rhythm in poetryIntroduction

Metre is a contentious subject, and studies based on mechanical, musical, organic and linguistic analogies have shown how little is currently understood.{1} Nonetheless, as defined as some pattern of phonological stress, pitch and/or length, rhythm is an inescapable element of poetry. Cultural conventions and literary history select their varying requirements from the individual features of a language, but rhythm also arises naturally from the simple exercise of breathing and the desire for shape and regularity in human affairs.

Metre is a systematic regularity in rhythm. In western literature there are two great metrical systems the quantitative (introduced by the Greeks) and the accentual (which appears in Latin of the third century AD) but metre of some sort is found in all poetry, east and west. {2}

Metrical skill comes from practice rather than any slavish following of rules, and rules indeed vary with the literary tradition and what poets are attempting to achieve. The ear is not the only judge. Swinburne and Chesterton appeal to the auditory imagination, but look bombastic on the page. The late blank verse of Shakespeare needs a trained actor to bring out its rough-hewn splendour, and the rhythmic subtleties of Geoffrey Hill are apt to vanish on public performance.

New metres are rarely created, but much more common is the importation and adaptation of metre from a foreign language, which is a good reason for reading beyond translations. Conventional English verse is usually (and confusedly) described in a terminology deriving from classical prosody: as iambic, trochaic, dactylic and anapaestic. On an elementary level it may be better to consider metre under two headings: whether the syllables or the stresses are being counted, and whether these counts are fixed or variable. Accentual verse has fixed counts of stress but variable syllables. Syllabic verse has fixed counts of syllables regardless of stresses. Accentual-syllabic is conventional metre with both stress and syllables fixed. Free verse has no restrictions on either. How readers recognize and respond to metre is unclear, but any particular metre seems to be a norm, a pattern intuited behind permissible examples. The examples are often irregular, and indeed the common iambic pentameter seems only to be exact in some 25% of cases overall. {3}

Accentual verse is found in popular verse, ballads, nursery rhymes, songs and doggerel. Syllabic verse as exemplified by the French alexandrine is not strictly metrical, and twentieth century attempts to write a pure syllabic verse in English have not caught on. Accentual-syllabic was developed by Chaucer from Italian models, and became the staple for English poetry from Elizabethan times till comparatively recently. Contemporary forms of verse originated in France around the middle of the nineteenth century, were championed (briefly) by the founders of Modernism, and have ramified into styles largely indistinguishable from prose.

Traditional verse is overshadowed by the achievements of the past. Chaucer, Spenser, Shakespeare, Milton, Pope and Wordsworth set standards difficult to emulate, and poets are nowadays hardly encouraged to try. Many of the better magazines, where the fledgling poet must start his publishing career, will not take traditional poetry, and those with more generous requirements may still lack readers or editors capable of telling the good from the merely facile. Nonetheless, strict verse enjoys periodic revivals, and has been a feature of several twentieth century schools: the Georgians, Neo-Romantics, the Movement poets and the New Formalists.

Free verse is very confused field, not properly understood or linguistically mapped. {4} Adoption may be more about pamphleteering and cultural aspects than poetic ends. Some of the speech rhythms claimed as "superior to metre" are not rhythms at all but an enviable dexterity in idiomatic expression. Some are loose assemblages of rhythmic expression to no constant base, and some an endearing tribute to their author's performance skills.

Why use the device? Because metre creates and organizes content, giving emphasis to words or elements that would otherwise escape attention: the tighter the metre, the more expressive can be small departures from the norm. Metre gives dignity and memorability, conveying tempo, mood, the subtle shifts in evidence, passion and persuasion beyond what is possible in prose. In the hands of great master like Shakespeare, metre provides grace, energy, elevation, expressiveness and a convincing approximation to everyday speech.

But metre is not diametrically opposed to free verse. Many contemporary poets write both, or served an apprenticeship in strict forms before creating something closer to their needs. Nonetheless, in the absence of this ability to highlight and compound meaning, free verse is often driven to expand in other directions. It prizes a convincing exactness of idiomatic expression, the line seeming exactly right in the circumstances: appropriate, authentic and sincere. It operates closely with syntax. It adopts a challenging layout on the page where line and syntax are rearranged to evade or exploit the usual expectations.

Metre in Practice

Poems need some supporting structure, and that in turn requires a decision: should you go for free verse or tackle the more demanding traditional forms?

Traditional metre and stanza shaping confer certain advantages, and certain disadvantages. They:

1. Please the reader by their display of skill, their variety within order, their continuity with the admired literature of the past.

2. Help the actual writing of the poem, either by invoking words from the unconscious, or by pushing the poem in new directions to escape the limitations of the form.

3. Provide a sense of completeness impossible in free verse. The author knows when the last word clicks into place.

4. Enforce dignity, emotional power and density of meaning.

5. Are more memorable.

The difficulties are equally apparent. Strict forms are:

1. Taxing to write, requiring inordinate amounts of time, plus literary skills not given to everyone.

2. Much more likely to go wrong and expose the blundering incompetence of their author.

3. Inappropriate to the throwaway nature of much of contemporary life.

4. More difficult to place in the small presses.

Theory: Introduction

Firstly, if, as we have said, poets write more by ear than rules, what's to be gained by formal study? Four answers:

1. Rules govern many art forms music, painting, choreography, etc. and are seen as aids to creation once they are so thoroughly ingested as to be second nature. Indeed, without some rules, art fails to be art and becomes instead a perplexity to everyone, not least its practitioners.

2. Poetry is now fragmented into diverse schools, each claiming indisputable truth. As rhythmic expressions are often made into shibboleths, the sensible writer will want an understanding of the issues, so as to choose between the rival claims.

3. Study sharpens the ear, and will locate examples useful to the practising poet.

4. An understanding of rhythm will create links to other art forms, and open the door to current studies in human physiology and behaviour.

Rhythm in poetry can be treated in many different ways. In fact, there exist no fewer than sixteen different theories. Most are in use by literary criticism, and each theory adds depth and significance the others.

Foot Substitution

We start with the most popular theory, that of the foot substitution prosodists. {5} The approach derives from classical scansion, and divides the line into feet. Each foot contains one dominant stress and a number of unstressed syllables. The metres are:
Iambic: rising duple: The cúr | few tólls | the knéll | of párt | ing dáy.
Dactylic: falling duple: Súm mer | ísles of | É den | lý ing | ín dark | púr ple | sphéres of | séa
Anapaestic: rising triple: The Ass ý | rian came dówn | like a wólf | on the fóld
Dactylic: falling triple: Tóuch her not | scórn full y

Verse is rarely completely regular, but if we a. add the pyrrhic (two unstressed syllables) and the spondee (two stressed syllables), and b. allow individual feet to be replaced (most commonly the iambic by a trochaic foot) then most verse can be scanned succinctly. The classification includes type of line (dimeter with two feet, trimeter with three feet, tetrameter with four feet, pentameter with five feet, hexameter with six feet and heptameter with seven feet), and groupings by stanzas of various types.

