Introduction: linking metre and meaning
A new system of metre notation (scansion)
The method in brief
Illustrations: Heaney, Donne, Wordsworth, Keats, Frost,      Shakespeare, Eliot, Hardy

Generative metrics
References a

See also the page

Introduction to {theme} theory

Introduction: linking metre and meaning

Musicality is the central, distinctive challenge for the poet. The distinctive musicality of a poem is achieved by metre rather than diction. The music of Tennyson's poetry owes more to his metrical skill than to his poetic use of poetic words. The forms taken by musicality are very varied: sensuous, attractive musicality and harsh, jagged, dissonant musicality, light musicality and intensely serious musicality, highly regular forms and very loose or highly irregular forms,just as musical musicality takes such very different forms as the polkas of Johann Strauss, the uncompromising Grosse Fuge of Beethoven, the grace of the Mozart concertos for solo violin and the profound music of the Sinfonia Concertante for violin and viola, Bach's 'Die Kunst der Fuge' and pieces whose architecture is hard to detect. All good poets have to be, to some extent, poet-musicians. I've argued at length, with evidence, that the poet Seamus Heaney, for all his achievement, was generally very weak as a poet-musician.

It's widely recognized that simple methods of examining metre and notating metre often amount to a serious distortion. A reader examines a line and may well find that there's a pattern of unstressed-stressed syllables with five stressed syllables in each line, and so the line is labelled, correctly, as an 'iambic pentameter.' But this traditional approach often claims stress for syllables which aren't stressed at all and regards syllables as either stressed or not, ignoring gradations of stress which are very significant. It also ignores the effect of pauses, ignores phrasing, which often make a crucial contribution to the structure and artistry of the poem, and above all separates meaning from metre. This is a failure too in such modern approaches as Generative Metrics and Bracketed Grid Theory.

Traditional analysis applied to this phrase from Book 1 of Wordsworth's 'The Prelude' (lines 425 - 6, 1805 version)

And in the frosty season, when the sun
was set,

focuses attention on the complete line 'And in the frosty season, when the sun' in isolation from the complete phrase. It ignores the importance of the term 'sun was set' which is carried over to the next line.

Simple-minded analysis, identifying, correctly, this line later in the same passage (line 446) as iambic

The orange sky of evening died away.

goes on to allocate strong and undifferentiated stresses mechanically from beginning to end of the line, giving the impression that the line ends decisively. It's obvious, though, that there's a gradient of stress: this is a line with diminuendo. Meaning - 'dying away' - is linked with the dying away of stress.

Traditional methods are very useful and convenient and, used well, they can help to make clearer the artistic success (or failure) of lines of poetry. They can be supplemented by separate studies of pauses, phrases and meaning - such aspects as the imagery of the line. Sensitive traditional analysts know very well that the metre of a poem is a thing of infinite subtlety, but although they may be very skilled at applying corrective interpretation, it's better to use more artistic - and realistic - methods at the outset.

The first stage in the study of the lines in my approach is to identify the primary sources of emphasis in the line. In a system based on stress, applicable to almost all  English poetry, these are (1) primary stress-emphasis (2) primary meaning-emphasis. This can include imagery, but in general, words with semantic force.  These may be prepositions and words such as 'not'  as well as language which is concrete, vivid, muscular, sensuous.

My approach makes use of new ideas, such as metre enjambment (this occurs in 'when the sun / was set') metrical inertia, perturbation and different stress volumes.

In the study of metre, attention has been focused on the periodicity of metre, the alternation of stressed and unstressed syllables. This is obviously the essence of metre, but there are forces which tend to resist the alternation too. There's also a tendency for strong stress to be continued and rapid light syllables to continue. 'Metrical inertia' is the counterpart of inertia in Physics, which, omitting any detail, is the property which causes a mass to resist changes - so, a body in motion tends to continue in motion. Metrical inertia, like inertia in Physics, has no associations of sluggishness.

Perturbation of metre is caused by close and vivid images and in general by the use of language with semantic force or significance. (In Physics, 'perturbation' is a secondary influence which brings about {modification} of simple behaviour. For example, the trajectories of comets are perturbed when they pass close to massive bodies, such as the planets of our solar system.)

I explain 'stress-volume,' which is concerned with the overall stress of lines, in the section on metre-notation.

I use established concepts as well, above all beats and offbeats, which are different from stressed and unstressed syllables. This distinction, widely used but not used widely enough, is very important. Often, stressed syllables in a wrongly applied traditional approach turn out to be wrongly stressed beats.

A system of metre notation ('scansion') should be able to show more than basic stressed and unstressed syllables. In the section 'A new system of metre notation,' I discuss the system I use. It can show degrees of stress, is very easy to implement and has other advantages.

No system of metre notation can show all nuances and no method of analysis can explain all of them. I make use of 'approximations,' beginning with metrical distinctions which are basic and surely beyond dispute. Image-terms and primary stresses play an important part in these first approximations. Then there are further approximations which can often be shown by the notation, and which concern matters of metre and meaning which are also beyond dispute, or practically beyond dispute. Then I proceed to matters which are disputable, which concern strikingly varied - or subtly varied - interpretations and performances of the lines.

A new system of metre notation (scansion)

The most common way of showing metre notation (scansion) is symbolic: the use of symbols such as x or w to indicate an unstressed syllable and / or s to indicate a stressed syllable. This has disadvantages: the difficulty and inconvenience of placing the symbols above the line when using a computer text program, even one with graphics facilities, and the fact that this system is binary, incapable of showing degrees of stress.

