Three Thousand Years of Hebrew Versification: Essays in Comparative Prosody

Three Thousand Years of Hebrew Versification: Essays in Comparative Prosody

by Benjamin Harshav

In this unparalleled study of the forms of Hebrew poetry, preeminent authority Benjamin Harshav examines Hebrew verse during three millennia of changing historical and cultural contexts. He takes us around the world of the Jewish Diaspora, comparing the changes in Hebrew verse as it came into contact with the Canaanite, Greek, Arabic, Italian, German, Russian,

…  See more details below


In this unparalleled study of the forms of Hebrew poetry, preeminent authority Benjamin Harshav examines Hebrew verse during three millennia of changing historical and cultural contexts. He takes us around the world of the Jewish Diaspora, comparing the changes in Hebrew verse as it came into contact with the Canaanite, Greek, Arabic, Italian, German, Russian, Yiddish, and English poetic forms.
Harshav explores the types and constraints of free rhythms, the meanings of sound patterns, the historical and linguistic frameworks that produced the first accentual iambs in English, German, Russian, and Hebrew, and the discovery of these iambs in a Yiddish romance written in Venice in 1508/09. In each chapter, the author presents an innovative analytical theory on a particular poetic domain, drawing on his close study of thousands of Hebrew poems.

Editorial Reviews

CHOICE - H. I. Needler

“Harshav's analyses are painstaking and precise, and the import of his discussions of the relations of metrics and rhythm, syntax, and semantics in a wide range of verse forms is far-reaching. The author's knowledge of his material, in many languages, is vast . . . this is a major study.”—CHOICE

Robert Alter

“For more than six decades, Benjamin Harshav instructed readers in the operations of poetic technique, but he was always mindful of the subtle and profound articulation of human experience it served. Three Thousand Years of Hebrew Versification is a vivid reminder of the precious legacy he has left us.”—Robert Alter, The Jewish Review of Books

Thomas G. Pavel

"Benjamin Harshav’s extraordinary book is the best available study of verse form, poetic metrics, and poetic rhythm, admirably applied to the unique wealth of Hebrew poetry.  A rigorous and friendly masterpiece authored by one of the most important contemporary specialists in poetics."—Thomas G. Pavel, Gordon J. Laing Distinguished Service Professor, University of Chicago

Paul H. Fry

“With immense learning and concise powers of summary, Harshav explores metrical constraint in verse. Hebrew and Yiddish poetry from the Bible to the present are his focus, but he also introduces the formal conventions of all the Western national poetries.”—Paul H. Fry, Yale University

Jerold C. Frakes

“With the art of a master story-teller and the analytical precision of a reference manual, this consummate cultural theorist unfolds the comparative cultural history of Hebrew and Yiddish verse forms from the Bible to the twentieth century. An immediate classic!”—Jerold C. Frakes, editor of Early Yiddish Texts

Product Details

Yale University Press
Publication date:
Product dimensions:
6.30(w) x 9.30(h) x 1.10(d)

Read an Excerpt

Three Thousand Years of Hebrew Versification

Essays in Comparative Prosody

By Benjamin Harshav


Copyright © 2014 Yale University
All rights reserved.
ISBN: 978-0-300-14487-1


Prelude, ix,
Acknowledgments, xvii,
A Note to the Reader, xix,
1 Basic Aspects of Meter and Rhythm, 1,
2 Do Sounds Have Meaning?, 14,
3 Rhythms of the Bible Revisited, 40,
4 The Systems of Hebrew Versification, 64,
5 The Discovery of Accentual Iambs in European Poetry, 149,
6 Basic Forms of Modern Yiddish Poetry, 198,
7 The Constraints of Freedom in Modern Yiddish Poetry, 216,
8 Free Rhythms in Yiddish Folksongs, 281,
Appendix A: Perception, Rhythmical Groups, and the Bible (1959), 297,
Appendix B: Toward a Critical Theory of the Structures and Functions of Modern Poetry (1959), 302,
Appendix C: A List of Spanish Hebrew Meters, 314,
Notes, 317,
Publications by Benjamin Harshav (Formerly Hrushovski) on Prosody and Related Fields, 337,
Index of Persons, 339,
Index of Topics in Prosody and Literature, 344,
Index of Hebrew Versification and Yiddish Poetry, 349,


Basic Aspects of Meter and Rhythm


In the Modern Hebrew and English traditions, meter can be seen as a regular pulse, as a wave of ups and downs, as a series of stressed and unstressed syllables.

Let us read aloud the first strophe of William Blake's poem "The Tyger."

Tyger! Tyger! burning bright
In the forests of the night,
What immortal hand or eye
Could frame thy fearful symmetry?

Most readers are drawn into it and convey the rhythmical beat smoothly but slow down or stumble in the fourth line. Why?

