of Near Eastern Languages and Civilizations
University of Chicago
Approved * 6 May, 1997

A copy of the completed dissertation is available via ProQuest ID 304544227
[Note on displaying fonts in this document]

Table of Contents

I. Abstract

Past attempts at poetic analysis of Ben Sira are either dated or suffer from methodological flaws. No specifically structural analysis of Ben Sira's poetry has ever been attempted. I propose such an examination, focusing in particular on the poems preserved in the Ben Sira scroll from Masada because of the relatively reliable reading offered by that manuscript and because such an approach conveniently limits the scope of the study. In order to describe more precisely the structure of Ben Sira's poetry, my analysis will borrow from recent methods of poetic analysis where the types of parallelism and their distributions are separated out and described independently through charts and prose analysis. Such analysis will throw into relief specific characteristics of Ben Sira's poetry that might otherwise go unnoticed. In addition, careful attention will be paid to how the parallelisms work together to communicate the theme or idea of the poem. After the analysis of this small portion of Ben Sira, comparison with other parts of the book will suggest whether or not the prosody of the book is consistent. Comparison to Biblical and other extra-Biblical poems from the Dead Sea Scrolls will be made to determine the relation of Ben Sira's poetry to earlier and to contemporaneous Hebrew poetry.

II. Introduction

With the discovery of Hebrew versions of Ben Sira among the Cairo Geniza manuscripts, scholars were for the first time able to analyze and discuss the qualities of Ben Sira's Hebrew. However, some debate surrounded the question of whether these manuscripts reflected a Hebrew original. It was not until the unearthing of fragments of a Ben Sira manuscript at Masada that a scholarly consensus has formed around the idea that the Hebrew versions, both from the Geniza and from Masada, point to a Hebrew Vorlage.1 Despite this new assurance as to the authority of the Hebrew versions, comparatively little work has been done on the poetry of Ben Sira. I propose a study of this poetry focusing in particular on the poems attested in the Masada fragments, borrowing from recent methods of poetic analysis.

III. Survey of Literature on the Poetry of Ben Sira

Among the ten full commentaries on Ben Sira that I have consulted only one attempts more than a cursory analysis of the poetry: Patrick Skehan and Alexander Di Lella's in the Anchor Bible Commentary.2 Only two encyclopedic articles even touch on the aesthetic qualities of this poetry.3 Exclusive treatments of Ben Sira's poetry are also very few,4 and I have found none that uses the kind of analysis that has emerged in the last 15-20 years. Among recent studies in Biblical Hebrew poetry, only Watson's Classical Hebrew Poetry includes references and citations from Sirach poetry.5 Geller's6 and O'Connor's7 work specifically focused on what they considered to be older elements of the Bible while Collins's8 study limited itself to prophetic works. Despite the more general scope of Kugel's9, Alter's10 and Berlin's11 books these authors did not cite any Sirach poetry.12 Among the older treatments of Hebrew poetry that I have consulted only Lowth made reference to Ben Sira.13

What work has been done on the poetry of Ben Sira has been carried out by Skehan and Di Lella. Skehan's analysis of Sirach poetry is comprised mostly of philological commentary.14 When he does address poetic issues it is usually to describe and isolate clusters of verses or stanzas. These stanzas are often set apart from one another by different formal elements. The clearest example of his method is seen in his analysis of Proverbs 8. The first stanza is set off by repetition of language and by assonance but the second stanza is marked by imperatives and synonymy.15 Because of the inconsistency of diagnostic elements I find the resulting poetic divisions unconvincing. Furthermore, these different elements are not always consistent with Skehan's division of Proverbs 8. For instance, Skehan asserts that imperatives are one of the distinguishing marks of the second stanza; however, imperatives appear immediately preceding stanza 2 in the last verse of the first stanza. Two of Skehan's articles analyze poems found among the Masada fragments.16 Both are fairly short and deal primarily with philological readings. Di Lella's contribution to the understanding of Ben Sira's poetry is seen principally in two articles.17 In the first article, the more general of the two, Di Lella isolates four features of the poetry: phonetic features (including alliteration, assonance and rhyme), chiasmus, inclusio and poem length of 22 or 23 lines. Most of Di Lella's analysis is comprised of listing examples which, with the exception of the last feature, are drawn from chapters 1 through 36. He does not discuss any features specifically of those chapters that are part of the Masada fragments. Di Lella's second article is an analysis of Sirach 51:1-12 and suffers from some of the same deficiencies as Skehan's, especially his division of stanzas on the basis of different formal elements which do not even align with the purported divisions of the poem. This is seen in stanzas II and III and stanzas IV and V of Sirach 51:1-12. According to him, each line of stanza III begins with a mem except the last. This is true, but its usefulness as a criterion is diminished because the last two lines of stanza II also begin with a mem. In stanza IV he asserts that each colon begins with /we-/ but so do the first three cola of stanza V. Di Lella does note where parallelisms arise and details specific phonetic complements. Like Skehan, Di Lella notes copious allusions to other parts of the book as well as to other passages from the Bible and this is very useful. However, I find the analyses of both Skehan and Di Lella outdated given the progress in poetic analysis over the past 15 years.18