The use of capital letters to show the stressed syllables overcomes the difficulty of placing the syllables above the line of poetry and is very easy to implement but again, the notation is binary. It's also obtrusive. To show gradations and degrees of stress, some analysts use a musical notation. This has the disadvantage of requiring a specialist program and the disadvantage that music and poetry have many differences in the stresses to be notated.

I make use of the gradations in the appearance of text which can easily be implemented in Web creation programs such as Dreamweaver and word-processing programs, a form of text-emphasis. These gradations include very faint print, faint print, 'default print' and bold print. To show the simple binary alternatives of unstressed and stressed syllables, faint print and default print or faint print and bold print need only be used, as in this example of iambic rhythm, the first two lines of Blake's 'The Divine Image,' from 'Songs of Innocence.

Faint print and default print:

To Mercy Pity Peace and Love,

Faint print and bold print:

All pray in their distress:

This shows gradations of metrical stress by gradations of text-emphasis, from faint to bold print:

tell tell tell tell

My system can indicate as many degrees of stress as are shown in Otto Jesperson's system, which uses the numbers 1 - 4 to show increasing degrees of stress. ('Notes on Metre,' 1900.) It's important that a notation should be able to show the degrees of stress revealed in this example, analyzed according to Bracketed Grid Theory (which I don't explain further).

For most purposes, though, I notate three stress levels, shown by the use of faint print, default print and bold print.

To notate four stress levels, I make use of a different font in bold print for the strongest stress level, for example rapture.

To indicate the metre independently of the words, I make use of these symbols. To show three stress levels,

stress level 1, weakest

stress level 2, intermediate

stress level 3, strongest

as in the line I discuss below:

It is December in Wicklow

Implementation: three gif images are created. This takes very little time. These images can easily be added to text publications, using a word processing program, or a Web site, using a program such as Dreamweaver. It's most convenient to add the images to the Library. They can then be added by simply dragging and dropping.

I use a different method to show wide {distance} in page-space of stresses and meanings. In these lines, lines 455 - 457 of Book 1 of Wordsworth's 'The Prelude' (1805 version) it should be clear that from the word 'happy' to the word 'rapture' no stresses are anywhere near as strong or as important as the strong stresses within these two words. These are primary stresses. The meaning-linkages between the two words are overwhelmingly important too, emphasized by the strong stresses.

... (2) happy time
It was indeed for all of us, to me
It was a time of (1) rapture. Clear and loud
The village clock tolled six ...

Yet sophisticated approaches to metre as well as simple traditional approaches ignore wider {distance}. The whole concern is with shorter {distance}, the linkages and contrasts in stress of syllables near to each other. It's widely recognized that non-free verse isn't usually completely regular or systematic in its stresses, but unsystematic stresses such as this example aren't taken into account.

I use a numerical notation to show stress linkages and relative stresses over a wide {distance}. The numbers, in brackets, are placed immediately before the syllables to which they refer, if necessary within a word. (1), (2), (3) ... show decreasing stress levels. If there's more than one set of stress-linkages and contrasts at {distance}, then I use different fonts, such as (1), (2), (3) ... in different sizes, as in this same passage, in connection with 'all' and 'me,' which are in the same line but at a significant {distance}, another unsystematic use of stresses:

... (2) happy time
It was indeed for (1) all of us, to (2)
It was a time of (1) rapture. Clear and loud
The village clock tolled six ...

Here, the stresses shown by (1) and (2) are linked and are distinguished from the stresses shown as linked by (1) and (2), the two sets being in a different font and size.

Stresses in poem-lines can be shown by using notation-lines. This system is used by Alan Holder in 'Rethinking Meter: a New Approach to the Verse Line,' but I use it for a very different purpose, to show volume.

My poem 'To a Lyric Poet' begins at high volume:

Useless - less - unless distress
drives or drove your work.

If three stress levels are notated, then there are three notation-lines, as here, showing upper stress, middle stress and lower stress. The syllables are distributed on the notation-lines according to stress. The higher the placing, the greater the stress.

The lowest notation-line is unoccupied, even by 'or.' The syllables of a poem at low volume would occupy completely or almost completely the lower two notation-lines. If four lines are used, a poem at middle volume would occupy the middle two lines.

An analogy with volume: the mountains of the English Lake District differ in their height - all their heights are very modest by the standards of many mountain ranges - but all of them start from a low altitude, not much greater than sea level. In other parts of the world, many mountains, also differing in height, of course, start from a much higher level. This shown by the lower and upper lines here:

Overall differences in volume are unaccounted for in sophisticated as well as simple approaches to metre. They are concerned only with local differences in syllable stress.

Lines also differ in their stress uniformity. Poems with greater stress uniformity occupy two out of three levels, if three levels are notated, completely or almost completely. Poems with less stress uniformity occupy all three levels. By considering more than three levels we can study more effectively lines which show a great degree of stress non-uniformity.

Whether the lines have a high degree of stress uniformity, like the lines 'Useless - less - unless distress \ drives or drove your work' or a low degree of stress uniformity, the pattern frequently continues for some distance, rather than being replaced by another pattern. Driving, emphatic rhythms and the lack of marked rhythm, at both high and low volume, may continue. This is 'poetic momentum.'

I show phrase boundaries using square brackets, [ ]. If it's necessary to show shorter or longer pauses between phrases, then the distance between brackets is increased. (In a Web creation program, which allows only one fixed space, the easiest way of implementing this is to insert more or fewer characters, the colour set to the colour of the background. The characters are then invisible.) Examples:

[phrase 1] [phrase 2]

[phrase 3] a[phrase 4]

[phrase 5] aa[phrase 6].