In this poem the first line has four words, each word has two syllables, and the first syllable is stressed /— [union]/— [union]/— [union]/— 0/. The language itself provides a metrical line. There is a regular beat, starting with the first syllable (ta-ta ta-ta ta-ta ta). By inertia, the beat is continued in the next line, "In the forests of the night." In this tradition of rhymed lyrical poems, the word "bright" in the rhyming position requires a response in the next line or lines. The reader is pulled forward to fulfill both requirements, of meter and of rhyme.

Yet, in the second line, only two words provide definitive stresses: "forests" and "night"; the other words may not be stressed at all in normal speech: "in the forests" is usually read as one word, although it can accept a secondary stress if the reader is swept up by the metrical impulse.

Meter is an abstract and regular pattern provided by the language of the text. It is abstract because it is imposed regardless of any concrete details about the language or degrees of stress; and it is regular because the same abstract unit, the foot /— [union]/, is repeated throughout the poem—in our case, four times in each line and four lines in a strophe. (Some critics would describe this metrical line as three and a half feet, but we use here the number of accents alone, leaving the number of syllables at the end of a line optional.) Now, let us reread the first strophe in its entirety:

Tyger! Tyger! burning bright
In the forests of the night,
What immortal hand or eye
Could frame thy fearful symmetry?

If read emphasizing the four-beat meter of "What immortal hand or eye" (ta-ta ta-ta ta-ta ta), the last line should read, "Could frame thy fearful symmetry?" This sounds grotesque. The poet added only one syllable in front of line 4, so we have to revise our whole reading of the poem. We turn the trochaic meter, going from the stress down /— [union]/, into an iambic meter, going upward /[union] —/, in a total reversal of the rhythmical move. We end up reading those words twice, slowing down, and evoking the "semantic density" of these lines. In adjusting to the deviant syllable, we come up with various solutions for the performance, for example, reading the last two lines as one uninterrupted unit, we get: "What immortal hand or eye could frame thy fearful symmetry?" /— [union]/ — [union]/ — [union]/ — [union]/ — [union]/ — [union]/ — [union] / — [union]/.

How much easier it would have been without the added syllable: "frame thy fearful symmetry?" But Blake does it intentionally: all strophes of the poem have the same additional syllable at the beginning of their last lines; hence it is part of the poem's design. Only with a background of precise regularity in the system of poetry can such a reversal be effective.

In addition to the regular meter and rhyme, there are several free sound patterns (not alliterations). I have in mind clusters of two or more consonants: BuRning BRight enforced by the paradox of a burning tiger. Such clusters are FoRests-FRame-FeaRFul and iMMoRTal-syMMeTRy, where the last two clusters converge in the final formula of the strophe. Thus we have a box covered on three sides: the lower and upper verse and the right rhyme, where every word has a double.

This change of one syllable also indicates a period in English poetry. It is part of a "Romantic" tendency (not necessarily fully Romantic poetry) and not to be expected, for example, in a Shakespearean sonnet.

This device may serve as a marker to differentiate between two shades of English Romanticism represented by Wordsworth and Shelley. Wordsworth is more "classical"; the meter is subdued and automatic. He does not play with detailed variations, whereas Shelley pays a great deal of attention to formal details of the verse and strophic structures. Here is the first strophe in Shelley's "Stanzas Written in Dejection, Near Naples":

The sun is warm, the sky is clear,
The waves are dancing fast and bright,
Blue isles and snowy mountains wear
The purple noon's transparent might,
The breath of the moist earth is light,
Around its unexpanded buds;
Like many a voice of one delight,
The winds, the birds, the ocean floods,
The City's voice itself, is soft like Solitude's.

Wordsworth would use the same pattern of a line throughout a poem, hence unobtrusively, almost automatically. The rhymes here are /abab/bcbc/c for two quatrains, while the second quatrain picks up a rhyme from the first (b), and the last rhyme of the second quatrain is repeated in a ninth-line closure (c). The graphic arrangement highlights the three rhymes in three levels of indentation. Some of the rhymes are precise, while others mix sound rhyme with "eye rhyme." In Blake's poem the rhyme eye/symmetry is a "sound rhyme." The reading symmetry ("symmetry") is forced by the meter. The rhyme buds/ floods/Solitude's has an "ear rhyme," buds/floods, and then a pseudo–eye rhyme in another direction (floods/Solitude's). The last word rhymes by the sound in one direction and by the eye in another. The meter is four iambs per line in eight lines, but expands to ten syllables in the last line in Shelley's self-conscious art of composition. The closure line is reinforced by a pattern of eight sibilants in eighteen syllables.