How real is this periodic stress? It is something instinctively felt on practice, but can be broken into amplitude (i.e. volume), duration and pitch in that decreasing order of importance. Laboratory study on adults and children (and indeed unborn children) has substantiated the reality of the sensation, and some schools of linguistics recognize stress as a feature in everyday speech. In that sense, stress pattern is a reality of the English language, and not something imposed by outmoded literary conventions. If its intricate employment in verse is culturally determined, then so of course is language.

The foot substitution system has many advantages. It is: 1. Widespread in earlier literary criticism, and still very popular. 2. Simple to learn and apply. 3. Sufficiently flexible to cover everything from advertising jingles to Shakespearean blank verse.

Unfortunately, the system also has serious limitations: 1. The classification is artificial, the terminology and to some extent the practice being taken from the wholly different system of classical verse. 2. The classification does not distinguish degrees of stress. 3. Prosodists vary (sometimes very markedly) in their interpretations, and there seems no way of deciding between claims. 4. The system is somewhat rough and ready, and so not over-helpful to the poet. Only the trained ear will distinguish between a good and the merely correct line of verse, and that training exceeds what is needed to apply the system. 5. The system does not properly capture the experience of reading verse. 6. The system enforces correctness at a very elementary level, but metre is not integrated into an overall and illuminating view of a poem's structure.

Temporal Prosodists

The traditional rival to foot substitution has been temporal prosody. Under this approach, which derives from classical verse, the lines are divided into measures like a musical score. {6} Each measure contains a major beat plus one or more subordinate beats. The beats are not marked by pitch or loudness but by the time taken to sound them.  Some are long (major) and some are short (subordinate), but in total the beats make up measures of approximately equal duration. One measure may consist of one major beat lasting six intervals of time, for example, and another consist of three equal but subordinate beats, each lasting two intervals of time. Milton's famous opening to Paradise Lost could be scanned thus: 2|42|222|312|222|42





























Many refinements are possible.  Some prosodies, as in the example above, allow for pauses. Some would insist that the bar-lines fall always before the major beat. Many have indeed employed a musical notation.

One attraction of temporal prosody is its application to prose rhythm. Verse in this view is not something divorced from natural speech, but merely speech made more regular and pleasing. The same measures can be applied to both, and there thus exists a gradual progression from strict forms, through free verse and emotive prose into everyday, spasmodic conversation.

Yet the temporalist approach is not without its problems. Most telling is the complete lack of evidence that such measures actually exist. Decades of testing in speech laboratories have failed to substantiate the notion. Whether in everyday speech, narrative prose or verse, human beings do not linger over syllables in such a way that measures of equal duration can be recognized, or not unless the lines are chanted, when all rhythmic subtlety is lost. Verse seems not to be a heightening of natural speech tendencies, but speech cultivated for a particular end.  "What does not ripen with cultivation?" said Quintilian (c. 90 AD), and "that which is most natural is that which nature permits to be done to the greatest perfection." Verse allows the formal powers of language to release, inflect and modulate meaning, and so needs extensive training to write, to speak and to appreciate


Phrasalists are not concerned with metre, but with rhythmic phrases, these phrases forming spatial patterns that give verse its aesthetic appeal. {7} Lines are scanned for combinations of phrasal shapes repetitions, variations, inversions, etc. and the attraction of a poem ascribed to its figured harmonies of thought.

Some phrasalists ignore metre, or deny it exists at all, particularly in modern free verse, but the majority ascribe the effectiveness of verse to interplay between phrase and metre.  Phrasal prosody may be directed to individual lines, stanzas, or to the poem as a whole. A few prosodists even extend the concept from units of sound to implied meaning, when their analysis encroaches on what has traditionally been considered rhetoric.  Some of Whitman's verse, for example, can be analyzed on four phrasal levels.  Stresses combine into pause groups. Pause groups combine into lines. Lines combine into verse paragraphs. These three levels interact to form a fourth level, which is our experience of reading a poem as a whole.         
Milton composed by verse paragraphs. The end of Paradise Lost, with its regular alternation of 2 and 3 beats phrases, is subdued and dignified, the beat strict as always with Milton: inversions but no extra syllables. The sequence in this passage is 3 || 2 || 3  || 2 | 2 | 3 || 2 || 3 || 2 | 3 :

           Som natural tears they drop'd, || but wip'd them soon; ||
           The World was all before them, || where to choose
           Thir place of rest, | and Providence thir guide
: ||
           They hand in hand || with wandring steps and slow, ||
           Through Eden took| thir solitarie way.

The drawbacks with the phrasalist approach?  Most obvious are those common to all such studies: knowing what is not knowing how.  Findings cannot be applied as recipes, only as guides to self-cultivation.  We also find ourselves recognizing the effectiveness of phrasal structures without entirely knowing why they are effective. Most of the phrasalist approaches until recently have indeed been limited, but the approach does form a basis for the sophisticated theory of rhythmic phrasing developed by David Cureton (see below).

Prose Rhythms

Affective and ornate prose has its own rhythms. {8} What can we learn from their classification?

The most ambitious study was George Saintsbury's A History of English Prose Rhythms (1912), which extended the approach of foot substitution to prose paragraphs. Containing up to five syllables, these prose feet were longer than verse feet, however, and required a complex terminology: 22 types in all. The classification did not catch on, but Saintsbury was able to draw some lessons from the ordered variety.   Juxtaposed feet were never of the same shape, but followed one or more of these patterns: grades moved from short to long and long to short; variations modified a common shape; shapes themselves reflected the content rising for inspiration, for example, and falling for paragraph endings. Always there was opposition between verse feet (recurring feet with some variation) and prose feet (varied feet with some recurrence.) 

