Forgotten Ground Regained

A Classic Sampler
Beowulf / Viking Poetry
Sir Gawain & the
Green Knight and Pearl

Poetry 'zine
Featured Poems
Editor's Notes

Other Translations
Medieval Texts
Modern Poetry
Fantasy Poetry
Poetic Techniques / Essays

Site Info
Masthead / Awards
New Changes & Old
Site References

Linking Letters:
A Poet's Guide to Alliterative Verse

Part XIV: A Prosodic Primer

    This page is something of a fast course in the theory of linguistic rhythm. Note that I said linguistic rhythm, not poetic rhythm: this is the kind of rhythm you find in any kind of language, not just poetry. We will go into more detail than earlier; once we have built up an understanding of the basic concepts, we will return to poetry, metrics, and Anglo-Saxon verse. The theory presented here is a somewhat generic version of modern linguistic accounts. It is phrased to apply specifically to modern English; the rules for Old English are similar, but differ in points of detail.

    Syllables, Beats and Off-beats

    At the lowest level, we can divide a word, phrase, sentence, or line of verse into syllables. For instance, we can count 14 syllables in the following sentence from a nursery rhyme:

      1   2   3  4  5   6    7    8      9   10  11  12 13 14
    The king is in his counting-house, counting out his money.

    But not all syllables are the same. Some syllables are heavy, and stressed; others are light, and unstressed, and there is a strong tendency to alternate between the two. This alternation can be analysed as a sequence of beats and off-beats. The heavier, more strongly stressed syllables become beats: the lighter, less stressed syllables become off-beats. In our nursery rhyme example, there is an almost exactly even pattern of beats and offbeats, though some of the beats are far stronger than others:

    The king is in his counting house, counting out his money.

    Beat and off-beat is essentially what is measured in accentual-syllabic verse. But there are several layers of rhythmic organization above it.

    Degrees of Stress

    Linguistically, we have to distinguish four levels of stress: primary stress, secondary stress, tertiary stress, and lack of stress. These four levels are not an absolute degree of loudness; they simply measure a scale from the strongest stresses to the weakest unstressed syllables. Let us use S to mark primary stress, s to mark secondary stress, W to mark tertiary stress, and w to mark lack of stress. Then our nursery rhyme example would be marked:

      w   S  w  W  w     S  w    s      S   w    W   w   S w
    The king is in his counting-house, counting out his money.
    The following rules apply:

    • Use S to mark the root syllables of content words.
    • Use s to mark (a) secondary stress of content words, and (b) primary stress of function words.
    • Use W to mark an unstressed syllable that is heavier than its neighbors. (I.e., a heavy syllable like out is stronger than the closed syllables ing and his.)
    • Also use W to mark an unstressed syllable that is slightly more strongly stressed than its neighbors
    • Use w for everything else.

    Following these rules, the first sentences of this page would be marked as follows:

      w    S   w   S   w   W  w  S     S    W   w    S  w  W   w  S  w   S  w
    This page is something of a fast course in the theory  of linguistic rhythm
      S    w  w  S    w   S  w    S  w   W   w S  w    S  w
    Note that I said linguistic rhythm, not po e tic rhythm:
      W  w   w    S   w   S  w  w   S   w  s w   S  w   S   w     W   w    S w  W
    this is the kind of rhythm you find in any kind of language, not just po e try.
     W   w   S  s w  W    S  w    w   S  w w
    We will go into more detail than earli er;
     W    w  w     S   W  w   s w   S   w  w    w  S w   S  w 
    once we have built up an understanding of the basic concepts,
      W  w    w S    w  S w W     S  w     w  S   w  s w   S
     we will return to po e try, metrics, and Anglo-Saxon verse.

