I

Schopenhauer said that ‘rhythm is to time as symmetry is to space’. Symmetry tends to be an overt or latent feature of most art which attempts to make an integrated unity of its material: as humans, it seems we can find no deeper sign of a thing’s structural integrity than the doubling, reflection, repetition or self-similarity of its own form. Poetry’s medium is time-based, being that of spoken or read language. The relatively recent innovation of representing it on the page has lent it very strong spatial features too, allowing us to not only sense but also see the poem as a ‘single-celled entity’, where its elements are related not just in syntagmatic series, but in paradigmatic parallel. Poets will now write directly for this spatialised medium; ‘symmetry’ can be formed through the kind of motivic, thematic and conceptual connections we looked at in the previous essay. But many of us feel that poetry is still written to be first read aloud – or at least ‘aloud in one’s head’ – in that time-driven, gappy noise humans use for making sense. The principal way in which symmetry manifests itself in spoken language is rhythm.

Rhythm is an inevitable feature of poetic speech. As representative products of nature, everything we create is imprinted with the steady beats and cycles of which we are ourselves composed in time, and language is no exception. Language is hopelessly rhythmic because we are hopelessly rhythmic – in our circadian cycles, footsteps, heartbeats, breathing, in the oscillations of our brainwaves and the circulation of the blood. We are rhythmic because physical law has created a rhythmic universe, and settles everywhere into periodicity: patterns of regular gravitational orbit, spin, electromagnetic pulse. (If there was a god or prime cause, they clearly had a hell of a groove on them.) The poet who would denounce all explicit rhythm in verse as so much artifice should affect the same distaste towards their own regular breathing and their habit of sleeping every night.

Furthermore, written and oral poetry is generally composed in lines or strophes, which not only constitute metrical units in themselves, but also draw forth, exaggerate and crystallise those rhythmic properties already latent in our speech. The poetic line pressurises language, which then reveals that which is most strongly characteristic of its structure and grain. (‘Free verse’ can never be ‘free’ in the sense of ‘free from all rhythmic patterning’, even if it wanted to be.) Rhythm is also the aspect of poetry which most directly declares its sisterhood with music – and through music, its connection to dance. Rhythm has the automatic effect of making our bodies dance, inwardly or outwardly: the metrical ‘foot’ is derived from the Latin pes or Greek πούς, in the sense of ‘what you beat time or tap out a rhythm with’.

Compared to the subjects of sound-patterning and trope, metre has taken up a disproportionately large amount of theoretical discussion; it is an easy subject for any academic suffering from ‘physics envy’ to turn their attention to, and – allied with some basic arithmetic – generate insight after barren insight, each one full of more nuance and detail than the last.1 However I believe the study of prosody to be a finite one, and hope to show why. Poetic prosody tends to be described in either a simple inaccurate way or a complex inaccurate way. The inaccuracies originate, straightforwardly, in a failure to accept its paradoxical nature: it is both a simple and a complex matter. As phonology has understood for many years, prosody is a rather broader project than one of merely describing ‘the pattern of stress’. Prosody describes a relationship between metre and sensemaking, one in which they are taken to be inextricably connected. This symbiosis has consequences: because sense is infinitely multiple, being non-intrinsic and residing nowhere, it is ultimately subjective; this means that in its real-world performance, stress is not accurately quantifiable, as its relative strength from one syllable to the next depends on the local, subjective interpretation of a fundamentally unstable sense. Quantifying the unquantifiable is an endless and pointless project which can nonetheless do a fine impersonation of a rich and insightful one, with the prosodist confirming his brilliant findings again and again – through what, on closer inspection, often turns out to be sheer subjective projection, and the circular confirmation of one’s own discoveries. What is required is a description of the system which includes its subjectivity as integral to it, and does not try to eliminate it. It must also take into account the dynamic dimension of both the poet’s rewriting and the reader’s rereading, both of which further destabilise the stress pattern.

So, an early warning: while musical rhythm can be intelligently discussed in the abstract, poets and theorists of verse should not make the mistake of discussing poetic metre outside the context of the speech act to which it is inextricably bound. In poetry, as in speech, a metrical event can’t be properly understood unless one takes account of the music and the sense it marks. Both music and sense directly impact on the quality of syllabic stress itself, and our perception of it.