When phrasing isn't indicated but it's important to show pauses (these can be regarded as positions in the lines which are not filled with syllables with a degree of stress) then I use ‚ and for longer pauses ‚‚ and ‚‚‚. An example, discussed in detail below:

To be, or not to be, that is the question:

I explain 'image-term' in detail below. An image-term has 'semantic force' or 'semantic significance.' It includes words which are vivid or sensuous but also words such as 'not' and the demonstrative 'that,' when they signify an important and sometimes crucial meaning. When it's important to distinguish the image-terms, or the main image-terms, I do this by underlining. The symbol for an image-term, given in isolation from words, is _ in my notation. So, three image-terms in a sequence are shown as _ _ _ .

I use \ to indicate a line break, as in 'To mercy pity peace and love, \ all pray in their distress. Although I never make use of / to indicate strong emphasis, in accounts which do there's an obvious risk of confusion with the symbol for line break, in the established notation also /.

The method in brief

This method integrates metre and meaning. In this section, I simply describe the method, very concisely. I don't give illustrative examples - I provide these in the next section, which for that reason may well be found more approachable than this summary.

Stage 1

The order 1a followed by 1b can be reversed. In the notation of my {theme} theory, « :- Ô.

1a. Identify the primary metrical stresses in the line, if any, or all the metrical stresses. (The 'stresses' are obviously 'emphases.') It may be that all the metrical stresses can be heard and are more or less identical in their force. If some metrical stresses are very prominent and the others far less prominent, or can't be heard at all, then by deliberate distortion these can often be brought out. This is one way of establishing the beats, the syllables which are stressed to a greater or lesser extent or which aren't actually stressed but seem to belong to a pattern. The other syllables are the offbeats.

1b. Identify the primary meaning emphases in the line, if any, or all the meaning emphases. I use image-terms for the primary meaning emphases. Image-terms have particular semantic force or semantic significance (the link is to the entry in my glossary: poetry.) This includes the vivid, sensuous language given excessive, if understandable, emphasis by so many commentators on poetry. If there are no obvious image-terms, distinguish the different contributions to meaning of nouns, verbs and adjectives from other grammatical categories, such as prepositions. The different contributions are usually very distinctive, but allow for the fact that prepositions, for example, can have semantic force, as in this line from my poem Sailing from Belfast, at the time of the Troubles.

I climb up and down unstable stairs.

Stage 2

Establish a linkage between stress and meaning for the primary instances and then for all the instances. Image-terms are treated as a unit with the metre of the image-terms also treated as a unit. What is outside the image-term is contrasted in meaning and metre.

Stage 3

Examine the line or lines for phrases and phrase boundaries. There will be pauses at the phrase boundaries. Examine the line or lines for any other pauses.

Stage 4

If you're analyzing more than one line, turn your attention to the successor lines. Note any enjambments, such as phrase enjambment, image-term enjambment and metre enjambment.

The traditional notion of feet, such as iambs and trochees, is still useful, I think. It focusses our attention on the alternation of weak and strong stresses which is the essence of poetic meter in most English poetry. But I emphasize too cross-linkages, such as the metrical linkage of syllables within an image-term, contrasted with the metrical linkage of syllables outside an image-term. Two syllables which belong to the same 'foot' may be contrasted - one has a linkage with other syllables within the image-term it belongs to, the other is outside the image-term and has a linkage with other syllables outside the image-term.

Often, it's not only the most obvious linkages which are important. Elements which are linked together in an obvious way may be contrasted in other ways. Another example to illustrate this concept of cross-linkage. The matters studied in the branch of logic called 'fuzzy logic' have linkages with matters studied in other branches of logic. For example, nothing in logic is tested for truth or falsity by observation or experiment. Science, on the other hand, depends upon observation or experiment to test and often falsify theories. But matters of fuzzy logic are contrasted with other branches of logic and linked with science by having a degree of vagueness (which amounts to a technical term in philosophy.)


Seamus Heaney: Exposure, from North

This is the first line of the poem, indicating the two stressed syllables of 'December' and 'Wicklow' by the use of bold print, the other syllables in these words by default print and the syllables 'it,' 'is' and 'in' by faint print. I distinguish, then, three stress levels here.

It is December in Wicklow:

Using symbols:

It's recognized that a study of the metrical emphases in lines of poetry amounts to isolation, although very often necessary isolation. There are other possible emphases in a line, such as the use of a vivid image, and the use in general of words which have, in my terminology, semantic force and semantic significance.

The word 'December' has many possible associations. Here, it refers not to, say, the Russian December, but the far less extreme Irish December. Some background knowledge is needed to appreciate the use of the word here. 'Wicklow' will have rich associations for some readers but not for the majority. As a place name, it has concreteness and at least the stimulus of curiosity: what sort of a place is Wicklow? Both words have semantic emphasis.

These two words, in their different ways, have far greater emphasis than 'it,' 'is' or 'in.' This can be shown by the use of colour-emphasis, which shows the semantic emphasis of complete words but the greater metrical emphasis of an accented syllable, the lesser emphasis of an unaccented syllable, can easily be shown as well:

It is December in Wicklow.

However, to accommodate readers with impaired colour vision, and for use in monochrome print publications, it's better to use a different method. I use underlining-emphasis:

It is December in Wicklow.

'December' and 'Wicklow' here are examples of what I call image-terms. Image-terms are instances of semantic emphasis or semantic significance, just as stressed syllables are instances of metre-emphasis. 'Image' is given a new meaning here (although often it has a linkage with established literary meanings): something with semantic force or semantic significance. In mathematics, a 'term' is an expression forming a separable part of a wider expression. An image-term forms a separable part of the line.

'December' and 'Wicklow' are image-terms, then, and the linkage of the syllables making up these units is given {pre-ordering}. The syllables outside may be linked, as in the case of 'It is' or may be isolated, as in the case of 'in.'

