A Field Guide to Alliterative Verse

Doing Old English Alliterative Verse

I am not going to take the time in this section to provide a detailed application of the concepts I have just sketched to Old English verse proper. To do that, I would have to provide an exhaustive tour through rather complex facts about Old English as a language, and there is an ample scholarly literature on the subject (see the Bibliography of Germanic Alliterative Metrics for details or some of the books on my Alliterative Metrics page.)
Instead, I am going to sketch how the concepts I have outlined thus far can be used to mimick Old English verse forms in Modern English. While it is impossible to capture exactly the Old English form, one can manage a very close approximation. The key is to capture the rhythmic essentials of the form while keeping the close connection between alliteration and rhythm on which that form depends.
So then, taking the rules already outlined above, let us take a quick look at what it takes to write alliterative verse in modern English. But first, a couple of caveats and general suggestions.
Excess Alliteration
The classic form only has one pivot (obligatorily alliterated syllable) per half-line. In the Anglo-Saxon form, an extra alliteration can be put into first half-line, but is not necessary; and if it is overdone the effect can be far too strong.
End-Stopping
People used to rhyming verse of the usual English type will be strongly tempted to end clauses and sentences at the end of an alliterative line. That's actually to be avoided. The Anglo-Saxon form works better if sentences do NOT line up neatly with the ends of lines.
Rhythmic Ambiguities
One of the difficulties of imitating the Anglo-Saxon meter in modern English is that modern English typically involves rising rhythm in which the second of two nearby stresses is stronger than the first. Old English typically illustrates a falling rhythm in which the first of two adjacent stresses is stronger. This can cause problems, especially in half-lines with iambic rhythm: it is natural to read the second stress as stronger, but since the form requires alliteration on the first stress, the metrical rhythm indicates that the first beat is the stronger. Either way of stressing the phrase is possible, but the falling stress suggests emphasis. This can lead to unnatural-sounding phrases unless care is taken to structure the sentence so that the emphasis suggested by the alliteration is natural.
Kennings
Kennings are cunning little metaphors in which little phrases like "The Whale Road" is used to mean the sea. Classical Anglo-Saxon verse is full of kennings. Imitating this has its pitfalls, however, as modern audiences do not expect them and a high concentration of kennings can make for dense, almost unreadable verse. I tend to use them sparingly.
With these caveats, let us take a quick look at one example in which poetry adhering very closely to the Anglo-Saxon style has succeeded in English: J.R.R. Tolkien's alliterative verse play, The Homecoming of Beorhtnoth. Consider the following lines:
To the left yonder. There's a shade creeping, a shadow darker than the western sky, there walking crouched! Two now together! Troll-shapes, I guess, or hell-walkers. They've a halting gait, groping groundwards with grisly arms.
The metrical rhythm can be diagrammed as follows; the alliterating pivot syllables are marked in all caps, and stress is marked both in terms of the grid (with x-marks) and in terms of the S s W w notation. I also note the Sievers type for convenience.
x x x x x x x x x x x To the LEFT yonder. W w S W w [Type C]
 x x x x x x x x x x x There's a SHADE creeping, W w S S w [Type C] x x x x x x x x a SHADow darker w S w S w [Type C; "shadow" counts as single resolved lift] x x x x x x x x x x x than the WESTern sky, W w S w S [Type B] x x x x x x x x x x there WALKing crouched! w S w S [Type B]
 x x x x x x x x x x x TWO now together! S W w s w [Type D; "gether" counts as a single resolved lift]
 x x x x x x x x x x TROLL-shapes, I guess, S s w S [Type D]
 x x x x x x x x or HELL-walkers. w S s w [Type C] x x x x x x x x x x x They've a HALTing gait, W w S w S [Type B] x x x x x x x x x x GROPing groundwards S w S W [Type A] x x x x x x x with GRISly arms. w S w S [Type B]
Notice what Tolkien has done with these lines:
  • He has been sparing with the alliteration, allowing double alliteration only in the last line quoted.
  • He mixes end-stopped lines with lines where the sentence break is in the middle of the line.
  • In iambic half-lines (a shadow darker, there walking crouched, with grisly arms) the emphasis implied by alliteration on the first stress makes sense in terms of the overall rhythm of the sentence.
  • He does not follow the Anglo-Saxon tendency toward filling the verse as full of kennings as a fruitcake. (No doubt medievalists will shudder when I put it like this. But kennings worked in Old English because the audience already knew most of them, and were likely to admire a poet who came up with a particularly clever new one. Without that kind of audience, kennings make alliterative poetry much less accessible).
  • While his lines all fit strictly into one of the five Sievers types, he's reinforced most of the pivot stresses with long or strong dips, which helps get the intended rhythm, which I've shown above. One of the risks of using the traditional alliteration pattern in modern English is that default English rhythms place the strongest stress at the end of the phrase/clause, but in Old (and Middle English alliterative verse, the final stress of the second half-line (sometimes called the b-verse or off-verse) IS NOT SUPPOSED TO ALLITERATE. Modern English poets are therefore strongly tempted to alliterate on the last strong stress of the line. Even Tolkien has sort of done that, in one line in this passage. The line "two now together: Troll shapes I guess" only fits Old English meter if we treat both "two" and now" as taking strong stress. It would be equally natural to read the line line like this:
x x x x x x x x x x x x x x x x x x Two now together: Troll shapes I guess
  • On this reading, the pivots would be "guess" and the "ge" in "together" and the line wold alliterate on "g". That's clearly NOT Tolkien's intention, but this particular rhythm is really easy to fall into writing a modern English alliterative line. Just be aware that this kind of rhythm, alliterating on the even-numbered lifts, isn't going to sound like Old English. Tolkien gets away with it because stress on "two" the more sensible speech rhythm, which makes both half-lines type D, as shown above, and makes the alliteration on "G" purely decorative. In the context of the whole poem, the first stress of almost everyhalf line is naturally read as the strongest, which is what is needed if the rhythm is to sound like Old English.
One note, however. Tolkien's usage tends strongly towards two strong stresses per half-line. While this is common in Anglo-Saxon verse, it is not strictly necessary. A single four-syllable word, with one strong stress and one secondary stress, makes a perfectly good Anglo-Saxon half-line. Comparably, in modern English, words like half-heartedly or unworthiness are also entirely acceptable as half-lines, creating potential for much more rhythmic lightness and flexibility than is typically allowed for the form. Thus, lines like the following are perfectly metrical:
x x x x x x x x x x HALFheartedly S S w W
 x x x x x x x he HELD his tongue, w S w S
 x x x x x x x aWARE of his w S w W
 x x x x x x x unWORTHiness. w S w W
In general, the key to making alliterative verse work well is keeping the rhythm natural and unforced, and making sure that the alliterating words which function metrically as pivots can bear the natural emphasis.
Copyright ©2000, Paul Deane
No part of this site may be used or reproduced in any manner for the purpose of training artificial intelligence technologies or systems