    Clashing Stress

    There is a strong tendency not to allow syllables of equal stress to appear next to each other. When they do, something happens to relieve the clash. Sometimes the stress pattern changes. For instance, thirteen has stress on the second syllable, but thirteen men is pronounced with stress on the first syllable of thirteen. Other times, one of the stresses is elevated, or a slight pause is introduced between the two stresses. We will look at this in more detail below

    Rhythmic Feet

    The next level of organization involves the concept of the rhythmic foot. A rhythmic foot is not quite the same thing as a poetic foot, but it is a unit of about the same size and method of organization: a beat paired with an off-beat. A foot may be iambic (offbeat plus beat), or trochaic (beat plus off-beat). If a beat or off-beat cannot be paired with another syllable, it makes a (single-syllable) foot by itself.

    Feet also tend to alternate between weak and strong elements. A strong foot is a foot that contains a strong stress (S). All other feet are weak.

    The following rules for constructing feet apply in Modern English, and generalize fairly well to the rhythm of Old English:

    1. Prefer to put two syllables together to make a foot if they belong to the same word.
    2. Prefer to put two syllables together to make a foot if the first syllable is stronger than the second.
    3. Prefer to make feet of two syllables (monosyllabic feet are allowed.)
    4. A foot is strong if it contains a strong stress, weak otherwise.
    5. unstressed syllables can be monosyllabic feet only if they are also independent words.

    If we use F for strong feet, f for weaker feet, we get groupings like the following:

         F     f     F       f   F    F            f      F      f      F       F
        / \    |    / \     / \  |    |           / \    / \    /  \   / \     / \
      w    S   w   S   w   W  w  S    S          W   w  S   w  W   w   S  w   S   w
    This page is something of a fast course     in the theory  of linguistic rhythm

    Note the clashing stress, "fast course", which has to be resolved either by stressing fast, making it stronger than course, or vice versa.

    Also note that the rhythmic feet are not consistently iambic or trochaic; they are constructed to show the local rhythm within the word, then in the immediate context of the word.

    Prosodic Words

    The next layer of rhythmic organization is the prosodic word, typically two rhythmic feet, one weak, one strong. The basic rules for putting together a prosodic word are as follows:

    1. Prefer to combine two feet that belong (or contain syllables from) the same word.
    2. Prefer to combine two feet that have the same rhythm (iambic or trochaic)
    3. Prefer to combine a strong foot (F) with a following weak foot (f)
    4. Otherwise, prefer to combine a weak foot (f) with a following strong foot (F).
    5. Otherwise, prefer prosodic words with two feet, both weak or both strong.
    6. Any feet that cannot be combined in this fashion are prosodic words on their own.

    So, applying these rules to the same sentence, we get:

            WD           WD        WD                 WD            WD         WD
           /  \        /    \     /  \               /   \         /   \        |
         F     f     F       f   F    F            f      F      f      F       F
        / \    |    / \     / \  |    |           / \    / \    /  \   / \     / \
      w    S   w   S   w   W  w  S    S          W   w  S   w  W   w   S  w   S   w
    This page is something of a fast course     in the theory  of linguistic rhythm

    Rhythmic Phrases (Breath Groups)

    The highest level of rhythmic organization in language is probably the rhythmic phrase -- what has sometimes been called a breath group in some theories of poetic meter. This is a unit roughly the same length as a short clause or long phrase in grammar, and encompasses two to three prosodic words. As a general rule, the component parts of a rhythmic phrase are grammatically related, corresponding to such units as simple clauses, prepositional phrases, and the like. In the example we have been working through, there are two rhythmic phrases:

    This page is something of a fast course


    in the theory of linguistic rhythm.

    Rhythmic Prominence

    Grouping is only one part of rhythm. The other part is the relative prominence of the individual syllables. But prominence and grouping are related: the strongest syllable in a rhythmic group has to be more prominent than any other element. In addition, prominence tends toward an alternating pattern: an alternation of strong and weak elements is most natural, and if two adjacent elements are equally strong (as in clashing stress) something usually gives to create an alternating pattern.