Thomas Browne's (1605 - 82) celebrated passage in Urn Burial was analyzed thus:

Now since these dead bones | have already | outlasted | the living ones | of Methuselah,| and in a yard | under ground,| and thin | walls | of clay,| outworn | all the strong | and specious | buildings | above it;| and quietly | rested | under the drums | and tramplings | of three | conquests;| what Prince | can promise | such | diuturnity | unto his reliques,| or might not | gladly | say,

The classification could be taken further. Croll (1919), {9} for example, proposed a system of ascending units. The lowest unit was a colon, a passage typically averaging some ten syllables that could be enunciated in one breath. The next unit was a commata, a melodic and rhythmic passage not necessarily contained by sense and syntax. Then came phrases, which were often twofold:   a first phrase consisting of an appeal with modifying relative clause, and a second phrase of imperative clause or clauses.   The highest unit was the period, defined as perfect in itself and easily comprehended by the understanding. Often these periods coincided with sentences, but they could include more than one sentence.   This, from the English Book of Common Prayer, has two phrases and one period:  

Almighty and everlasting God who dost govern all things in heaven and earth;  

mercifully hear the supplications of thy people, and grant us peace all the days of our life.   

Prose analysts were also interested in the cursus, a rhythmical formula that ended the commata or cola in liturgical writing. The cursus began with a strong accent, continued through a long span of unstressed syllables to a subordinate accent, and then tailed away in (optional) unstressed syllables. There was a recognized terminology for the various types: The planus had no subordinate accent and ended in one unstressed syllable: help and defend us. The tardus ended with two unstressed syllables:   governed and sanctified. The velox had a subordinate accent: punished for our offences.  

Another form of periodicity studied by prosodists was the overall shape of sentences. Periodic sentences place adverbial elaboration before the main clause. Loose sentences place them after. Balanced sentences have initial and final elaborations in equal proportions. Suspended or mid-branching sentences have elaboration between the mandatory elements of the main clause. The matter is complex, and the classifications even more so.  

How does this affect poets? The exact terminology may not matter, but poetry of a high order was obtained by careful elaboration of such phrasing, and might be so again. The elaboration is inappropriate for demotic speech, as for writing of any naturalness or spontaneity, but fine prose shows that free verse is not bound to simple speech rhythms. French poets in particular have created marvels in the twentieth century with such devices, though the prose poem is admittedly easier in loose hexameters than irregular blank verse.

Mais ce soir de grânde age et de grânde patience, dans. . . (St. John Perse) 

More can be done.   Forster divided the rhythmic shape of an extended piece of prose into two categories. Easy rhythms were analogous to a musical motif. An event, action or scene recurred frequently, and each recurrence gained from previous incarnations, so that the prose was provided with a clear internal structure. Barthes  formalized this in semic, symbolic and referential codes. Difficult rhythms are more analogous to the deep structural organization of a musical composition: they can be sensed but not tapped or spelt out. Most good prose belongs to the second category, of course: heterogeneous, unwieldy but somehow pleasing, which Barthes labelled as proaeretic and hermeneutic. {10} 

The advantages of such study? It gets us reading good prose, developing an ear for rhythmic subtleties and structural organization that are needed for the longer poem. We appreciate the poet's art better, understanding what Shakespeare did to Sir Thomas North's prose to arrive at The barge she sat in like a burnished throne...

But the difficulties are again legion. Saintsbury's terminology is daunting, and based on a never-substantiated temporal prosody. Scholars disagree on interpretation, very markedly, and even Saintsbury, in his usual disarming way, could not decide whether the Urn Burial piece quoted starts with a heavily weighted docmaic or four monosyllabic feet.

Free Verse

Free verse is the form most widely used in contemporary poetry, but is anything but new.  Many of its features can be found in ancient Sumerian, Akkadian, Egyptian, Sanskrit and Hebrew literature. As a loosening of traditional metre, the form reappears in 17th century French poetry, and again in 19th century poetry throughout the European languages.

Pinning down the form has caused endless problems. {11} Free verse has been defined as unmetered, metered but unscannable, non-conventionally metered, partially or loosely metered, complexly metrical verse, and so on.  Perhaps all that the many studies have demonstrated is how readily the various forms of free verse evade any simple notation. Foot substitution in particular needs some metric norm to return to, but a norm is what free verse avoids. The lack of normative organization may indeed be the one defining characteristic of this complex and somewhat perplexing form.

Regardless of the facts, modern free verse sees itself as innovative, making a important break with the past.  Two traditions run through American and British poetry. The first originates in Walt Whitman and comes down through Allen Ginsberg and Robert Bly. This is a personal style tending to use asymmetric and often long lines, parallelism, repetition of words and phrases, stresses in unexpected places and mixtures of idiom. The second is more tightly written, with lineation coinciding with grammatical units (D.H. Lawrence, Wallace Stevens and Carl Sandburg) or not so coinciding (William Carlos Williams, Carl Sandburg and Robert Creeley).

Freedom applies not only to freedom from traditional metre, but freedom to use visual and sound effects for surprise, thickening of meaning, symmetry, repetition, or simply for pleasure. Lines can be shortened for speed, or segmented into clots of words or syllables to slow down the reading or comprehension.  Continuously long lines are more the preserve of Whitman (with David Jones, Robinson Jeffers and A.R. Ammons) but alternating long and short have been popular for long poems (Pound, Dorn, Reznikoff and others). Similarly the line breaks: some writers break their lines in unusual and sometimes arbitrary places (Williams), while others run them on (Lawrence).

Equally varied has been the arrangement of the poem on the page. Eleanor Berry {12} in The New Princeton Encyclopedia of Poetry and Poetics, for example, distinguishes some twelve ways in which such visual layouts may support the irony, openness, dissonance, anticlimax, tension, surprise, fragmentation, ambiguity and self-reference of modern poetry.

Those emphasizing the poem’s autonomy or unity may:

1. Give prominence to the sound or structure in the text.
2. Indicate the juxtaposition of ideas or images.
3. Signal shifts in tone, meaning or perspective.
4. Frame a word or phrase.
5. Emphasize (foreground) the text as an aesthetic object.
6. Provide an abstract shape of energy.

The textural and disintegrative may:

1. Signal a reference to poetic tradition.
2. Allude to genres of printed texts.
3. Sustain interest through textural differences.
4. Create counterpoint between structures occupying the same words.
5. Heighten awareness of the reading process.
6. Defamiliarize certain areas or aspects of the text.

Behind these devices generally lies a belief that poetry should be a representation of lived experience. The poetry tries to replicate the perceptual, cognitive, emotional or imaginative processes which gave rise to the poem, and to do so in a manner that seems natural and everyday. Living and writing are matters not to be confused or fused completely, of course, but free verse at least tries to overcome the nineteenth century dichotomy between art and life.