IV. Methodology: Parallelism

Since parallelism is a key component of Hebrew poetry, an outline of my methodology requires some brief statement as regards parallelism. In general, scholars agree that parallelism can take many forms.19 Beyond this, however, there is little consensus, particularly as regards a definition. One of the broadest definitions of the idea of parallelism is found in O'Connor's words: "any single word of a language can be paired with any other."20 This is reminiscent of Roman Jakobson's idea of "pervasive parallelism" which, he believed, "activates all the levels of language."21 "All the levels of language" is a vague reference, but what it refers to, among other things, is phonological, grammatical and semantic elements which interact with each other between parallel lines, and throughout the poem.22 I agree in principle with both O'Connor's statement and Jakobson's idea, but I do not believe that all the connections that can be made between elements of a poem are necessarily valuable for an interpretation - either from a thematic or a structural perspective. It is simply not profitable to compare each element of a poem to every other. In addition to describing the pervasiveness of parallelism, Jakobson asserted that every parallelism was "striking." For this he has been criticized by, among others, Paul Werth who writes:

The mere existence of such patterns [i.e., "pervasive parallelisms"] guarantees neither their effectiveness nor their usefulness.23

To further elucidate what each parallelism contributes Werth distinguishes between semantic, emphatic and euphonious effects (though he admits the idea of "effect" remains a subjective judgment).24 He suggests that what I call phonetic parallelism usually effects euphony.25 What I call syntactic parallelism effects emphasis. Homonyms and pairs of polysemous words also effect emphasis. Morphological parallelism, for him, is of marginal importance. Semantic parallelism has, of course, a semantic effect which, when examined, reveals the theme(s) of the work. This is a slightly more nuanced theory than that which states that parallelism simply unifies the poem or binds its different parts together.26 Furthermore, Berlin has noted that Jakobson was taken to task for suggesting that every linguistic equivalence was perceptible.27 In order to distinguish what is perceptible from what is not, Berlin sets out four criteria that help make an equivalence perceptible: 1) the proximity of linguistic equivalences, 2) the similarity of their form,28 3) the number of these equivalences and 4) the expectation of equivalences. By observing these four criteria the critic can more cogently make the argument for parallelistic patterns. Finally, Dennis Pardee in his work on Ugaritic and Hebrew poetry has separated the types of parallelism from their distribution.29 The types of parallelism are semantic, grammatical (morphological and syntactic) and phonetic; the distributions are of four kinds: half-line (within a single colon), regular (between cola of a verse30), near (in the adjacent verse) and distant (beyond the adjacent verse). By being able to distinguish between types of parallelism, their distribution and their basic effects, as well as being able to argue for their perceptibility, the critic is enabled to talk more precisely and convincingly about specific patterns. This does not necessarily limit the nuances that can be brought out through such an analysis but, rather, provides a means to describe and discuss them.

V. Methodology: Literary Analysis

Some brief statement concerning literary analysis must also be made. I would follow Berlin in her general assumptions for the interpretation of prophetic works:

I assume that a prophet's power is in his rhetoric, and that words ascribed to a prophet must not be trite or trivial. He must be saying something meaningful, and saying it in a manner that will have impact... A literary approach tries to get at the distinctiveness of a prophet's rhetoric and to explain how it achieves its impact.31

I would only add, that a wisdom writer's power also rests in his logic and ability to condense this logic into a pithy form, which is an essential aspect of poetic rhetoric. Issues related to the method of the poem's composition (i.e., oral or written) and performance (i.e., sung, chanted, read) will be suspended as we know little of these aspects. For the same reason specific questions of the poet's intentions will also be suspended.

VI. Methodology: Outline of Individual Poems

Alongside the analysis of the parallelisms, the study will attempt to outline individual poems. This will be based, for the most part, on the themes of these poems first as perceptible in a rapid reading, then as the themes take shape from an analysis of the parallelistic structure. The assumption here, again, is that the determination of the shape and length of Biblical poems can be more surely effected through a thematic analysis than analysis based solely or primarily on formal criteria, as seems to be the method of Skehan as well as van der Meer and de Moor.32 I do not mean to suggest, however, that there are not significant formal markers that can coincide with thematic units.33 These formal elements are a useful way of elucidating different thematic units and will be looked at closely.

VII. Methodology: Quantitative Analysis

My analysis of each poem will include a transliteration, a translation and brief philological notes concerning the reading. After the philological notes will come a quantitative analysis in which approximate length of line34 will be indicated by a comparison of the number of consonants,35 syllables,36 words37 as well as clauses, constituents and units38 in each colon. As has been pointed out by G. D. Young39 and several scholars since,40 Biblical poetry is not amenable to what most term metrical analysis. No system has been devised which describes the sequence of accents, syllables, words or consonants in any Biblical Hebrew poem as conforming to a prescriptive meter of any known types.41 Therefore, it is important to remember that this cocktail of methods to measure lines does not attempt to predict line length, nor to act as the single means for determining the stichometry of a verse but will be used along with parallelism, particularly semantic and grammatical, for such determination.