While this part of the book concerns itself with ‘form’, inasmuch as stanzas are metrical units too, it does not provide a ‘handbook of forms’. The ‘received forms’ are a metrically trivial matter. Their interest lies in their literary-historical evolution, and in the kind of pressure their structures place upon language in every aspect, not just its rhythm. Terza rima, the standard habbie, the pantoum and especially the sonnet all have rich tales to tell, and there are many books which tell them. Other forms consist of little more than arbitrary rules and strategies. (For the life of me I have never understood how the rich interior tensions generated by the sonnet form can possibly be discussed in the same breath as the merely apparent sophistications of the villanelle and the sestina. [For a short rant on the frequent silliness of the troubadour forms, see endnote 17.]

The second half of this essay will not be easily accessible to the lay reader. While I’ll endeavour to define the many technical terms as clearly as possible, they will soon have to be hurled around with a fair degree of abandon. For better or worse, this is a treatise, not a primer.2 It is an essay on poetic form considered as temporal and rhythmic structure, and is an attempt to explain how and why we shape poems to templates of stanza, line and metrical rhythm, and how these templates might work in the mind and in the ear. As before, its sole novelty is that it is written with someone with a little experience of how such effects are generated from the inside out, and knows them as process rather than effect. I am nonetheless aware that I have occasionally strayed from my own very narrow expertise into fields of scholarly research in which I am barely more expert than any other reader. However, authorities in those fields have also blundered into mine for long enough. My hope is that we soon might meet in the middle, and in the meantime be tolerant of our respective ignorances.

II

The first thing we must address is what form actually does. We know versification is somehow at the heart of the enterprise, but have been pretty bad at articulating why this is so. In his influential paper ‘A Disciplinary Map for Verse Study’, Richard Cureton described three possible explanations of the function of verse-form and metre.3 The first and the most commonly propounded is that it essentially performs a kind of rhetorical function or extends an existing one, and merely reinforces sense. This is predicated on the idea that poetry is primarily valued for its prose-like virtues, and that its value lies in its content (its physical descriptions, fictional speakers, speeches and verisimilitudes), the main difference from prose lying only in its aesthetic orientation, the intensity of its effects and the manner of its rhythmic organisation. One can see that such an approach would immediately trivialise the whole study of form and metre, or at least relegate it to a resolutely secondary and supporting role.

The second, favoured by more post-structural and critical-theoretical descriptions, is that verse form actually disrupts meaning, destabilising our interpretation of the text by providing a kind of playground of artificial effect that goes far beyond mere rhetoric, and declares an extraliterary agenda – deliberately drawing our attention to the very failure of normal language to provide adequate symbols for our contemporary experience. The first position is essentially that verse performs a merely rhetorical function; the second that it provides a disruptive one (or an aestheticising, ludic or empty one).

Cureton rejects these as alternately weak and inadequate, and posits a third approach, which he claims ‘turns the standard view of verse function on its head’, though I think this is a unnecessarily violent corrective. He says:

The more productive alternative to these standard conceptions of verse function, I would maintain, are the claims (1) that poetry does indeed ‘tell’ us various complex and significant things about human sensibility and (2) that these things derive primarily from the verse form itself, not from the meanings arranged in these forms and the fictional representations that we infer from those meanings.

His first point seems to me unarguably true. His second is overstated to the point of absurdity. It is self-evident that ‘the poetic experience’ is derived from both form and semantic content, and indeed any quick poll of readers and poets – neither of whom have been especially mystified by poetry’s effects, even if its machinery has remained obscure – at any point in the last 1,000 years would have established precisely that.