Slavic Theorists

It is Russian and Polish theorists, the Slavic School, who have created the largest body of metrical analysis in the twentieth century. {13} A few are well known in the west (Jacobson, Wellek, Tarjinskaya) but the work of others lies buried in foreign journals, where even here it was fortunate to survive Stalinist purges. From the sixties onwards, however, Slavic approaches have been applied to English verse, initially with considerable success. The approach is characterized by being a) minutely attentive to the texts, b) thorough and wide-ranging, c) supportive of theoretical considerations that were later transported to the west: semiotics and structuralism d) based on linguistics and e) aimed at normative analysis: i.e. not the preferred reading of a particular line but the quantitative generalizations that could be made about all verse of a similar type.

Such studies do not make for easy reading, and their technicalities will not help the poet struggling with a particular line, since correct lines are not necessarily good lines. But some objectivity and clarity does emerge. We don't have to lay down rules in absence of the facts: we can analyze large bodies of poetry and see what actually happens. Tarjinskaya in 1976 {14}, for example, analyzed the non-dramatic iambic pentameters of 30 major poets, from Chaucer to Swinburne. The finding was that 1. ictic (rhythmical or metrical) stress features in 75 - 87% of cases, that 2. non-ictic stress occurs three times more frequently in the first position of a line than elsewhere, 3. non-ictic stress declines progressively across a line, 4. ictic non-stress occurs most commonly with monosyllables and 5. lines having more than one non-ictic stress caused by polysyllables are generally avoided. Not compelling reading, and not a revelation, but a prophylactic against any rigid imposition of rules in the New Formalist fashion. Verse types do indeed fall into normative patterns, but the patterns are derived from the analyses, not the other way round.

In their analyses, the Slavic metrists divided lines horizontally and vertically. In addition to rhythmical stress, for example, there is the part played by meaning, intonation, sound orchestration, syntax and visual form. These aspects are depicted vertically above the line, as secondary rhythmic features that are nonetheless important. Across the lines appear the sorts of findings noted by Tarjinskaya above. They give a horizontal shaping, so that a line of verse is not seen as an unvarying pattern of stressed and unstressed syllables.

Additionally, there is a hierarchical grouping, generally represented by syllable, word, phrase member, phrase and sentence (also called utterance or totality). The higher levels of syntax are also brought into the picture, so that these analyses merge with the structuralism of Levi-Strauss. How are analyses decided upon in the first place? They may be quantitative and objective, but someone has to decide on what is relevant and worth measuring, as indeed happens in all the sciences. So enters the speculative part, the theories that the evidence must support or refute. The Slavic metrists were often audacious theorists, and none more so than Ramon Jakobson. {15} In his famous definition of poetry (1960), the poetic function of language projects the principle of equivalence from the axis of selection into the axis of combination.

The idea is this. The order of words in a sentence is governed by syntax. In The dark night closed around them the syntax is quite straightforward: adjective subject noun, verb, adverb and object pronoun. We can replace the noun by another noun, an equivalent part of speech: The dark animal closed around them. Ditto for any other part of speech, or all of them together: The overpowering smell seeped into him. These replacements are in a vertical sense, along the axis of equivalence. Replacements in the horizontal sense are much more difficult. We can't write Around the night dark closed them, for instance, because words in this horizontal direction have to combine in certain ways. But we can in literary language carry out some replacements: Around them, the night was close and dark. The close night darkened around them. And so on: it is very typical of poetry, and imaginative literature generally.

Characteristics are one thing, but to make a definition of poetry in this way was preposterous. Poetry is an art, and art is governed by a good deal more than linguistics. There is no clear demarcation between literal and imaginative language, moreover, and the most compelling poems can include apparent statements of fact: witness The Divine Comedy and thousands more. A case can indeed be made for all language being metaphoric at base, so poetry and non-poetry cannot be distinguished on such a basis anyway. Jacobson's notion gained an undeserved currency in the sixties because it coincided with the obscurantism of Structuralism, and the hope that artistic creation could be placed on a scientific footing.

Jacobson did in fact elaborate the notion and provide detailed examples, but the moral should be clear. Poetry is the most subtle use of words, and the lumping together of reductionism destroys precisely what is worth distinguishing. To the extent that linguistics is a science, and science deals with propositions that must be falsifiable, linguistics does not provide a royal road to certainty in metrical analysis. The work of the Slavic metrists has been invaluable to clearing out dogma, however, and their stress on normative analysis is well worth remembering.


Verse rhythm is not wholly to be represented by simple stressed and unstressed syllables. In their seminal work of 1951, Trager and Smith {16} distinguished four levels in three elements: stress, pitch and juncture. Stress or loudness was denoted as primary, secondary, tertiary or weak. Pitch (as in music, technically the frequency of vibration of a column of air) was denoted as very high, high, mid or low. Juncture was denoted as open, sustained, rising or falling. All four levels of stress were needed to say elevator-operator, for example, and we indicate a question by raising the pitch at the end of a sentence. (I'm late. versus I'm late?). Open juncture allows us to distinguish nitrate from night rate. And so on. Other workers developed similar systems, using them to argue that 1. variation within regularity so characteristic of verse is elegantly captured by such systems, 2. verse rhythm is more easily seen an alternation of stress levels, and 3. the approach can be adapted to a normative approach: rhythm is best studied by statistical analysis of what seems acceptable to practiced readers.

Such systems can be extended vertically, moreover. Elizabeth Hewitt (1965) {17} recognized what she termed metrical foot, measure, lower cadence and upper cadence in the poetry of T.S. Eliot, an analysis only possible with pitch and juncture as elements in the grouping. The pronunciation of American English is very different from that of British English, and has brought differences in approach. Americans tend to move from segment through phrase to meaning, where British intonationalists are top down, moving from overall meaning to phrase and segment. Important to the British are tone units, which divide a text into informational segments. Each tone unit includes a prominent pitch movement (tonic syllable) that coincides with the information focus. Preceding the tonic syllable is information recoverable from context, and following is a short post-tonic element. Features making the pitch movement are tones that 1. rise, 2 fall, 3. remain level, 4. fall then rise and 5. rise then fall. British linguists also divide the tone unit into rhythmic feet, comparable to prosodic feet, where the stresses can be implied rather sounded. Where does this lead?

British prosodists have used the tone unit and the melodic contour as alternatives to simple verse measures. Most poetic lines, it is claimed, end with a tonic syllable and a tone unit boundary. Unconsciously, readers expect lines to be intonationally delimited, unified and parallelled, and these are more practicable than standard rules of versification. More particularly, this approach finds patterns inherent in the many types of free verse that break the rules but are nonetheless pleasing. William Carlos Williams' "chopped-up prose" is often abrupt and jagged, for example, but has passages of stateliness when these deeper patterns are followed.