There's an increase in stress with the first syllable of 'December' but only with the fourth syllable of the poem do we have emphatic stress.

Derek Attridge's 'Promotion Rule' applied to a sequence of three unstressed syllables doesn't have the effect of stressing any of the syllables (the name 'promotion' isn't well chosen) but is used to explain the speeding up of tempo. I refer instead to a 'triplet' of syllables, with a linkage with the triplet of music. Three syllables take up more or less the time of two syllables. They are pronounced more rapidly. However, I treat the first two syllables with weak emphasis, 'it is' as the unit: These form a 'doublet.' The 'Promotion Rule' appeals only to metre. My own explanation is in terms of momentum and meaning. Three unstressed syllables have high metrical momentum. Once this group of syllables is set into rapid motion, then it has a tendency to continue in rapid motion. As the syllables are short, it's easy to set them into motion. Their contribution to the meaning is less weighty too. 'It is' can be omitted. The poem could have begun simply, 'December in Wicklow.'

There are three weak accents before the strong accent on the second syllable of 'December.' A very wide ((survey)) of poem openings shows the rarity of an opening which delays for so long the first strong accent of the poem. The accents in the first two bars of Beethoven's Pastoral Symphony - the theme played by the first violins - are the same as in the first line of 'Exposure.' In the Symphony, the first theme begins after a quaver rest and the main stress is delayed until the fourth note, after the bar line.

Here, as always, musical notation can't convey the varied possibilities of performance. The nuances of performances are almost limitless. The notation I use for metrical analysis of poetry could be used by composers to give greater guidance to performers concerning nuances of emphasis. Musical notes could be printed with gradations from faint to bold.

John Donne: A nocturnall upon S. Lucies day, Being the shortest day

This, like 'Exposure,' is a 'December poem,' linking diminished light with the feeling of weakness, although Seamus Heaney's poem also mentions the dispiriting effects of rain and cold.

The first clause is : Tis the yeares midnight,

'Midnight' has very strong associations and is given great semantic emphasis.

Tis the yeares midnight,

'yeares' has great emphasis too, brought about by extension-contrast. Not only is a year very extended by comparison with 'midnight,' the word itself can be extended to a greater extent than 'midnight.' The second vowel of 'midnight' can be extended slightly, but to nothing like the extent of the vowels in 'yeares.' 'Yeares,' then, has its own emphasis, which may be more or less forceful than those of 'midnight' for different readers:

Tis the yeares midnight,

The two words 'yeares' and 'midnight' can be considered as forming a compound term with semantic emphasis, even if the reasons why the words have semantic emphasis are very different. They can be included in a compound term, with underlining of the whole term:

Tis the yeares midnight,

'Tis' has greater stress than the opening 'It is' of Seamus Heaney's 'Exposure.' 'It is' is a quiet and uneventful opening, 'Tis' more emphatic, but in the opening of 'A nocturnall...' the effect of the two simple image-terms or the compound image-term is {restriction} :- metre-emphasis for 'Tis' and 'the.' (:- is read as 'applied to.' 'metre-emphasis' is here the application-sphere for the theme {restriction}).

This shows the stresses in the light of meaning:

Tis the yeares midnight,

Later in the poem, we can mark other terms with semantic emphasis, simple and compound. The stresses are clear, too, and can easily be added. (I omit in this first approximation the differing stresses in ' ... and I am.'

...and I am re-begot
Of absence, darkness, death, things which are not.

Here, there is the powerful piling up of image terms, the lack - apart from the word 'of,' of any words not part of this sequence of image-terms. Image terms may be proximate or not. They may be very dispersed. There may be a rhythm of image-terms and non-image-terms which has a linkage with the rhythm of stressed and unstressed syllables.

Words which are ordinarily without semantic force in a poem may show the powerful effect of a word with semantic force and becoming incorporated into a compound image term. The word 'things' is ordinarily so general as to lack semantic force, and 'which' and 'are' lack semantic force too. 'not' very often does have great semantic force, as here, its effect intensified by placement at the end of the line. It exerts a perturbing force on these proximate words.

Some word classes are far more likely to have semantic emphasis than others. I use 'word classes' in preference to 'parts of speech' or 'lexical categories.' The exact number of word classes in English varies according to the commentator. The word classes I take account of are: noun, verb, adjective, conjunction, determiner (eg 'a,' 'the' and 'my'), preposition and pronoun.

Nouns, verbs and adjectives are far more likely to have semantic emphasis in a poem than members of the other word classes. There are exceptions, of course. Parts of the verb 'to have' or 'to be' are more often than not without semantic emphasis, as in the opening of Seamus Heaney's 'Exposure.'

Stress-emphasis in general intensifies the effect of image-emphasis. Terms with semantic emphasis which are already powerful may have their power intensified so much by one, or more than one, strong stress within them that they dominate their lines, reducing other words and other strong stresses, not to insignificance but to something muted. Metre is often akin to an orderly progression but it may be a matter more of intense fragments. This is perturbation of the lines.

Wordsworth: The Prelude, Book 1

These lines, 452 - 457 (1805 version), are about the ecstatic excitement of skating on a frozen lake, Esthwaite Water. Just as poetic intensity may disrupt ordinary syntax, it may also disrupt ordinary metre. I use 'perturbation,' for the disruption of syntax as well as metre, even though I focus attention on perturbation of metre here.

This is a first approximation. I only indicate below what seem to me the main image-terms and I only indicate the metrical stresses within these image-terms. See also my sectional analysis of these same lines.

And in the frosty season, when the sun
was set
, and visible for many a mile
The cottage windows through the twilight blazed,
I heeded not the summons; (2) happy time
it was indeed for all of us, to me
it was a time of (1) rapture...