However, his statement that ‘the major gesture in poetry, this theory of verse function claims, is to semanticise the temporal significances of its linguistic forms and to temporalise the spatial significances of its linguistic meanings and their fictional representations’ seems far more nuanced and balanced, and is certainly close to my own experience of both reading and writing verse. My immediate instinct is to link this – yes, again – to Jakobson’s notorious formula: ‘the poetic function projects the principle of equivalence from the axis of selection into the axis of combination’. Once more, with feeling: what Jakobson means here is that the paradigmatic axis, the parallel rule of correspondence between things which governs our alternate choices, comparisons, lyric echoes, metaphors and so on, is developed along the axis of the syntagm, i.e. the serial rules whereby we combine words with other words along poetic sentences. In poetry we both find and ‘project’ a whole host of effects where the similarity between sequential elements is crucial to our experience of them as ‘poetic’. These ‘effects of equivalence’ might include rhyme, consonance, metricality, syntactic parallelism, and so on. (As well as, I would argue, a fundamental increase in the connotative function of language, developed often by phonosemantic means, blurring the boundaries between words and so enhancing our sense of their semantic overlap.) In poetry we see a high degree of connection, integration and a consequent interchangeability of form and content, leaving us with a horizontal structure with vertical depth, where the melodies of surface-sense are underpinned by complex and shifting harmonies of semantic resonance, association and correspondence. And, if Jakobson’s rule is true, it would certainly be no surprise to see an increase not only in the mere horizontal, tap-along presence of metrical rhythm, but also deep involvement of that rhythm in the vertical and recursive structuring of the very spaces it encloses. (I think the often-quoted statement that ‘rhythm creates time’ in the human experience is a little ambitious; we do not yet understand the nature of time well enough to assert anything of the sort. ‘Rhythm structures time’ seems the more defensible claim.)

III

Watching the lengths to which some authors go to maintain the integrity of their prosodic theory is like watching the last days of the Ptolemaic system, with more and more epicycles, adjustments, exceptions and retrograde motions being introduced in order to justify the one big error at the centre. I think their mistake is this: many theorists have completely lost track of the fact that this material has been produced by poets, whose relationship to the raw material of poetry – language itself – is of such a fundamentally strange nature I doubt they would quite believe it. Ignorance of the impulses and processes that have produced the work have led to a grave misreading of its intent; misreading of intent leads to misdiagnoses of function. There is still a fashion for the kind of pure-text readings inherited from the Leavisite, New Critical and post-structural schools (and, as I have occasionally discovered to my cost, from a certain breed of older scholar who continues to regard the poet as a bewildered naïf through which the work is channelled). However, such coldly forensic explorations – they remind one of a team of scientists gathered in bewilderment round a spade, wisely concluding that it is a giant teaspoon – are steadily losing their credibility. Unless each poem is used for the rough purpose for which its author intended it, its full power is unlikely to be released. Therefore any genuinely sensitive, useful reading must also posit what those authorial intentions actually were. Poetry is an act of communication; like any other, you misread the cues at your peril.

An effort to remove messy intent and affect from the study of metre has led to a failure to read stress and rhythm as a sufficiently complex and subjective phenomenon. This is compounded by attempts to notate it in binary or ternary forms nowhere near sophisticated enough to represent accurately either ‘what’s going on’, or the variety of legitimate alternative scansions. Furthermore, while there is often great confusion about what is being measured, there is even more over why it is being measured.

It is, nonetheless, important for the prosodist to remember T. F. Brogan’s wise warning:

This sounds unchallengeably wise, but the whispering goes on. While metre is binary, language is not. (There is also a specific technical objection to Brogan’s statement: sometimes ‘timing’ syllables is a procedural necessity. This is discussed later under ‘the displaced weak rule’, and elsewhere.) As I’ll show, mere ‘metrical prominence’ is not a phenomenon that can be successfully decoupled either from a qualitative and quantitative description of its nature, or from the semantic context that often produces it. As soon as one registers an agreement or disagreement between metrical position and syllable stress, one is already claiming a quality and meaning for that syllable. The problem is this: stress is an inseparable performative aspect of meaning; meaning is in flux, because words have no intrinsicality; and nothing, no other form of human discourse, undermines and revises the connection between a word and its meaning quite like a poem. Lexical stress is already highly destabilised in poetry, but if one also takes no account of the intonational properties of sense-performance – while they might still claim to be working with poetry, the theorist is treating it as mere versified language.

Ideally, poetry is radical thought expressed in appropriately radical language. It often finds this language directly through the attractions of an inflexible metrical frame. The frame additionally provides a safetynet, a set of co-ordinates, a zoo-cage, a musical structure, a matrix – that simultaneously licenses, contains, mitigates and encourages poetry’s excesses. Metrical frame grants words a substitute structure – a scaffold, an earth, a ground, a bearing – that emboldens them to take liberties with even such basic qualities as their own part of speech, and even to sever their link with their own primary denotation. But by the time the reader gets to it, it has also served to arrange these very words, which have been drawn into the pattern of the metre much in the way scattered iron filings are drawn into the nodal patterns of a vibrating membrane. The grievous mistake of many prosodic systems (of which I regard certain kinds of phrase rhythm analysis and generative metrics the worst culprits) is to treat poetic language as if it were any language that happened to find itself versified, and merely ‘thrown up against’ metre, rather than having been born directly from it.