VIII. Methodology: Semantic Analysis

Following the quantitative analysis I will indicate in a chart semantic parallelism for each colon within a bi-colon as well as for each unit of the colon,42 employing capital letters to describe the former43 and lower case letters for the latter. 'Prime' marks above the lower case letters will indicate semantic parallels. If the letter is simply repeated this shows repetition of the exact word or words. Semantic parallelism includes synonyms, antonyms (both gradable and ungradable opposites), hyponyms, part-whole relations and non-binary contrasts (though no attempt will be made to distinguish between these different types in the chart itself).44 Repetitive parallelism, that is the repetition of a word,45 will first be apparent here. Contextual parallels,46 that is pairs of words or phrases that are only semantically parallel because of a specific context, will be noted in the descriptive prose analysis. For the unit-semantic parallelism superscript numbers will indicate how many words are being grouped together under one letter (e.g., t#b rswm in the appendixed passage is listed as "a2").47 The notation of semantic pairs in this distribution will demonstrate if this poetry exhibits regular semantic parallelism, as has been suggested might be the case for late Biblical poetry or, perhaps, for Biblical wisdom poetry in general.48

Micro-semantic-parallelism, in other words, the noting of each word and the other words semantically related to it will follow Pardee's model49 where each word (including conjunctions and prepositions but not pronominal suffixes) is assigned an Arabic numeral and any word that is parallel to it is assigned the same numeral. To distinguish individual words in semantic sets each parallel word is also assigned a Roman numeral. Repetitions of the same word are indicated by superscript Arabic numerals.50 Then, a list accompanies this chart where the words are listed in the order of their first Arabic numeral. The list is easier to use than the chart because at one glance one can see what the semantic sets are, where they occur and with what frequency. The list includes the verse numbers where each word occurs - repeated if the word occurs more than once in a verse. Repetitive parallelism will also be apparent here. The micro-semantic charts and lists will demonstrate the web of semantic relationships woven throughout the poem, suggesting the poem's theme(s). In addition, it allows us to map the near and distant semantic parallels between bi-cola, something the macro-analysis does not allow us to do.

IX. Methodology: Grammatical and Phonetic Analysis

Grammatical macro- and micro-analysis will be done together using Collins's system as augmented by Pardee51 with a short description of each form following the majuscule letter. For instance, wrbd as the subject of the sentence might elicit S (noun, 3ms suff).52 In this way both syntactic and morphological parallelism will be shown together. Phonetic parallelism will follow the method of Pardee where each significant consonant is isolated in a column to the right of the poem.53 All the charts will appear in association with a transcription of the poem. An example of this notation for a small portion of Ben Sira is provided in the appendix to this paper.

Some short justification for what might appear a pedantic use of charts seems called for. Often in older analyses the critic would elucidate what he felt to be most interesting, outlining only elements of style and then sometimes cataloging these and sometimes not. Furthermore, even those studies by the likes of Jakobson who believed in the pervasive aspect of parallelism, leave one less than convinced that he has taken into account all the proofs and arguments for and against his analysis that the text provides. Charts, in short, can act as a control to keep the critic honest. Their benefit to the poetic analysis is to isolate specific types of parallelisms. How the different kinds of parallelisms are working together must be taken up in a descriptive prose analysis. The effectiveness of such a structural approach depends, as Werth has pointed out, on the discrimination of the critic. If the critic includes too many words or phrases as parallel, then the method loses its ability to isolate the theme, for instance. If the critic includes too few he or she runs the risk of finding no pattern or no identifiable pattern.

Following each chart will be a descriptive prose analysis elucidating any of their ambiguities. The explications will be as brief as possible, saving the majority of the discussion for the analysis that will follow all of the charts. This final prose analysis will seek to elucidate the thematic ramifications of all the parallels. Particular parallels will be analyzed in terms of their type, distribution and effect. In this way I hope to present a prose analysis more unified than are Pardee's analyses which tend to be spread out over different sections and then summarized in a concluding paragraph.54 The structure of the poem will be discussed in terms of the content, how this is reflected in the semantic parallels in addition to grammatical and phonetic parallels. The theme of one poem will then be compared with the themes of the surrounding poems.

X. Corpus Definition

As can be ascertained from the above description of the method of poetic analysis that I plan to use, a limitation of the corpus to be studied is a necessary practical consideration. This is especially pertinent given the fact that Ben Sira is one of the largest books of the Bible and approximately 70% of this is preserved in Hebrew.55 A focus on the poems preserved in the Masada fragments presents a convenient limitation. The problems posed by the readings of the Geniza manuscripts are often very complex and hard to resolve56 and, thus, these poems do not immediately lend themselves to translation, let alone poetic analysis.

On the other hand, the Masada manuscript, where undamaged, can be read reliably and where only partially preserved, it provides an important control to the Geniza manuscripts. The number of poems in the Masada scroll depends on which commentary or translation is consulted.57 Between chapter 40, verse 1158 where the first legible column begins and 44: 17d where the text breaks off, there are 116 verses, including fully attested, partially missing and entirely missing verses. There are 300 cola, 172 of which59 require no restoration or only the partial restoration of words that are extant in the Geniza manuscripts. Of the 128 cola (43%) that require the restoration of a word or more only 2360 are not attested in the Geniza texts and have to be reconstructed on the basis of the Greek and Syriac. What is more, the Masada fragments are particularly interesting from a poetic perspective because of their stichometric arrangement and also because of the marks in the margins of the scroll that appear to mark off larger semantic units. The photographs of the scroll in Yadin's edition61 will be used in conjunction with Strugnell's corrections to Yadin's reading.62 Facsimiles of Geniza MSS are found in The Book of Ecclesiasticus in Hebrew.63 A precedent for the use of a similarly sized text for a dissertation-length analysis is provided by Cotter's work on Job which analyzed only 48 verses.