The approaches are technical and somewhat narrow, but Kenneth Pike (1959, 1982), {18} on the American side, extended intonation into a larger study of rhythmic organization. Hierarchically, there are three 'standpoints': particles, field and wave. Any particle can have a nucleus, plus optional margins, and these margins can have a variety of roles: setting, denouement, action, etc. Each particle belongs to a class of waves in a vertical relationship with superordinate and subordinate waves. These waves in turn are associated with meaning and a field of structural and functional expectancies.

Poetic Closure

Less study has been given to movement among sentences, i.e. across the text as a whole. One exception is Barbara Herranstein Smith's (1968) {19} work on poetic closure. Poems please us because they arouse our appetites and expectations, continually defer and evade those appetites and expectations, and then finally close when the structure is coherent, complete and stable. Smith believed that such processes give us psychological or even physiological gratification, and are of two types: formal and thematic.

Formal structures arise from the physical nature of words, and include such things as rhyme, alliteration, metre and stanza shaping. Thematic structures arise from the symbolic nature of language, and include syntax, genres, sense imagery and rhetorical figures. Formal and thematic structures interact to effect closure. Formal patterning depends on repetition, and can always be extended, at least until we feel an overwhelming need for change. These terminations are not random, moreover, but emphasize a thematic revelation at the close, antithesis, puns, hyperbole, aphoristic brevity and parallels often being employed for this purpose.

Thematic structures at their most basic are unordered and expandable, and can only be closed by adding some 'paratactic listing'. These may be a paraphrase of the opening line, a reference to natural stopping places in human experience (death, sleep, heave, etc.) or some element of narrative or dramatic ordering. Poetry uses these novelistic devices, but theme dominates plot: the poem usually comments rather than simply presents or portrays. Smith in fact recognized several types of thematic structure, which she termed paractic, temporal, sequential and associative/dialectical. The last are interior monologues, the wavering of minds on trains of thought that may not reach closure.

Generative Metrists

The success of Chomsky’s generative grammar approach encouraged metrists to establish a similar system for well-formed verse.   Comprehensive rules would specify what was acceptable and what was not acceptable as verse.

Various systems were proposed. Hall and Keyser (1972) {20} demonstrated, for example, that iambic pentameters were specified by a set of correspondence rules operating on a metrical pattern of (W)SWSWSWSWS(X)(X) where ( ) indicates an optional element and X can only be occupied by an unstressed syllable. The correspondence rules are:

1. S, W or X corresponds to a single syllable or a sounded sequence of no more than two syllables.

2. Fully stressed syllables occur in all S and only in S positions, or

Fully stressed syllables occur only in S positions but not in all S positions, or

Stress maxima occur in S positions only, but not in all S positions. (A stress maximum is a fully stressed syllable occurring between two unstressed syllables in the same syntactic constituent.)

Where does that take us? Well, firstly, the rules are specified. Then the rules can establish a norm and a hierarchy of departures from the norm. Thirdly, they emphasize the importance of word boundaries since monosyllabic words have metrical freedom whereas polysyllabic words usually have their lexical break in a strong position.

And so on: this early work has been much built on and modified. Kiparsky (1975) {21} developed a concept of phonological word to examine metrical breaks that appear on syntactical and phonological grounds. (Phonological has lexical stress on a major category word plus syntactically associated unstressed syllables, a concept deriving from work of Chomsky and Halle in 1968. {22} ) It is well known, for example, that poets tend to place stressed syllables in weakly stressed positions only after strong breaks, and Dillon (1977) {23} suggested a sevenfold hierarchy of such phrasal breaks.

The drawbacks? The first is the difficulty with all generative grammars, that well-formed examples are not objectively derived. Second, the rules are not over-useful to practitioners. Like native speakers, poets construct their verses intuitively, not by following rules or blueprints. And finally, it must be admitted that disputes have frequently arise among generative metrists themselves, and that even their science has not escaped radical transformations and schisms. 

Reuven Tsur's Work

Rueven Tsur’s {24} cognitive poetics tries to fit the generative metrists approach into the larger framework of brain functioning. Literature has aesthetic effects when normal cognitive processing is interrupted or delayed.   Something complex and irregular is therefore scanned for similarity, regularity and balance, and recoded into the simple, economical patterns the brain requires. Conversely, however, input which already has these properties (has what gestalt psychologists call strong shapes) is not immediately accepted. Indeed, these strong shapes stand out from the overall complexity of other inputs, and so (in the larger framework) constitute irregular or weak shapes.   From this interaction between strong and weak arises the aesthetic phenomenon.

Some broad distinctions can be made. Convergent or conclusive styles are marked by clear-cut shapes, both in contents and structure. Divergent or suspensive styles have weak shapes: blurred, atmospheric and emotive.   Into one or other of these styles can be placed most of English poetry, with very different consequences for rhythm. Misplaced stresses are not tolerated in iambic pentameters with convergent styles, for example, but can be acceptable in divergent styles provided they affect only certain positions, usually position 7.   The brain, Tsur argues, integrates inputs into groupings on various levels based on word boundaries, syntactical breaks and pauses (caesurae). Some groupings are more common than others, and influence a good deal more than rhythm.  

Derek Attridge

Attridge {25} has united rhythmic experience with well-formedness and universals.   The focus is on what the trained reader senses, and to this end Attridge identifies an underlying rhythm consisting of a small number of abstract patterns. Such patterns are given particularity by the verse traditions in question, and can be related to the language of the text by realization rules. The underlying rhythm is a perceived temporal pattern reinforced by repetition and periodicity. (The pulses of energy that make the rhythm itself are difficult to analyze further, but seem to have both cognitive and muscular components.) Strong pulses (beats) and weaker pulses (off-beats) set up patterns of expectation, which are heightened by deviations from regularity. Beats can be sensed but unrealized, as in Donne’s  

                For I am every dead thing 

Where the offbeat between dead and thing is not realized, but adds very powerfully to the metrical tension.

Art requires shapes or groupings. In English verse the most popular groupings are four- and five-beat lines.   Four-beat lines are the most natural and popular, imparting a song-like quality. Five-beat lines are the more various, allowing the greater range of effects and are generally employed by serious poetry. Besides the underlying rhythm, however, experienced readers also recognize metrical patterns that particularize certain genres. These metrical patterns go well beyond rhythm, and include the shaping of lines, stanzas and the interplay of rhythm with content and syntax.

The theory uses a simple notation, is readily grasped, and overcomes the knotty problem of distinguishing between rising (iambic) and falling (trochaic) movement. Individual lines can be perplexing, as good line of verse commonly has both rising and falling sections. Attridge’s theory regards the distinction as unimportant, the key features lying elsewhere.