In the first syllable of 'rapture,' metrical stress and poetic intensity are at its highest.

I next discuss the new concept of metre enjambment, using as the illustrative example these lines from 'The Prelude:'

In a traditional analysis in terms of feet, the weak and strong stresses of 'when the sun' belong together, as do the weak and strong stress 'was set' in the next line. In my analysis, the image-term 'sun was set' links stresses in a different way: 'sun
was set
' and what is significant is the metrical group within the image-term, strong-weak-strong. 'sun' is linked by metre with 'was set,' not, as in the traditional analysis, with 'when the.' The metrical unit making up the image term is carried over to the next line.

The established term 'enjambment' includes what I refer to as 'line-enjambment' and 'stanza-enjambment.' This is a classification which refers to the boundary which is crossed, the end of a line or the end of a stanza. My classification also refers to the material which is carried over. In the case of 'metrical unit line-enjambment,' for example, there's carrying over from one line to another of a metrical unit. All the references to enjambment on this page are to line-enjambment, so I can simply use 'metrical unit enjambment.' A short list showing some of the things which may be carried over:

Metrical unit enjambment
Image-term enjambment
Sentence enjambment
Phrase enjambment

The last of these is important in the passage from 'The Prelude.' Above, I give the link with my discussion of Sectional Analysis on another page, but for convenience, I quote here the lines discussed.

452 [And in the frosty season,] [when the sun
453 was set,] [and visible for many a mile
454 The cottage windows through the twilight blazed,]
455 [I heeded not the summons;] [happy time
456 it was indeed for all of us,] [to me
457 it was a time of rapture.] ...

Square brackets show the opening and closing of phrases. A phrase may be opened and closed in the same line, for example 'And in the frosty season,' or opened in one line and carried on to the next, for example 'when the sun / was set.' A section is a phrase or part-phrase within a line. Line 452, for example, has two sections, the first ending with 'season' and the second beginning with 'when' and ending with 'sun' on the next line.

The image-terms are part of the phrases. They are never carried over from phrase to phrase. The metre within the image-terms is embedded in the metre within the phrases. In poems without prominent image-terms, the phrasing may be a main source of emphasis. There are other possible sources of emphasis in a poem, which may have the effect of creating other units, for example a strongly marked caesura, which brings about a degree of {separation} between what is before the caesura and what is after it, or a a spondee which interrupts a regular iambic metre.

The main units of a poem are these. Only the grammatical sentence is necessarily present, if 'prose poetry' is accepted as a form of poetry.

grammatical sentence.
poetic line (but not in 'prose poetry')

Although it's necessary that there should be poetic lines for the use of metre and the application of metrical analysis - prose poetry with anything like regular metre should be set out as poetry, not prose - I see no reason for making the line of poetry the invariable fundamental unit for metrical analysis. Of course, in free verse, the line may become identical with the phrase. For convenience only, I use the lines from The Prelude to show this - my appreciation for the greatness of the poetry, and other considerations, make it impossible to consider this to be a 're-writing.'

And in the frosty season,
when the sun was set,
and visible for many a mile the cottage windows [through the twilight blazed,
I heeded not the summons;
happy time it was indeed for all of us,
to me it was a time of rapture ...

This does have the advantage that it removes all instances of image-term enjambment, as well as phrase enjambment. The transition from line to line gives a pause, sometimes amounting to a gulf. This gives opportunities to the poet. For a brief discussion of the transition and its opportunities, see my discussion of Seamus Heaney's poetry. What has not been discussed so far, I think, either in the essay of Christopher Ricks from which I quote on that page or anywhere else, are the disadvantages of the pause or gulf. Specifically, dividing an image-term, so that part of it is one line, part of it on the line below, gives {restriction} of its impact.

Keats: To Autumn

This is the very well known first line:

Season of mists and mellow fruitfulness

In traditional analysis in terms of feet, 'of mists' forms a unit, with weak-strong stress. In my analysis, 'mists' is an image-term which has semantic force and makes necessary a different analysis of the metrical linkages. The 'of' which is the first syllable of 'of mists' has semantic weakness - not, of course, any weakness of phrasing, grammar or artistry. The 'of' is indispensable.

In my analysis 'mists,' a semantically very emphatic term, has a strong stress which is preceded by the weak stress of 'of' but has a high degree of {separation} from it, despite the fact that 'of' has {direction}: it 'leads to' or 'points to' 'mists.' The successor of 'mists,' 'and,' has a high degree of {separation} from the proximate image-terms 'mists' and 'mellow fruitfulness:' another strong contrast with the traditional analysis in terms of feet, which takes no account of semantic factors and the linkages between semantic and metrical factors.

The units of metrical analysis here are not feet, then, but metrical units linked with image-terms and metrical units not linked with image-terms. These metrical units linked with image-terms may be made up of a single stressed syllable, such as 'mists' or may be a multi-syllabic unit with stressed and unstressed syllables, such as 'mellow fruitfulness' - but the unstressed syllables of image-terms tend to receive more stress than the unstressed syllables outside image-terms.

Robert Frost: Desert Places

The first line is

Snow falling, and night falling fast, oh, fast

This is an interesting opening: emphatic, despite its quietness. A traditional analysis fails to show the fact that all the syllables have a degree of emphasis, except for 'and' and 'oh,' including the so-called unstressed syllables within the image-terms. This analysis wouldn't be a gross exaggeration:

Snow falling and night falling fast, oh, fast

even if the first syllable of 'falling' has a greater degree of emphasis than the first.

The next line is without obvious image-terms, although some people find resonance in 'fields.' They may associate fields with growth, productivity, income and other things.