The concluding section of the dissertation will first attempt to determine whether or not the prosody of the Masada poems is similar to that of other poems of Ben Sira found in the Geniza manuscripts. Then, it will compare the poems of this study with samples of the poetry of the Book of Proverbs and with some of the extra-Biblical poetry found among the Dead Sea Scrolls (e.g., the Hodayot and the Songs of the Sabbath Sacrifice), attempting to discern any commonalties. In addition, comparison will be made to earlier Biblical poetry to determine whether or not there appears to be a development in poetic expression.

XI. Conclusion

My study will hopefully improve upon the work of Skehan and Di Lella, covering specific poems that have not received a great deal of their attention. The analysis will focus on the structure and form of the poems but will not be a taxonomic listing, rather, an attempt to describe the network of parallelisms in the text. My method will borrow heavily from Pardee's analyses, distinguishing between types and distributions of parallelisms. If what Pardee has said of Proverbs 2 is also true for Ben Sira, then we might expect Ben Sira's poetry to show, in general, a regular distribution and frequency of semantic and grammatical parallelism, chiastic structure, but little repetitive parallelism in regular distribution. In addition to this, descriptive prose analysis will attempt to elucidate the semantic relationship of the different pairs, how these work together, and how the order in which these are encountered expresses the poem's structure and theme.

Appendix: Sample Analysis

Poetic Analysis of 41: 14a, 16a-16c:

Mynb w(m# t#b rswm
y+p#m l( wml[khw]
#wbl hw)n t#b lk )l
rxbn Mlkh lk )lw
Hear, O children, the instruction on shame,
[be hu]mble (lit. humiliated) according to my judgment(s).
Not every shame is worthy of being ashamed of;
not every humiliation should be chosen.

Semantic analysis:

Mynb w(m# t#b rswm
y+p#m l( wml[khw]
#wbl hw)n t#b lk )l
rxbn Mlkh lk )lw
A a2 b c 1I 2I 3 4
B d a, 5 2II 6 1II
A a2 b a, 7 8 2I2 9 10 2I3
A a2,, b, 52 72 82 2II2 11

N.B. The first letter of the unit-semantic analysis for vs. 14a represents the construct chain t#b rswm and, therefore, in contrast to the micro-analysis, the unit-analysis does not indicate the parallelism between the noun t#b and the verb wmlkh.

Semantic sets:

  1. Instruction
    • I rswm (vs. 14a)
    • II +p#m (vs. 16a)
  2. Shame
    • I t#b // #wb (vss. 14a, 16b) // (vs. 16b)
    • II Mlkh // Mlkh (vss. 16a) // (vs. 16c)
  3. (m# (vs. 14a)
  4. Mynb (vs. 14a)
  5. w (vss. 16a, 16c)
  6. l( (vs. 16a)
  7. )l (vss. 16b, 16c)
  8. lk (vss. 16b, 16c)
  9. hw)n (vs. 16b)
  10. l (vs. 16b)
  11. rxbn (vs. 16c)


  • Inner:
    • t#b // #wb
  • Regular:
    • rswm // +p#m
    • t#b // Mlkh
    • )l // )l
    • lk //lk
    • t#b // #wb // Mlkh
  • Near:
    • t#b // t#b // #wb // Mlkh
    • (m# //rxbn
  • Distant:
    • not applicable.

Grammatical Analysis:

Mynb w(m# t#b rswm
y+p#m l( wml[khw]
#wbl hw)n t#b lk )l
rxbn Mlkh lk )lw
O ( cstr. chain: noun + noun) V (G imv. pl.) Voc. (noun pl.)
V (N imv. pl.) M (prep. + noun pl. + 1 sing. suffix)
S (cstr. chain: kl + noun) P (adj. sing. [negatived])64M (prep. + G inf. cstr.)
S (cstr. chain: kl + N inf. abs.65) V (Nperf. 3ms [negatived]/N part.)

Phonetic Analysis:

Mynb w(m# t#b rswm
y+p#m l( wml[khw]
#wbl hw)n t#b lk )l
rxbn Mlkh lk )lw
s s s m m m b b
kl l s m m
l kl l bs bs
l kl kl


* In accordance with the rules of the Department of Near Eastern Languages and Civilizations of the University of Chicago this dissertation proposal was approved by the dissertation committee and successfully defended at a public hearing. The members of the committee are:

  • Dennis Pardee (Chairman)
  • Norman Golb
  • John Collins

This document was published on-line for the first time on 21 April 1997, courtesy of the Oriental Institute Research Archives. The only changes from the version approved by the Faculty of the Department of Near Eastern Languages and Civilizations are some small changes to accommodate the HTML encoding, and correction of minor typographical errors. HTML encoding was done by Charles E. Jones. [Return to text]

This article requires a Hebrew font and a Transliteration font. The Hebrew font used is SPTiberian. The transliteration font is SPAtlantis. These fonts, developted for Web use by Scholars Press, are available for Mac and Windows platforms, and are in the public domain. Please click here to download them. [Return to text]

1 Patrick Skehan and Alexander Di Lella, The Wisdom of Ben Sira, Anchor Bible 39 (New York: Doubleday, 1987), 54. [Return to text]