The problems? Attridge’s realization rules are complicated enough (two-base rules, deviation rules, sets of conditions) but gloss over the finer points of verse structure. Syllables are either stressed or unstressed.   Abnormally stressed syllables are called rhetorical stress, and phonetic elision is also oversimplified. Many theorists are also unhappy with the extensive use of unrealized beats, as this opens the door to regarding trimeter verse as tetrameter verse with a stress unsounded, etc.

Metrical Phonologists

Why do we say The word is misspelt, but a misspelt word? What rules make us shift the stress? One approach to understanding is to construct a metrical grid. The grid is hierarchical and includes progressively larger elements upwards: syllable, word, phrase, sentence. At each level we recognize weak and strong elements (syllable level mis spelt: word level the word is misspelt). Then we can find the rules or constraints that apply to each level of the grid. It may be, for example, that an alternation of weak and strong at one level needs to be matched by a similar alternation at a higher level.

Such was the approach of Liberman (1975), {26} and has been very influential, many workers refining and adding to the original notion. Hayes (1984) {27} added three rules to that of simple alternation. One was that four syllables should occur between stresses at some highly salient level in the grid. A second required symmetrical division at a level below that of the highly salient level. The third required a strict polarization of stresses at the highest level. Hayes also identified five layers: word, clictic group, phonological phrase, intonational phrase and utterance.

How are these rules derived? Essentially by observation and induction, the scientific approach of trial and error. But they have some linguistic basis i.e. form part of a larger set of linguistic rules and aren’t therefore purely arbitrary. Often the linguistic basis is that of generative grammar. Kiparsky (1977) {28} for instance claimed that iambic pentameters had a structure like this:

Level 5: strong
Level 4: weak strong
Level 3: weak strong
Level 2: weak strong strong weak
Level 1: weak strong weak strong weak strong weak strong weak strong.

Constraints on metrical styles could then be derived by comparing well-formed verse instances with untrammelled possibilities. Hayes himself generalized a wide set of rules that governed the interaction of phrase with metre. Bounding rules governed stress peaks within a line. Right-edge rules forbade stress clusters at the end of phrases to appear in weak positions. Left-edge rules countermanded right-edge rules and allowed clusters of stresses at the beginning of phrases.

The importance? Much of this work confirmed the theories of the Slavic metrists that the beginnings of phrases are metrically lax and the ends of phrases are metrically strict. But they did so with greater precision, laying down rules for well-formed verse.

David Gil

David Gil and coworkers at Tel Aviv University {29} underline an important aspect of the generative grammar approach, that of choosing examples of well formedness. We should not restrict ourselves to art forms, to the work of distinguished poets, they argue, but consider mankind’s general competence at verse, what they call non-canonical verse. That competence, moreover, should extend from the smallest units of rhythm to texts as a whole, including nonlinguistic matters. Like Liberman, Gill and his coworkers employ a metrical grid, with constituents at each level composed of one strong unit and one or more weak units. These constituents are given the traditional labels of iambic (I), trochaic (T), anapaestic (A), or dactylic (D). Looking at the vertical organization of the grid, the central level is the line, labelled 0. The four levels below extend to the “sub-position” or 1/16 of the line. The four levels above extend to the “bi-super-stanza” or 16 lines.   Individual constituents ( I, T, A or D) are identified by superscripts for level and position within that level.

By analyzing a good deal of material in this way, a number of unexpected features arise. Non-canonical verse operates as a continuous series of layers, with rules or constraints that control what can appear on adjacent lines. Depending on the verse type or genre, patterns appear in the syllables that can take strong or weak stresses, and there is also a patterning in the vertical sense. Most important of all, this patterning appears in the genres of other languages, and is a feature of broader linguistic features. Principles like “small precedes large” and “focus falls at the end of the unit” are well known in the phonological, intonational and narrative organization of language, but also apply to metre. Anyone who disputes the importance of metre, and most particularly its objective existence, has solid linguistic evidence to contend with.


As should be apparent from these short notes (and would be overwhelmingly so on reference to textbooks or research papers) the linguistic analysis of rhythm has become a very technical subject, divorced not only from the writing of poetry but the study of literature as a whole. Donald Wesling’s {30} contention is that we need to look at the larger purposes of language, and be content with approximate answers to important questions. Linguistic prosodists have elaborated brilliant theories, but only related sign to sign, not sign to signifier: they have failed to ask what the signs mean. Literary critics have equally avoided reality by making a fetish of metrical regularity, recasting poetry in simplistic models that do not explain why poetry was ever written in the first place.   For poetics to be useful today, we need to accept that contemporary writers are attempting to do away with stable forms, to dissolve genre and technique back into ordinary language. All aspects of a poem are now indexes of personality, a correlation between work and universe that Wesling calls “voice”.

What does that amount to? We need to examine how voice is realized:

1. In the reader’s experience of the poem.
2. In the language of the text.
3. In the relation between the two.

Wesling has a Modernist conception for the first. Poetry defamiliarizes, defacilitates and retards our normal expectations in language, and does so because aesthetic and cognitive aspects interpenetrate.

He bases much of these contentions on the theoretical work of Jan Meijer (1973) {31}, who recognized specific aesthetic and cognitive structures. The first create relations of equivalence and work towards self-perpetuation. The second create unequal relations of dependence and work towards closure. From the mutual interference of the two structures arises the artistic tension and the poet’s individual voice.

And the language of the text? Wesling proposes a grid marking the intersections of syntax and a poem’s structure.   One axis of the grid has morpheme, word, phrase, clause, sentence and group of sentences.   The other has syllable, foot, part-line, line, stanza and whole poem.   The size of a circle drawn at each intersection can show its importance: an approach Wesling calls grammetrics.

How is that importance assessed? For an answer to question 3 above, Wesling uses the work of the Czech theorist Jiri Levy, who pointed to morphological analogies between prosodic forms and semantic effects. The linear flow in poetry, for example, follows three formative principles: continuity-discontinuity, equivalence-hierarchy and regularity-irregularity. Each plays its part in the formal aspects of poetry (pauses, intonation, rhyme, repetition and rhythm) but also impacts on the semantics.   Each intersection can be used to form hypotheses about higher level interpretation (tone of voice, point of view) so that stylistic concepts can be loaded with semantic possibilities.

Henri Meschonnic

Henri Meschonnic {32} also takes a broad, not to say speculative, view of rhythm. His Critique du Rythme (1982) encompasses language, society and the individual, and in much of its considerable bulk Meschonnic attempts to show how narrow, vague and wrong-headed are conventional views. Rhythm is not binary, periodic or repetitive, but pervades all discourse. Rhythm is not valued for its mnemonic properties, but remembered because it is valuable. Rhythm is not decoration but a constituent of meaning. The essence of rhythm is form, not movement, as an examination of its etymology demonstrates.