Logically, there can be only five kinds of meters:

Binary meters
Ternary meters iamb /[union] —/ anapest //[union][union] —/
trochee /— [union]/ amphibrach //[union] — /[union]/
dactyl /— /[union][union]/

The length of the line is labeled in Greek terms: tetrameter (four feet), pentameter (five), and hexameter (six). Sometimes we use a larger foot of four syllables, a paeon. We mark the place of the stress with roman numbers—here, paeon III: /[union][union] — [union]/[union][union] —/ (in the forests / of the night). A paeon easily disintegrates into two iambs or two trochees. In Blake's poem, the reader is hovering between two readings: either four trochees or two paeons III. It is a matter of expressive reading: between a mechanical, monotone beat, on the one end, and the stresses of a content-oriented performance, on the other. We could also retain two levels of accents: in the fOrests of the NIGHT /— [union] [??] [union] /— [union] [??] /. However, this precise matching of language stress and metrical accents in the first line of Blake's poem is rare in English poetry and is used for special rhythmical effects.

Here is another example, of the most widespread meter in English—iambic pentameter—which encompasses all traditional narrative and dramatic poetry, all sonnets, and many lyrical poems. The first sentence in Milton's Paradise Lost is spread out over sixteen lines of verse.

Of man's first disobedience, and the fruit
Of that forbidden tree whose mortal taste
Brought death into the world, and all our woe,
With loss of Eden, till one greater Man
Restore us, and regain the blissful seat,
Sing, Heavenly Muse, that, on the secret top
Of Oreb, or of Sinai, didst inspire
That shepherd who first taught the chosen seed
In the beginning how the Heavens and Earth
Rose out of Chaos: or, if Sion hill
Delight thee more, and Siloa's brook that flowed
Fast by the oracle of God, I thence
Invoke thy aid to my adventurous song,
That with no middle flight intends to soar
Above th' Aonian mount, while it pursues
Things unattempted yet in prose or rhyme.

Every line has ten syllables when read as intended. The first line reads: "Of man's first disobedience, and the fruit" |x x|x| x x x x|x x x|. The syllables ending in a stress are 2-1-3-4. No regularity here. Observing these variegated words and stresses, some critics claimed that there is no iambic pentameter and that all there is are four accents, i.e. the same Old English "native" verse, without any regularity in the number of syllables.

Here, however, is the same line with exactly the same words and the same number of syllables (ten), but with the word "first" moved a few syllables forward: "Of man's disobedience, and the first fruit" /[union] — [union] — [union] — [union] — [union] —/. I only moved the word "first" a few syllables forward. True, Blake's version [1] may fulfill the four-accent requirement of Old English meters, and if we read it stressing only the four accents, we may feel no contradiction between the two versions. But if we read the line in five iambs, my "correction" does not work out. Concretely, this reading is correct:

Of man's first disoBEdience, and the fruit /[union] — [union] — [union] — [union] — [union] —/

And this reading is not. The line is unreadable:

Of man's disObedience, and the first fruit /[union] — [union] — [union] — [union] — [union] —/

I distinguish between language stress and metrical accent. For a verse line to have a certain meter, there must be a correlation between the language and the metrical matrix. The rules of this correlation are as follows. When a meter is applied to language, the feet must adapt to the text.

1) Some words use their stress for an accent (man's).

2) In some words the stress is used in a non-accented position (first).

3) If any metrical accents fall on a word, at least one of them must fall on the stressed syllable.

In fact, the third rule covers all three points and is the final rule of correlation between verse and language. In the word disobedience /— [union] — [union]/, the stress falls on the syllable be, so one or more accents can be added: dis-o-be-dience /— [union] — [union]/. On the other hand, in the incorrect version the stressed syllable be is not accented in the meter; there is a clash between language and meter: disobedience /[union] — [??] —/ has a stress with no accent and an accent without a stress.

In sum: the meter is not given as such in the language of the poem; it is a construct based in the traditional institution of verse. The last line reads: "Things unattempted yet in prose or rhyme" /[union] — [union] — [union] —/ [union] — [union] —/. The long word "unattempted" has two accents, one of them falling on the stressed syllable "temp."

This correlation requirement is not as widely used in English as in Hebrew or Russian; because of the overwhelming amount of monosyllabic words in English, the emphasis shifts to phrase stress and the oscillation between the two, as in "To be or not to be, that is the question" /[union] — [union] — [union] —/ [union] — [union] — [union]/. For the phrase stress, we may emphasize the "is"; for local stress, "that is." Here Shakespeare uses the word-stress rule. But he does not activate the phrase stress and does not emphasize the "is," so we may read "that is the question," "that is the question" or "that is the question." The first line of Paradise Lost reads: "Of man's first disobedience, and the fruit" /[union] [??] [??] — [union] [??][union] — [union] [??]/. "Man's" is accented (rule 1); "first" is not (rule 2); disobedience has two accents, and the second is stressed in the language (rule 3).