2 The full length commentaries that I consulted, in addition to the one cited above, are: Moses Zevi Segal, Sefer Ben Sira (Jerusalem: Bialik Institute, 1953); Rudolf Smend, Die Weisheit des Jesus Sirach (Berlin: George Reimer, 1906); W. O. E. Oesterley, The Wisdom of Ben Sira (London: Society for Promoting Christian Knowledge, 1916); idem, The Wisdom of Ben-Sira (Ecclesiasticus), Cambridge Bible for Schools and Colleges (Cambridge: Cambridge University Press, 1912); Antonino Minissale, Siracide (Brescia: Editrice Queriniana, 1988); John G. Snaith, Ecclesiasticus or The Wisdom of Jesus, Son of Sirach, Cambridge Bible Commentary (Cambridge: Cambridge University Press, 1974); A. Eberharter, Das Buch Jesus Sirach oder Ecclesiasticus, Die Heilige Schrift des Alten Testamentes ubersetzt und erklärt in Verbindung mit fachgelehrten 6, 5 (Bonn: P. Hanstein, 1925); Bruce Vawter, The Book of Sirach (New York: Paulist Press, 1955); Helmut Lamparter, Die Apokryphen I: Das Buch Jesus Sirach, Die Botschaft des Alten Testaments 25 (Stuttgart: Calwer Verlag, 1972); and Norbert Peters, Das Buch Jesus Sirach oder Ecclesiasticus, Exegetisches Handbuch zum Alten Testament 25 (M,nster: Aschendorff, 1913). This last book does contain in the introduction a brief discussion of Ben Sira's poetry, in particular metrical and strophic structure, but is less a description of the poetry than an outline of the critic's methodology. [Return to text]

3 These were in The Anchor Bible Dictionary (Alexander Di Lella, "Wisdom of Ben Sira, The" in The Anchor Bible Dictionary, vol. 6 [New York: Doubleday, 1992], 931-945) and in The Interpreter's Dictionary of the Bible (T. A. Burkill "Ecclesiasticus" in The Interpreter's Dictionary of the Bible, vol. 2 [New York: Abingdon Press 1962], 13-21). Those articles which did not treat the aesthetics of the poetry were Malka Hillel Shulewitz, "Ben Sira, Wisdom of" in Encyclopedia Judaica, vol. 4 (Jerusalem: Macmillan and Co., 1971), cols. 550-553; Alexander Di Lella "Ecclesiasticus" in New Catholic Encyclopedia, vol. 5 (New York: McGraw Hill, 1967), 33-34; idem, "Sirach, Book of" in the New Catholic Encyclopedia, vol. 13 (New York: McGraw Hill, 1967), 257-58; A. Gelin, "Ecclesiastique (Livre de l')" in Dictionnaire de theologie catholique, vol. 4 (Paris: Letouzy et AnÈ, 1956), cols. 2028-54; F. E. Gigot, "Ecclesiasticus" in The Catholic Encyclopedia, vol. 5 (New York: Robert Appleton, 1909), 263-69; and F. Spadafora, "Ecclesiastico" in Enciclopedia Cattolica, vol. 5 (Vatican City: Ente per l'Enciclopedia Cattolica e per il Libra Cattolica 1950), 40-45. [Return to text]

4 The only one I could find was a comparison of Sirach 44-50 to Greek poetry: Thomas R. Lee, Studies in the Form of Sirach 44-50, Society of Biblical Literature Dissertation Series 75 (Atlanta, Geo.: Scholar's Press, 1986). [Return to text]

5 Wilfred G. E. Watson, Classical Hebrew Poetry: A Guide to its Techniques, Journal for the Study of the Old Testament Supplement Series 26 (Sheffield, England: Journal for the Study of the Old Testament Press, 1984). References to Ben Sira are made throughout the book and are drawn from every part of Ben Sira. [Return to text]

6 Stephen Geller, Parallelism in Early Biblical Poetry, Harvard Semitic Monograph 20 (Missoula, Mont.:Scholar's Press, 1979). [Return to text]

7 M. O'Connor, Hebrew Verse Structure (Winona Lake, Ind.: Eisenbrauns, 1980). [Return to text]

8 Terence Collins, Line Forms in Hebrew Poetry, Studia Pohl: Series Major 7 (Rome: Pontifical Biblical Institute Press, 1978). [Return to text]

9 James L. Kugel, The Idea of Biblical Poetry (New Haven: Yale University Press, 1981). [Return to text]

10 Robert Alter, The Art of Biblical Poetry (New York: Basic Books, 1985). [Return to text]

11 Adele Berlin, The Dynamics of Biblical Parallelism (Bloomington, Ind.: Indiana University Press, 1985). [Return to text]

12 Other, more recent, books on poetry that do not cite Ben Sira are Pieter Van der Lugt, Strofische Structuren in de Bijbels-Hebreeuwse Pözie, Dissertationes Neerlandicae Series Theologica (Kampen: Kok, 1980); Elaine R. Follis, ed., Directions in Biblical Hebrew Poetry, Journal for the Study of the Old Testament Supplement Series 40 (Sheffield, England: Sheffield Academic Press, 1987); Luis Alonso-Schökel, A Manual of Hebrew Poetics, Subsidia Biblica 11 (Rome: Pontifical Biblical Institute Press, 1988), a translation and adaption by Alonso-Schökel and Adrian Graffy of Alonso-Schökel's work Hermeneutica de la palabro. II. Iterpración literaria de textos biblicos, Academia christiana 38 (Madrid: Ediciones Cristiandad, 1987); Wilfred G. E. Watson, Traditional Techniques in Classical Hebrew Verse, Journal for the Study of the Old Testament Supplement Series 170 (Sheffield, England: Sheffield Academic Press, 1994); Oswald Loretz and Ingo Kottsieper, Colometry in Ugaritic and Biblical Poetry, Ugaritisch-Biblische Literatur 5 (Altenberge, Germany: CIS Verlag, 1987); Willem van der Meer and Johannes C. de Moor, eds., The Structural Analysis of Biblical and Canaanite Poetry, Journal for the Study of the Old Testament Supplement Series 74 (Sheffield, England: Sheffield Academic Press, 1988). [Return to text]