What is Meschonnic’s definition of rhythm? He doesn’t really say, except to claim that rhythm is the most archaic part of language and human subjectivity, all those aspects of discourse that do not include signification. Rhythm is fragmentary and diverse, moreover, making definitions impracticable and unneeded.

Examples? Meschonnic postulates three sorts of rhythm: linguistic rhythms, rhetorical rhythms and poetic rhythms. Unfortunately, his Critique elaborates only the first, which does seem to involve just those matters that escape rigorous linguistic description. Here belong paralinguistic gesture, visual form, sound orchestration, metaphor, ambiguity, choriambic patterns and a host of others. Meschonnic’s normalization of this category is standard but very elementary, its purpose being largely to demonstrate the inadequacy of traditional scansion.

Richard Cureton

Rhythmic Phrasing in English Verse, Richard Cureton’s important book of 1992 {33}, brought together contemporary thought on verse rhythms, but did so in extensive and carefully thought out schemes. The inspiration was Jackendoff and Lerdahl’s  (1983) treatment of rhythm in western tonal music, which recognized three distinct but interacting components: a cyclic beating (metre), peaks of structural salience (grouping) and arrival at expected goals (prolongation). Musical theorists had argued that what we call rhythm is the human ability to combine complex and diverse inputs into relatively simple patterns, and Cureton was equally determined not to oversimplify matters. Metre certainly existed, and consisted of duple or triple patterns.  Grouping extracted more irregular shapes from structures inherent in the medium, and ordered these shapes in a hierarchical fashion over all levels. Prolongation was the reader’s expectation of where the text was heading, and was therefore more general and subjective. Metre, grouping and prolongation were a feature of rhythm, but all three were not required for rhythm to exist.

An illustration may help. Cureton uses an extensive bracketing and tree notation, but here is a flavour of the metrical system, rather freely transposed into a scheme for web pages, the bold type indicating a beat. The text is from William Carlos William’s Paterson:


Level 5   Without invention nothing is well spaced

Level 4   Without invention nothing is well spaced

Level 3   Without invention nothing is well spaced

Level 2   Without invention nothing is well spaced

Level 1   Without invention nothing is well spaced


Level 10 Without invention nothing is well spaced

Level 4   Without invention | nothing is well spaced

Level 3   Without invention |   nothing |   is well spaced

Word      Without invention nothing is well spaced

Syllable  With out in ven tion no thing is well spaced

Readers will have to consult Cureton’s book for a proper treatment. His scheme is relatively compact and simple, but daunting to those unfamiliar with linguistic notation. Moreover, by giving emphasis to grouping, Cureton’s 1992 book rather neglected prolongation and metre, though work in progress (A Temporal Theory of Poetic Rhythm {34}) apparently corrects the imbalance. Language, literary convention and historical content are also to be brought into the new system, which is indeed the full set of metrical preference rules detailing how prosody and versification affect our metrical reading of a text. A fourth rhythmic component is also added (theme), and Cureton attempts wider correlations with cultural history, neurobiology, ethics and so forth.


Are we home? Probably not. Cureton’s may be the most convincing and comprehensive treatment we have of rhythm in English verse, but it is worth dwelling on the assumptions and limitations.

1. Does a grouping of vertically arranged layers of progressively larger units really exist? Do the layers relate to realities of consciousness or brain processing, or are they only the artifacts of a convenient notation? Chomsky’s grammar is vexed by the same questions, and they are not empty sophistry. Chomsky and Cureton have rather a computer-processing view of the brain, which is certainly open to question

2. Is Cureton’s scheme able to distinguish the good line of verse from the merely correct? No. It is not intended to. The scheme simply aims to establish the rules to comprehensively distinguish between what is acceptable and what is not acceptable as verse. Who then makes the correct reading in the first place, which Cureton objectifies in comprehensive rules? Who indeed? Academia is not noted for consensus.

3.  Does metre apply to the same phenomenon in different cultures? Is the Chinese experience of metre in their verse (which is very strong) actually the same as ours? We can’t be sure, and experts in fact disagree.

What attitude should we adopt to this brief survey of a very technical and disputatious field? Some patience and honest doubt. Rhythm is not an easy matter, and we should treat with as much caution the proselytizing of the free-verse movement as we do the simplistic rules of the New Formalists. More important than correctness is excellence, and that is only acquired by a deep love and knowledge of poetry. Perhaps, to be reactionary, we could adopt what poetry lovers of earlier generations recommended: recitation, the speaking of verse as an art form. That art has now made an encouraging return in the many recordings now available on CDs or over the Internet.


1. Chapter 1 of Derek Attridge's The Rhythms of English Poetry (1982) and Annie Finch's The Ghost of Meter (1993).
2. See articles in Preminger and Brogan's The New Princeton Encyclopedia of Poetry and Poetics (1993).
3. TVF Brogan's Meter in The New Princeton Encyclopedia of Poetry and Poetics (1993).
4. Section 1.3.5 in Richard Cureton's Rhythmic Phrasing in English Verse ( 1992).
5. Section 1.3.1 of Richard Cureton's Rhythmic Phrasing in English Verse ( 1992) and Section 1.1 of Derek Attridge's The Rhythms of English Poetry (1982). Succeeding sections are modelled on Richard Cureton's work.
6. See section 1.3.2 of Richard Cureton for references.
7. See section 1.3.3 of Richard Cureton for references.
8. See section 1.3.4 of Richard Cureton for references.
9. M. Croll's The cadence of English oratorical prose in Patrick and Evan's (Ed.) Style, Rhetoric and Rhythm (1919).
10. Roland Barthe's S/Z (1974).
11. See section 1.3.5 of Richard Cureton for references.
12. Eleanor Berry's Visual Poetry in The New Princeton Encyclopedia of Poetry and Poetics (1993).
13. See section 1.3.6 of Richard Cureton for references.
14. M. Tarjinskaya's English Verse: Theory and History (1976).
15. Ramon Jakobson's Linguistics and Poetics:1960 in Jakobsons's Language and Literature (1987).
16. G.L. Trager and H.L. Smith's An Outline of English Structure (1951).
17. Elizabeth Hewitt's Structure and Meaning in T.S. Eliot's 'Ash Wednesday' (1965).
18. Kenneth Pike's Language as Particle, Wave and Field (1959) and Linguistic Concepts: An Introduction to Tagmemics (1982).
19. Barbara Harranstein Smith's Poetic Closure: A Study of How Poems End (1968).
20. M. Halle and S.J. Keyser The iambic pentameter in W.K. Wimsatt's (Ed.) Versification: Major Language Types (1972).
21. P. Kiparski's Stress, syntax and meter in Language: 51(1975).
22. N. Chomsky and M. Halle The Sound Pattern of English (1968).
23. G. Dillon's Kames and Kiparsky on syntactic boundaries in Language and Style 10 (1977).
24. References in Richard Cureton for bibliography on Rueven Tsur.
25. Derek Attridge's The Rhythms of English Poetry (1982).
26. M. Liberman's The intonational system of English (1975).
27. P. Kiparsky's The rhythmic structure of English verse in Linguistic Inquiry 8 (1977).
28. B. Hayes' The phonology of rhythm in English in Linguistic Inquiry 15 (1984).
29. See references in Richard Cureton for bibliography on David Gil and co-workers.
30. See references in Richard Cureton for bibliography on Donald Wesling.
31. Jan Meijer's Verbal art as interference between a cognitive and aesthetic structure in Van der Eng and Grygar's (Eds.) Structure of Texts and Semiotics of Culture (1973).
32. Henri Meschonnic's Critique du Rythme: Anthropologie Historique du Langage (1982).
33. Richard Cureton's Rhythmic Phrasing in English Verse (1992).