Metre is the perfect minefield. However, we should be guided as far as possible by what it is, exactly, we want to achieve from this exercise, and my aim here is not to clear the minefield but to map it, so we can dance through it. I want to account for the interpretative and compositional possibilities of the metrical line in a dynamic frame, one which will see stress alter with interpreted and reinterpreted sense within a developing thematic domain. However, in attempting to give an account of such a rich field using only the meagre, binary tools of scansion, one is faced with a technical problem similar in nature to that of labanotation, or encountered by the narrator of ‘Flatland’. I needed all the help I could get, and the approach taken here owes something – perhaps nearly everything – to those scholars of metre I’ve studied over the years, from Otto Jesperson and Paul Fussell to Richard Cureton and Nigel Fabb. Without their clear and insightful articulations to completely disagree with, there would be little to say.

However, for better or worse, I have written this chapter from scratch, from the ground up, without their books at my side – mainly to force myself to define every term and account for every step of a process whose integrity can lie only in the hideous interdependence of its parts. Where there are agreements between my system and another, I acknowledge them if I feel they are borrowed; if I don’t, I have probably arrived there by my own route. (A wholly original metrical theory should be treated with great suspicion.) The exception is Derek Attridge’s method. Attridge has, in my opinion, come closest to a full explanation of ‘what’s going on exactly’ in English metrical rhythm, but I hope the methods I use here sound more like broad concurrence than plagiarism. I often arrive at a similar conclusion by different means, though in several key regards our ideas appear to differ substantially. However, I still suspect that – since Attridge’s perspective is that of a scholar, and mine that of a poet – we are merely looking at the same lumbering, asymmetrical mountain from different angles.

Poetry is just speech placed under a spatiotemporal constraint, behind which act lies an emotional impulse: to say something which will maximise language’s power and density. No other poetic technique facilitates this more effectively than measure and number, and an understanding of metre’s complexity is coterminous with an understanding of some of poetry’s most challenging mysteries. We don’t need to understand them to either write or read poetry, and I very much doubt they make much difference to the results either way. But there are, as they say, two kinds of people in the world – and personally, I find material explanations stranger, wilder and altogether far more slack-jawed amazing than those which invoke shadowy and invisible forces. If we learn nothing else, I believe it will be this: that there is little in the phenomenon of metre but our own determined dreaming, and the crystalline projection of the rhythmic structures of our own minds.

1 This is still preferable to the bewildering ignorance displayed by certain other experts in poetics, for whom the subject of prosody still appears to be something of a mystery. Here, for example, is Marjorie Perloff attempting to destroy a poem by the early twentieth-century African-American poet Georgia Douglas Johnson. She quotes the following:

[…] The heart of a woman falls back with the night,

And enters some alien cage in its plight,

And tries to forget it has dreamed of the stars

While it breaks, breaks, breaks, on the sheltering bars.

– ‘The Heart of a Woman’ (1918)

And continues: ‘These chug-chug iambic pentameter stanzas rhyming aabb remind one of a Hallmark card; indeed, so slack is the diction, so hackneyed the phraseology and sentiment …’ – Marjorie Perloff, ‘Janus-Faced Blockbuster’, review of Cary Nelson (ed.), Anthology of Modern American Poetry, Symploké 8, nos. 1/2 (2000): 206.

2 Readers after a crash-course in metrics or a handbook of forms should turn to the excellent books for the general reader by Derek Attridge, James Fenton and John Hollander.

3 Richard D. Cureton, ‘A Disciplinary Map for Verse Study’, Versification: An Electronic Journal of Literary Prosody 1 (1997).

4 Alex Preminger and T. V. F. Brogan (eds), ‘Scansion’, in The New Princeton Encylopedia of Poetry and Poetics (Princeton, NJ: Princeton University Press, 1993), 1118.