13 Robert Lowth, Lectures on the Sacred Poetry of the Hebrews, trans. G. Gregory (Andover: Crocher and Brewster and J. Leavitt, 1829); Alex R. Gordon, The Poets of the Old Testament (New York: Hodder and Stoughton, 1912); Eduard Sievers, Studien zur Hebraischen Metrik, Metrische Studien I (Leipzig: B. G. Teubner, 1901); George Buchanan Gray, The Forms of Hebrew Poetry (London: Hodder and Stoughton, 1915). [Return to text]

14 Patrick Skehan, "Sirach 40:11-17" in Patrick Skehan, Studies in Israelite Poetry and Wisdom, Catholic Biblical Quarterly Monograph Series 1(Worcester, Mass.: Heffernan Press, 1971), 129-31; idem, "Staves and Nails and Scribal Slips (Ben Sira 44:2-5)," Bulletin of the American School of Oriental Research 200 (1970): 66-71; idem, "Structures in Poems on Wisdom: Proverbs 8 and Sirach 24," Catholic Biblical Quarterly 41 (1979): 365-79; idem, "The Acrostic Poem in Sirach 51:13-30," Harvard Theological Review 64 (1971): 387-400. [Return to text]

15 Skehan, CBQ 41 (1979). [Return to text]

16 Skehan, BASOR 200, (1970) is the first article. The second is "Sirach 40:11-17," chapter 21 in his book of essays Studies, (1971). [Return to text]

17 The first article, which also appears in his Anchor Bible Commentary on Ben Sira, is his "The Poetry of Ben Sira," Eretz Israel 16 (1982): *26-*33. The second article is "Sirach 51:1-12: Poetic Structure and Analysis of Ben-Sira's Psalm," Catholic Biblical Quarterly 48 (1986): 395-407. [Return to text]

18 Di Lella even comments, inter alia, that he does not wish to enter in on the debate over certain issues of poetic analysis, and prefers to follow "scholarly consensus" (EI 16 [1982]: *26). [Return to text]

19 Karl Budde is quoted by Kugel: "'The variety of possible relations between the stichoi is endless'" (Idea [1981], 15). O'Connor writes of parallelism that it is a "congeries of phenomena" (Verse [1980], 5). Berlin writes of the "multiaspect or multilevel nature of parallelism" (Dynamics [1985], 25) and Jakobson speaks of a "network of multifarious compelling affinities" (Roman Jakobson, "Grammatical Parallelism and its Russian Facet," Langage 42, [1966]: 429). Hrushovski writes: "In most cases there is an overlapping of several such parallelisms with a mutual reinforcement..." (Benjamin Hrushovski, "Prosody (Hebrew)," in Encyclopaedia Judaica, vol. 13 [Jerusalem: The Macmillan Company, 1971], cols. 1200-01). [Return to text]

20 O'Connor, Verse (1980), 96. Associated with this is the expectancy that, according to Berlin, is established by the poetic form, an expectancy in which the reader/listener perceives "equivalence" where there is only a partial equivalence (Berlin, Dynamics [1985], 11-12). [Return to text]

21 Roman Jakobson, Language 42 (1966): 423. [Return to text]

22 "This focusing upon phonological, grammatical and semantic structures in their multiform interplay does not remain confined to the limits of parallel lines but expands throughout their distribution within the entire context..."(Jakobson, Language 42 (1966): 423.). [Return to text]

23 Paul Werth, "Roman Jakobson's Verbal Analysis of Poetry," Journal of Linguistics 12 (1976): 54. [Return to text]

24 The distinction between euphonious and emphatic and between emphatic and semantic is a little obscure in this article, but I take 'emphatic' to mean that words eliciting this effect draw attention to themselves and to like words, not necessarily contributing to the theme or idea of the poem as a whole, only to their specific colon. [Return to text]

25 Exceptions to this are when phonetic parallelism is combined with syntactic parallelism or the phonetic parallel exhibits onomatopoeia. [Return to text]

26 David W. Cotter, A Study of Job 4-5 in the light of Contemporary Literary Theory, Society of Biblical Literature Dissertation Series 124 (Atlanta, Geo.: Scholar's Press, 1992), 82. Note, however, that Werth's euphonious effect of parallelism would have, essentially, the same end-result. [Return to text]

27 Berlin, Dynamics (1985), 130. [Return to text]

28 i.e., repeated words or repeated syntactic or surface structure. [Return to text]

29 He writes : "What I see as my contribution is the insistence that the possible distributions be classified systematically in any study of the macrostructure of a given poem and that all types of parallelism (repetitive, semantic grammatical [including morphological and syntactic] and phonetic) be systematically sought in each of their distributions" (Dennis Pardee, Ugaritic and Hebrew Poetic Parallelism, Supplements to Vetus Testamentum 39 [Leiden: E. J. Brill, 1988], 7 n. 13). [Return to text]