Internet Resources

1. The Ghost of Meter: Culture and Prosody in American Free Verse by Annie Finch. 1994. http://www.users.muohio.edu/finchar/criticism/
. Review by Timothy Morris noting the formal elements in free verse.
2. Poetry: Meter, Form, and Rhythm. H.T. Kirby-Smith. Nov. 2001. http://www.uncg.edu/%7Ehtkirbys/. Instructional programs online in the meters and forms of poetry.
3. The Neural Lyre: Poetic Meter, the Brain, and Time. Frederick Turner and Ernst Pöppel. Oct. 2001. http://www.cosmoetica.com/B22-FT2.htm. Speculative article on the larger implications of structured verse.
4. Rhythm and Linguistic Form: Toward a Temporal Theory of Poetic Language. Richard D. Cureton. 1997. http://depts.washington.edu/versif/resources/
. The element of time in stylistics.
5. Milton’s Metrical Development. 1921. http://www.bartleby.com/218/0908.html#note7. Note in The Cambridge History of English and American Literature.
6. A Review of 'Rethinking Meter: A New Approach to the Verse Line' by Alan Holder. Oct. 1997. http://depts.washington.edu/versif/backissues/vol1/
. Critical review by G. Burns Cooper of Holder's phrasalist approach.
7. On the Difference between Verse and Prose. Arthur Quiller-Couch. 1916. http://www.bartleby.com/190/3.html. An older view, distinguishing clearly between them.
8. 'The Tunnel: Selected Poems' by Russell Edson. http://www.webdelsol.com/tpp/mm2-tpp.htm. Review of Edson's prose poems by Morton Marcus.
9. An Interview with David Antin. Charles Bernstein. 2000. http://www.centerforbookculture.org/interviews/
. Long interview in which Antin explains his approach to speech rhythms in his work.
10. 'The Granite Butterfly. A Poem in Nine Cantos' by Parker Tyler. 1994. http://wings.buffalo.edu/epc/authors/perloff/parker.html. Marjorie Perloff's review examining Tyler's free verse techniques.
11. Thoughts on Rexroth's Prosody. Bradford Morrow. 1984. http://www.english.uiuc.edu/maps/poets/m_r/
. Analysis of his free verse.
12. Sounds of Poetry. Ismail Talib. Jul. 2000. http://courses.nus.edu.sg/course/ellibst/lsl05.html. Halliday's phonometric approach.
13. Cognitive Poetics Project. Reuven Tsur. http://www.tau.ac.il/~tsurxx/. Several articles of interest on Tsur's homesite.
14. Poetic Rhythm: Structure and Performance An Empirical Study in Cognitive Poetics. Reuven Tsur. Oct. 1997. http://www2.bc.edu/~richarad/lcb/wip/rt.html. Outline of forthcoming book, surveying field and mentioning Wellek, Halle and Keyser, etc.
15. Interactive Tutorial On Rhythm Analysis. Ellen Stauder. 2000. http://academic.reed.edu/english/intra/index.html#TOC. Uses Java applets: approach of Attridge and Richard Cureton.
16. Review: Liberman: Speech - A special Code. Jun. 1997. http://www.linguistlist.org/issues/8/8-954.html. Some background to Liberman's work: technical.
17. 'The Scissors of Meter: Grammetrics and Reading' by Donald Wessling. Jun. 1997. http://depts.washington.edu/versif/backissues/
. Review by Brennan O'Donnell stressing that metrics is a branch of cognition.
18. Temporal Theory and Poetics. Richard Cureton. http://www2.bc.edu/~richarad/lcb/wip/rc.html. Abstracts of three works in progress.
19. Accentual Verse. Dana Gioia. 2001. http://www.danagioia.net/essays/eaccentual.htm. One of many sensible essays on this site.
20. Versification: an Electronic Journal of Literary Prosody. http://depts.washington.edu/versif/. Articles, reviews and short listing: possibly not continued beyond 1998.
21. Ancient Rhythmicians and Modern Prosodists: Searching for the Location of Meter. Steven J. Willett. 1997. http://depts.washington.edu/versif/resources/
. Differences between metre and music.
22. A Disciplinary Map for Verse Study. Richard D. Cureton. 1997. http://www.depts.washington.edu/versif/backissues/
. Different ways of looking and accounting for verse.
23. Little boxes: the effects of the stanza on poetic narrative. Catherine Addison. 2003. http://www.findarticles.com/cf_dls/m2342/2_37/
. 17 page article on many aspects of poetic organization.
24. Prosice: A Spoken English Database for Prosody Research. Mark Huckvale and Alex Chengyu Fang. http://www.phon.ucl.ac.uk/home/mark/papers/prosice.pdf. Details of what is needed and available.
25. Linguistics and Poetics Revisited: Response. Derek Attridge. 1997. http://depts.washington.edu/versif/resources/papers/
. A look at current difficulties.
26. The Art of English Poetry by Edward Bysshe. 1702. http://etext.lib.virginia.edu/toc/modeng/public/BysEngl.html. Electronic version of classic text: exhibits from older poets but still helpful.


C. John Holcombe   |  About the Author    | ©     2007 2012 2013 2015.   Material can be freely used for non-commercial purposes if properly referenced.