In a field I looked into going past,

It would be possible to analyze this line in traditional terms, as iambic, with only moderate stresses. This would be to overlook the perturbing effect of the first line and the metrical momentum of the first line. The emphatic rhythm of the first line is carried over to a significant extent to this line: emphasis-enjambment. This second line is surely far more emphatic in its rhythms, in the emphasis even of unstressed syllables, than it would be in isolation.

Robert Frost: Stopping by Woods on a Snowy Evening

Whose woods these are I think I know.
His house is in the village, though;
He will not see me stopping here
To watch his woods fill up with snow.

The metre of the whole poem is regular but not completely regular iambic (the opening two syllables are both strongly stressed, so that the poem begins decisively, and 'fill' is strongly stressed). The rhythm is in slow and steady motion, like the horse's hooves printing the snow. The fact that the horse does stop, and the stopping is central to the poem, is in contrast with the unmodified rhythm. This contrast could even be viewed as a fault - here, metre and meaning are at {distance} - but this would be hyper-critical, I think. I share Ian Hamilton's admiration for the poem (see his introduction to the Penguin 'Selected Poems.' '... the hesitation, the confrontation of the choice, is neither laboured nor inflated but on the contrary is sensitively dramatized in its own slight, fleeting terms.' After a preliminary examination, this is a poem where traditional analysis is applicable: metrical analysis in isolation from meaning, image-terms without perturbing effect. Binary analysis does justice to the metre, in general. 'in', as in 'in the village' and 'up' have strong stress. Further examination shows that stress-meaning linkages account to a large extent for the observed regularity of stresses. In the first line, there's a scrupulousness in the desire to establish ownership of the woods, and semantic inertia in the carrying over of this scrupulousness to the second line, concerned with the extenuating circumstance that the owner's house is in the village. This care and scrupulousness lead to distributed and regular stress, not metrical stress linked with image terms at {distance} from each other.

Shakespeare: Hamlet, Act III, scene i, line 56

To be, or not to be, that is the Question:

This celebrated line has a semantic structure which makes the traditional metrical analysis wide of the mark. To make clear the 'semantic structure' is to make clear the terrible dilemma. To give a better 'metrical analysis' is to show the unforgettable impact of metre. The use of technical terms doesn't reduce the impact of the poetry at all.

The semantic structure of this line, reduced to its essentials, is 'To carry on living? ('To BE?) NOT to carry on living? THAT'S the essential question.' (1) The alternative of living, posed as a question. (2) The negative alternative (3) A statement of the extreme importance of the dilemma.

The established notation of symbolic logic gives an account of 'to be, or not to be, which can be used to state concisely the semantic structure of the line: p p, where p is the proposition 'to be,' is the disjunction connective and is the negation operator. The poetic realization has some distinctive differences of emphasis and de-emphasis. p, and are given emphasis. The second occurrence of 'to be' is part of the basic semantic structure but not of primary importance in the poetic structure. The negation of 'to be,' by means of the simple word 'not' establishes this, clearly and memorably. 'the Question' isn't part of this terrible dilemma. It's already clear that this extreme dilemma is posed in the form of a question. The 'is' which precedes it has no semantic force. The 'to' which precedes the first occurrence of 'be' has no semantic force. An infinitive form of a verb has to be preceded by 'to' but that's taken for granted.

My analysis has metrical-stresses linked with the semantic-force-stresses of the first occurrence of BE, NOT and THAT. The first occurrence of 'be' also has emphasis by extension-contrast.

To be, or not to be, that is the question:

This gives, then, only three fundamental stresses, which are enough to give overwhelming impact to the dilemma. The stress on the first syllable of 'question,' insisted upon in a traditional analysis, is insignificant by comparison.

Words with semantic significance include, as this example makes clear, not only obviously 'vivid' words but words which mark a strong contrast, such as negation, and ostensive words, such as 'that.'

T S Eliot: opening of The Waste Land

April is the cruellest month, breeding
Lilacs out of the dead land, mixing
Memory and desire, stirring
Dull roots with spring rain.

The poem-lines are at medium volume, with small vertical stress intervals. The fact that there are small stress intervals is declared. It isn't sufficient to show the notation-lines as being close together.

Perturbation can be forward or backward. In forward perturbation, the perturbing element has an effect on what lies after it. In backward perturbation, the perturbing element has an effect on what lies before it. The assumption, of course, is that a proper interpretation should never be based on a first and only reading. 'April' and particularly its first syllable is back-perturbed by the rest of its line and the later lines. The opening word of a poem, and particularly the opening syllable of the word, are very often given great emphasis, if the opening is emphatic at all, rather than belonging to the category 'quiet and uneventful.' But 'April' shouldn't be interpreted as receiving great emphasis here, otherwise there would be too great a contrast with what comes next. These lines should be at middle volume in their entirety.

Thomas Hardy: The Self-Unseeing

The metre of these lines is analyzed by Tom Paulin in his 'Thomas Hardy: The Poetry of Perception.'

Here is the ancient floor,
Footworn and hollowed and thin,
Here was the ancient door
Where the dead feet walked in.

This is how he indicates the stressed syllables:

Here is the ancient floor,
and hollowed and thin
was the former door
the dead feet walked in.

He writes, 'The first line consists of a trochee and two iambs; the second line of a spondee, an iamb, and an anapaest; the third has the same pattern as the first line, a trochee followed by two iambs, but with an internal rhyme ('for' and 'door'); while the fourth line is a most marvellous and surprising line - a trochee followed by two spondees.