30 A verse being a bi-colon or a tri-colon. [Return to text]

31 Adele Berlin, Zephaniah, Anchor Bible 25 A (New York: Doubleday, 1994), 20. [Return to text]

32 An exception to this may be seen in the acrostic poems of the Bible. The distinction is that the acrostics follow a prescribed pattern that determines their beginning, middle and end. The analysis of van der Meer and de Moor and their followers (The Structural Analysis of Biblical and Canaanite Poetry, [1988]) suffers from their belief that the structure of the poem can be discerned, primarily, through "markers of separation": vocatives, imperatives, deictic particles and tri-cola none of which need appear consistently. Supposedly content also plays a role, though this is only a "provisional" role. Additionally, they seem obsessed with the idea that each poem can be divided into larger parts which can then be subdivided into smaller parts which in turn can be subdivided. [Return to text]

33 For instance, although I have yet to be convinced of Greenstein's five formula for poetic closure in Biblical Hebrew, I find his method much more convincing for the simple reason that closure is indicated, for him, by the structure of the verses themselves not by relatively minor elements such as particles, morphological forms or assonance (Ed Greenstein, "How Does Parallelism Mean?" in A Sense of Text: The Art and Language in the Study of Biblical Literature, Jewish Quarterly Review Supplement (Winona Lake, Ind.: Eisenbrauns, 1982). [Return to text]

34 The concept of "approximate length" is taken from Pardee. He notes: "In general, however, one can refer to parallelism as the main structural element in Hebrew poetry and approximately comparable length of line as the main quantitative element" (Dennis Pardee, "Ugaritic and Hebrew Metrics" in Ugarit in Retrospect, ed., G. D. Young [Winona Lake, Ind.: Eisenbrauns, 1981], 113-30). [Return to text]

35 First used by O. Loretz ("Die Analyse der ugaritischen und hebraischen Poesie mittels Stichometrie und Konsonantenzo/oohnung," Ugarit Forschungen 7 [1975]: 265-269) then by Loretz and Kottsieper (Colometry [1987], 26) and indicated experimentally by Pardee (Pardee, VTS 39 [1988], 71; idem, "Structure and Meaning in Hebrew Poetry: the Example of Psalm 23," Maarav 5-6 [1990]: 242). [Return to text]

36 Used by D. N. Freedman ("Pottery, Poetry and Prophecy," Journal of Biblical Literature 96 [1977]: 5-26) and indicated experimentally by Pardee (Pardee, VTS 39 [1988], 71; idem, "Structure," Maarav 5-6 [1990]: 242; idem, "Accrostics and Parallelism: the Parallelistic Structure of Psalm 111," Maarav 8 [1992]: 119). This is in contradistinction to Stuart's attempt to recover a syllabic meter in Hebrew poetry through use of numerous emendations (D. K. Stuart, Studies in Early Hebrew Meter, Harvard Semitic Monograph 13 [Missoula, Mont.: Scholar's Press, 1976]). [Return to text]

37 Following Pardee's example particles do not count as words but compound particles do (Pardee, Maarav 5-6 [1990]: 242). [Return to text]

38 The last three are part of O'Connor's constriction system. [Return to text]

39 G. D. Young, "Ugaritic Prosody", Journal of Near Eastern Studies 9 (1950): 124-133. [Return to text]

40 Cotter, Job (1992), 48-73. [Return to text]

41 Pardee, Retrospect (1981), 113-30. [Return to text]

42 A unit will be a noun (subject, object, predicate), construct chain, noun + adjective, verb (finite or otherwise) or modifier (adverb, preposition plus noun). [Return to text]

43 Not all semantic, colon-length parallels need be synonymous. They can be stipulating opposite things while using semantically parallel, or identical words. This will still be considered semantically parallel on the level of the bi-colon. E.g., Ben Sira 41:13 t[obat h[ay mispar yamim // wet[obat [s]em] )eyn mispar (The benefit of a living thing is of a limited number of days//but the benefit of a name has no such limit). [Return to text]

44 This list condenses and slightly modifies the list of sense relations outlined in chapter 9 of John Lyon's book Semantics (Cambridge: Cambridge University Press, 1977), 270-335. [Return to text]

45 Regardless of suffixes or definiteness of any kind, nouns of different number, adjectives of different number and gender are considered the same word, as are verbs of differing gender, number or mood. Depending on the verb and the conjugation, verbs of differing conjugations may be grouped together or separately. It follows, therefore, that I do not include root parallelism with repetitive parallelism as Pardee does (Pardee, VTS 39 [1988], 6 n 12). However, I do not follow O'Connor who denies the status of "repetitive" to any two verbs of the same root in different stems (O'Connor, Verse [1980], 109). Determination of semantic sets will be made on a more ad hoc basis. [Return to text]