    One way to represent this is to build what some theorists call a rhythmic grid, adding an extra level of prominence to each syllable based on its strength in the overall pattern. We construct such a grid by following these rules:

    1. Put an x over each syllable, an extra x over each beat.
    2. Two beats are adjacent if they are separated by no more than one unstressed beat. If two beats are adjacent, and one is stronger (i.e., S and s, S and W, or s and W) then add an extra mark over the stronger of the two.
    3. Two beats clash if they have are immediately next to each other and are the same or almost the same stress level (i.e., they have the same number of x's, or differ by one.) Prefer an analysis of the rhythm that avoids stress clash. But if it occurs, put extra marks over one the two clashing stresses to eliminate the clash. (Which one depends on various factors, including emphasis; the default in modern English is generally the second.)

    Applying these rules to our example, this is what we get:

                   x             x    x                  x             x       
          x        x        x    x    x         x        x     x       x       x
      x   x    x   x   x    x x  x    x         x    x   x  x  x   x   x  x    x  x
    This page is something of a fast course     in the theory  of linguistic rhythm

    This represents the natural, neutral way to say the sentence, without any special emphasis added. Notice that the most prominent beats are those which clearly dominate another beat, which means (in turn) that they usually belong to "heavy" prosodic words -- units which contain more than one stress.

    And this leads to a final definition: prominent stresses. As I will use the term, a prominent stress has two closely related characteristics: (1) it dominates an adjacent beat (which means in turn that it has three or more x's, not the two associated with a lighter stress; (2) it is the strongest beat in its prosodic word. So in our example, the prominent stresses are some in something, course, the first syllable of theory, and the second syllable of linguistic. The stressed syllables in page and rhythm are strong stresses, but not prominent stresses. The beats on of, in, and of are neither prominent stresses nor strong stresses.

    Notice, by the way, that the rhythms I have described are in some sense defaults; they can change considerably if a word gets special emphasis -- what is sometimes called rhetorical stress. We can easily give page or rhythm stronger emphasis than the naturally prominent stresses, but this isn't the same as the built-in, natural prominence of the stresses that really have to be stronger because of their position in the rhythm of word and phrase.

    How this applies

    At this point the application of these concepts to alliterative verse should be fairly obvious: strong stresses are the syllables we have been marking S, and which function as the chief stresses in prosodic words. Only strong stresses matter for alliteration. And the pivots, or chief alliterating stresses of the half-line, must be prominent stresses, the strongest element in their prosodic phrase. Strong dips matter because their extra syllable count will necessarily contain a very light stress; thus, a prominent stress next to a strong dip will dominate that weak beat and have the added prominence necessary for it to function as an alliterating strong-stress pivot. In other words, the special rhythmical role of strong stresses as lifts, and of strong or heavy dips, follows rhythmically from the requirement that alliteration be placed on the most prominent stress in the prosodic phrase that constitutes a half-line. For instance, if we have the half-line:

    	as he spoke to me

    the strong dip "as he" forms a foot with weak stress on as. Since the next syllable is the intrinsically strong-stress syllable spoke, they combine to form a prosodic word, with spoke dominating the weaker stress in the word. The final stress on me, since it is neither an intrinsically strong stress, nor a stress that dominates another stress, is also subordinated to the strong stress on spoke. Thus the only possible pivot syllable on this half-line is spoke. Or to take another example, a different Sievers type, consider the phrase:

    	hard-hearted men
    Here, there is intrinsic stress on hard, heart, and men. However, hard has the primary stress in the compound word hard-hearted, and so it is both (a) and strong stress, and (b) it dominates another stress, the secondary stress on heart. It is therefore a stronger stress than the intrinsic strong stress on men, which does not dominate another stressed syllable, and hard is the only possible pivot syllable in the half-line.

    This story is complicated for Old English, where resolution must be kept in mind (i.e., Old English treats a long stressed syllable as equivalent to the combination of a short stressed syllable followed by an unstressed syllable) and by various other facts about Old English as a language. But the basic analysis does, I believe, apply fairly generally.

    Back: A Framework
    Next: Mimicking Old English Alliterative Verse

    Copyright ©2000, Paul Deane