In my analysis, the first line has two image-terms, 'Here' and 'ancient floor.' Although the second syllable of 'ancient' has weak stress, as part of the image term it has a stronger stress than either syllable in 'is the,' which is outside both image-terms. As so often, 'is' and 'the' are weakly stressed. English might almost be the Latin of 'Ars longa' in the phrase about medical education, 'Ars longa, vita brevis,' 'The art is long, life short.' In the Latin, 'est' or 'is' is missing and there is no word in Latin for 'the' (or 'a').

The third line is analysed in the same way, except that the second syllable of 'former' has weaker stress than the second syllable of 'ancient' in the first line. This is because 'ancient' ends with a consonant and 'former' with a vowel, so that 'former' is subject to greater 'fading' than 'ancient.'

In the second line, there are three image-terms separated by two instances of the conjunction, 'and.' The two disyllabic image-terms, 'footworn' and 'hollowed' both end in consonants and they are strictly equivalent in their stresses.The second syllable of each receives almost exactly the same stress, and both receive slightly less stress than the first syllable.

The fourth line consists entirely of image-terms, except for 'the.' 'dead feet' and 'walked in' are both image terms. Prepositions are often not image-terms but 'in' here has emphasis, important in showing the direction of walking. It's feet which walk, not the adjective 'dead.' The feet are obviously semantically important. The stronger stress on 'dead' than 'feet' in Tom Paulin's analysis isn't part of my analysis. Both words are stressed almost equally. All four words in 'dead feet walked in' are stressed almost equally.

Generative Metrics

Poets need to know nothing about Generative Metrics (GM) for the purposes of poetic practice, any more than writers in general need to know anything about generative grammar, but GM, as a rigorous theory of metre, is very interesting and significant in its own right. I don't give an introductory explanation here of GM. My discussion is confined to just a few aspects.

Theories of metre, like scientific theory, have an empirical basis. It's characteristic of GM, like Generative Grammar, that it goes beyond generalization of observed instances to give a set of rules.

There are very often significant linkages between things which are strongly contrasted in some other ways. I refer to these as cross-linkages. There are significant linkages, I think, between GM and mathematical structuralist views (MSV) in the philosophy of mathematics.

GM views the line as 'a sequence of positions,' (entry in the 'Princeton Encyclopedia of Poetry and Poetics.') According to a simple binary system or a more complex graduated system, syllables are allocated differences of emphasis.

In his discussion of 'Structuralism' in 'The Oxford Handbook of Philosophy of Mathematics and Logic,' Geoffrey Hellman states that 'objects' 'serve only to mark "positions" in a relational system ... their "individual nature" is of no mathematical concern.' The linkage with the account of the poetic line, for metrical purposes, as a 'sequence of positions' is obvious and there would seem to be the possibility of a more general theory of sequences of positions, at a higher level of generality than MSV or GM.

A theory of this kind is unlikely. There may be linkages between theories within a sphere, such as mathematics or stylistics - intra-linkages - but not inter-linkages, other than the most obvious. The epistemological contrasts are too great. (Scientism is the ignoring of inter-linkages.) The linkages which are important are at the thematic level. Themes are intended to be applicable to inter-linkages of all kinds as well as intra-linkages.

Thematic linkages are applicable to the otherwise unlinked spheres of mathematics and metre. Cross-linkage is important here, the linkage between a mathematical sequence, a mathematical structure which shows {direction}, and the sequence of positions in Generative Metrics, which also shows {direction}. The mathematical structure here has a linkage with the non-mathematical structure, and this particular linkage is lacking with many other mathematical structures - for example some geometrical structures not showing {direction} or the specific characteristics of a sequence.

In very simple terms, a mathematical sequence is an ordered set of existents. Using my terminology, the generalization of 'sequence' which includes both mathematical sequences and the sequences of generative metrics involves {ordering}: - E.

If the existents in the sequences of generative metrics have {direction}, what is the 'telos'? (I use the Aristotelian term for 'end' or 'goal.') But although the telos is important, it isn't important, surely, for the purposes of analysis in terms of Generative Metrics. What is important, obviously, is the fact that there is {ordering} in time, so that the weak-strong pattern of an iamb is different from the strong-weak pattern of a trochee. Compare the mathematical concept of an ordered pair, which again is a sequence: < a, b > is not identical with < b, a > unless a = b.

In the mathematical example, distinguishing a and b, or deciding that a = b, is unambiguous. Is it as easy to distinguish unstressed and stressed syllables? I only deal with one difficulty here, and only very briefly, the difficulty of perception. Generative metrics is based on phonological principles and not phonetic ones. In his introductory but very detailed book, 'A Guide to Musical Analysis,' about Schenkerian and other approaches to analysis, Nicholas Cook claims 'music as it appears to the listener and music as it appears to the analyst may not necessarily be quite the same thing.' To relate this comment to my previous discussion of image-terms and their effect on the metre of the poem, I do claim that the listener (or reader) of a poem is likely in many cases to perceive the image-terms so much more easily or readily or vividly than the other poetic material in the lines that both the traditional analysis in terms of feet and an analysis in terms of Generative Metrics become, not non-applicable but 'subsumed.' If a is an unstressed syllable and b is a stressed syllable according to these analyses, then perception may be different from analysis. The listener or reader may not distinguish them or may even invert them. The general lessening of stress, both strong and weak outside the image-terms may be one reason why it becomes difficult to distinguish them, even if contrasts remain. It's often stated - but not argued - that the contrasts are fundamental.

Contemporary philosophical analysis, like the analysis of the past, so often, is immensely sophisticated in general. Practitioners in a sphere are usually unaware of more than a fraction of the difficulties, paradoxes, complexities which are generated within their sphere, with implications, very often, for other spheres. They have enough to do within their sphere. Philosophers of science and mathematics are aware of so many issues which usually escape the attention of working scientists and mathematicians.