46 Contextual parallelism is how I will refer to Pardee's sequential/functional parallelism. Pardee defines these as "words or phrases that are grammatically or positionally parallel but of which the semantic proximity is so tenuous that only the context indicates a form of synonymity" (Maarav 5-6 [1990]: 249-50). He gives as examples lqh[ //msk from Anat I (take//mix), two steps in the process of making wine; byt // m(gll from Proverbs 2:18 (house // path), two parts of the strange woman's residence (kiy s]ah[ah )el mawet beytah // we)el repa)im ma(gelotehah Because, her house leads to death // her paths to the Repaim); and bi ne(ot des]e // (al mey menuh[ot from Psalm 23 (in grass pastures // over quiet waters), two places appropriate for flocks to be (bi ne)wot des]e) yarbis[eniy // (al mey menuh[ot yenahhaleniy In pastures of grass he will cause me to lie down// to waters of quietness he will guide me). [Return to text]

47 This follows Pardee's method in VTS 39 (1988), 77-78; idem, Maarav 5-6 (1990): 246-47; idem, Maarav 8 (1992): 121. Particles, prepositions, and conjunctions are not counted as "words" in this notation. [Return to text]

48 Cf., Pardee, VTS 39 (1988), 164 [Return to text]

49 Pardee, VTS 39 (1988), 77-81; idem, Maarav 5-6 (1990): 246-48; idem, Maarav 8 (1992): 121-25. [Return to text]

50 Semantic grades, like those used by Geller will not be used because of the very subjective nature of these distinctions, as well as the further encumbrance they would generate for the notation. [Return to text]

51 Collins, Line Forms, (1978) and Pardee, Maarav 5-6 (1990): 256-57. The letters M, O, P, S, V are used to indicate modifier clauses (including adverbs and prepositional phrases), objects of verbs, predicates of nominal phrases, subjects of verbal phrases and finite verbs, respectively. The placement of parsing in parentheses is my own innovation. [Return to text]

52 Only the marked forms will receive mention of number, i.e., plural nouns will be described as plural but singular nouns will only be described as a noun. [Return to text]

53 Pardee, VTS 39 (1988), 71; idem, Maarav 5-6 (1990): 258-59; idem, Maarav 8 (1992): 135. It should be noted that I will follow a definition of alliteration closer to Cotter's and that of the New Princeton Encyclopedia of Poetics' than to Pardee's. Pardee defines alliteration, citing the Oxford English Dictionary: "repetition of consonants in initial and/or accented syllables..." (Pardee, VTS 39 [1988], 52). Cotter quotes part of the entry under alliteration from the New Princeton Encyclopedia: "...any repetition of the same sound(s) or syllable in two or more words of a line (or line group), which produces artistic effect" (Cotter, Job [1992], 29). The euphonious effect of phonetic parallelism is something I will recognize and is, therefore, in contrast to Berlin's definition of phonetic parallelism which excludes any sound pairs that do not appear with other "linguistic correspondences" (Berlin, Dynamics [1985], 152, n. 2). My view seems closer to Hrushovski's : "Possibly the most widespread use of sound patterning has no direct relation to the meanings of the immediate words" (Benjamin Hrushovski, "The Meaning of Sound Patterns in Poetry," Poetics Today 2 [1980]: 53). I do not agree entirely with him, however, that it is part of a Jakobsonian "poetic function." [Return to text]

54 Cf. Pardee, VTS 39 (1988); idem, Maarav 5-6 (1990); idem, Maarav 8 (1992). [Return to text]

55 Skehan and Di Lella, Ben Sira (1987), 53. [Return to text]

56 Note Di Lella's comment on the subject of the B Manuscript, one of the largest and best preserved: "The Cairo Geniza MSS, especially MS B, have more than the usual share of scribal errors-dittography, haplography, misspellings and misreadings of the exemplar being copied" (Skehan and Di Lella, Ben-Sira, (1987), 59). [Return to text]

57 Skehan and Di Lella count eight (Ben Sira, AB 39 [1987], 462-502) but Segal six (Sefer [1953], 266-306). [Return to text]

58 Following the verse numbers of Ziegler (Sapientia Iesu Filii Sirach, Septuagina:Vetus Testamentum Graecum, 12:2, ed., Joseph Ziegler [G7ouml;ttingen: Vandenhoeck and Ruprecht, 1965]). [Return to text]

59 That is, approximately, 57%. [Return to text]

60 That is, only 7% of the total 300 cola. [Return to text]

61 Yigael Yadin, The Ben Sira Scroll from Masada (Jerusalem: Israel Exploration Society, 1965). [Return to text]

62 John Strugnell, "Notes and Queries on 'The Ben Sira Scroll from Masada'," Eretz Israel 9 (1969): 109-19. Unfortunately, there is only one photo of the unrolled Ben Sira scroll in The Dead Sea Scrolls on Microfiche, ed., Emanuel Tov (Leaden: E. J. Brill, 1993) and it is only of a portion of the actual scroll. [Return to text]

63 The Book of Ecclesiasticus in Hebrew (London: Henry Frowde, 1901). [Return to text]

64 Or, alternatively, V (N perf. [negatived]). [Return to text]

65 The distinction between a previously unattested noun klm and the N stem inf. abs. of this verb is veiled by the absence of vowels. According to Waltke and O'Connor infinitives absolute appear as the subjects of verbal clauses whereas infinitives construct appear as the subjects of non-verbal clauses (Bruce K. Waltke and M. O'Connor, An Introduction to Biblical Hebrew Syntax [Winona Lake, Indiana: Eisenbrauns, 1990], 591, 601). [Return to text]

ERIC REYMOND ©1998-2000
Department of Near Eastern Languages and Civilizations