Saying that a verse line has a certain meter means that the text can be read by that meter (so-called "scanning"). The meter lies in the base of the poetic text; it is the infrastructure of any rhythmical reading. But the actual rhythmical pattern is woven out of the irregularities that permeate the whole body of the verse. The regular meter is the symmetrical warp, and the free local configurations make up the colorful woof.

This is a good example of verse ambiguity. The same line can enable a construct of an iambic pentameter or a four-accentual paeon. The decision of which to select depends on the literary system to which it belongs. In the first case, the rule of language/verse correlation must be observed. In the second case, the number of syllables between stresses is immaterial.


A "natural" rhythmical unit is one that cannot divide into further regular groups—i.e. it consists of two or three subunits (see Chapter 3). The iambic pentameter has five subunits ("feet") and will normally divide into two asymmetrical subunits (2 + 3 feet or 3 + 2 feet). A key role is performed by the colon, the smallest syntactic-rhythmical unit. The first colon covers the first three iambs; then a second colon begins ("and the fruit"—two iambs). Clearly, this colon is interrupted. In general, the strongest divide between words in Milton's passage is not at the end of a line, but in its asymmetrical middle. Most lines here end close to the middle of a colon and require complementary grammatical parts of speech, or complementary objects in the fictional world, or both. This is not an individual case.

A tension is established between two forces: verse and grammar. As soon as the first colon is finished, the verse line pulls us forward to the completion of the ten-syllable pentameter. And as soon as we reach the end of a pentameter, the colon pulls us beyond the verse boundary, and so on—consistent concatenation. Two functions are at work here: the forward move and the impeding dam. These are accompanied by other rhythmical factors, for example, FiRst DisoBeDieNce/FRuit/FoRBiDDeN TRee—all capitalized consonants are repeated twice. This is an anti-narrative force.

The whole sentence unfolds over sixteen verse lines. That is a powerful pull forward, subordinating a large number of segments, which dam the flow. But the forward pull encounters another force: the division into cola, which constantly break up and stall the forward drive of our reading. This is the reverse of enjambment. Small devices support this double energy; e.g. the first line starts with "Of"—we don't understand what "of" is of until line 6. By poetic inversion we read: "Sing, Heavenly Muse, of ..." We move on, but there are further dams on the way. In the end, a whole world opens up for us: the network of a mythological world based on the Bible, to be unfolded and described in detail later (see Chapter 3).

At first glance, Milton's poetics are similar to the Bible's poetics in that they construct a world: a move from naming a frame of reference, land or sky, to unfolding (through expanding and detailed description). In both texts, names, especially sanctified names (of persons and cities), serve as accumulators of information.

But a major difference lies between them: In the Bible the verse units and the syntactic units overlap; indeed, the syntactic unit determines the rhythmical divisions. In metrical verses a tension is created between the two. The two systematically crisscross each other, creating tensions and non-narrative advancements of the text. This imaginary constructed world has evolved through the dynamization of the syntactic frames. This is hardly an accidental device but a countersystem, which accompanies every precise system.

RHYTHM Meter is the metronome, the regular wave that subordinates all irregular elements in a poetic text (through the rule of correlation). We have: "Of man's first disobedience // and the fruit" /[union] [??] [??] — [union] [??][union]/— [union] [??]/. The physical articulation of language is irregular |x x|x|x x x x x x x|. The five-iamb line is made of irregular units (words) of 2-1-4-3 syllables. The distances between stresses are changing as well. A text is said to have a meter when the metrical matrix can be read in that meter. However, for expressive purposes we may not read it metrically.

Rhythm is the result of all rhythmical inputs that converge in one text or in one local intersection. The total rhythmical impact is a result of a complex performance, which is always different and subordinated to the total reading of a poem. Three forces converge here: (1) the regular meter, absent in free verse; (2) the rhythmical units diverging from regularity in such matters as the size of the words and the place of their stresses; and (3) all non-rhythmical factors that influence the tone and meaning of our reading, for example, the voice of the speaker or narrator, the allusion to a biblical world, the impact of syntax, the suspended syntactic resolutions, and the changed order of words, the total of a rhythmical "image."


Excerpted from Three Thousand Years of Hebrew Versification by Benjamin Harshav. Copyright © 2014 Yale University. Excerpted by permission of Yale UNIVERSITY PRESS.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

Meet the Author

Benjamin Harshav is professor emeritus of comparative literature and J. & H. Blaustein Professor of Hebrew Language and Literature, Yale University, and professor emeritus of literary theory, Tel Aviv University. He lives in North Haven, CT.

Customer Reviews

Average Review:

Write a Review

and post it to your social network


Most Helpful Customer Reviews

See all customer reviews >