So far, philosophers have left Generative Metrics alone. Practitioners within the sphere will have had an easier time as a result. But there are surely many, many difficulties with the foundations of Generative Metrics. In time, philosophers may will turn their attention to GM and point out some obvious difficulties. Here, all I do is allude to one or two obvious ones.

The most obvious of all is that GM is surely an inductive generalization from empirical data - metrical data, in this case. As such, in accordance with the influential views of Popper, which I myself accept, GM is falsifiable. The fact that GM appeals to 'rules' seems over-confident and a confession of weakness at the same time: over-confident because the rules are based on inductive generalizations, a confession of weakness because 'rules' which are based on inadequately justified theory are no better than 'rules of thumb.' In the definition from Collins English Dictionary, a rule of thumb is 'a rough and practical approach, based on experience, rather than a scientific or precise one based on theory.'


Metre, as I see it, has to be considered in its linkage with meaning, generally, although there can obviously be separate studies of the two. Metre in English is predominantly accentual. The stresses, strong and weak, are periodic, like the peaks and troughs of a transverse wave, but of course with many, many differences. The periodicity of the metre is modified, as I see it, by the fact that in so many lines, there are regions made up of image-terms and regions outside the image-terms, these outside-regions often made up of prepositions and the definite and indefinite article, amongst other words.

I've mentioned extension-contrast in a few places in my discussion of individual lines of poems but now I give a more detailed account, although not a very detailed account. English poetry, unlike Greek and Latin poetry, isn't qualitative, with the length of syllables as the source of metrical contrast. But I think that syllable length is an important source of emphasis in the line, and in some lines is almost as important as stress as a source of emphasis. Qualitative stress, unlike quantitative stress, is almost always non-periodic.

In the first section of my page Line length in poetry, I discuss differences of extension in my poem 'Sifted, unfallen snow. I quote next, with modification and some omissions, from some sections of this page where I mention extension-contrast:

From my discussion of John Donne's 'A nocturnall upon S. Lucies day, Being the shortest day.'

The first clause is : Tis the yeares midnight,

'Midnight' has very strong associations and is given great semantic emphasis. The emphasis is shown by underlining.

Tis the yeares midnight,

'yeares' has great emphasis too, obtained by extension-contrast. Not only is a year very extended by comparison with 'midnight,' the word itself can be extended to a greater extent than 'midnight.' The second vowel of 'midnight' can be extended slightly, but to nothing like the extent of the vowels in 'yeares.'

Tis the yeares midnight,

From my discussion of Shakespeare's Hamlet, Act III, scene i, line 56.

To be, or not to be, that is the Question:

My analysis has metrical-stresses linked with the semantic-force-stresses of the first occurrence of BE, NOT and THAT. The first occurrence of 'be' also has the emphasis of extension. Although 'be' in both occurrences can be drawn out, extended, the different semantic roles of the two give the first occurrence greater extensibility in this line. This occurrence receives the accentual-metrical stress and the extension-stress, whereas in 'or not to be' it's 'not' which receives the metrical stress, but no extension-stress. My detailed analysis above gives further discussion.

In general, systematic uses of metrical stress, rhyme and qualitative contrast (closely linked with extension-contrast) have been much more fully explored and discussed than non-systematic uses. This is for obvious reasons, but in the case of qualitative contrast (and extension-contrast), this has entailed the neglect of a very important technique in poetry in English. Because almost all poetry in English makes no use of quantitative metre and English isn't at all well adapted to the use of qualitative metre, the fact that English allows very significant use of non-systematic qualitative contrast hasn't been sufficiently recognized.

In the section above, A new system of metre notation (scansion), I outline the system I use for indicating strong and weak accents and gradations of accent.

The extensibility of a word is clear enough by considering the pronunciation of the word. As in the case of metrical stress, it often helps to employ deliberate exaggeration. The extension of a word in poetry, the extension as linked with meaning, the extension as linked with poetic success, or lack of success - these make a notation for extension very useful.

In the International Phonetic Alphabet, ':' is used to indicate length of vowels in some cases and is used when the vowels are long. I make use of the colon too, in bold, italic print. The colon shows extension in the preceding syllable. Contrasts in extension, extended or not extended, can be shown by inserting the colon after syllables which are extended. Contrasts between degrees of extension, extended or extended to a greater extent, can be shown by inserting a single colon : or two colons :: (A declaration before the discussion can make it clear which system is being used.) When a colon is used in its primary, established way, as a punctuation mark, then I use non-bold, non-italic print.

So, the line above from Hamlet, which includes a 'primary, established colon' at its end, can be notated (Dn: one or two colons to indicate degrees of extension) as,

To be::, or not to be:, that is the Question:

My translation of the first verse-paragraph of Dante's Inferno uses extensions which aren't pronounced at all but which give significant contrast:

Midway in the jour:ney of our life:
I found myself in a dark: for:est:,
where the road: I should follow was lost.


Selections from Donald Wesling's 'The scissors of meter: grammetrics and reading'

'Versification,' 'an electronic journal of literary prosody:

And this Introduction to Rhythm Analysis by Ellen Stauder. It gives a clear and concise account of the approaches of Derek Attridge and Generative Metrics, but with hardly any attempt to assess their limitations:

The site includes a short bibliography, including:

Derek Attridge, 'The Rhythms of English Poetry.' (Longman)

I'd add another  book by Derek Attridge, but shorter and less detailed,

'Poetic Rhythm: an Introduction.'

And (although only a few of the chapters specifically concern metre in English poetry - the other chapters will be of interest more to linguists with specialist interests):

Morris Halle and Nigel Fabb, 'Meter in poetry - A New Theory.' (Cambridge University Press).