Beowulf: Alliteration
Several of the translations, including the Ruth P.M. Lehmann, Frederick Rebsamen, Marijane Osborn, Seamus Heaney, James Garnett, Charles W. Kennedy, and Alan Sullivan/Timothy Murphy versions, have maintained the same sort of alliteration as is present in the original Old English version.

Frederick Rebsamen describes this alliteration on page xviii of his translation:

Old English poetry has no stanzic form and no rhyme (with the exception of a few later poems) except by accident. It consists of lines which run on to form sentences, each line composed of two half-lines, or verses, with a natural pause between them, so that the sentences may end at line-end or between half-lines. There is no set number of syllables per line - in Beowulf a normal line contains between eight and twelve. The half-lines are tied together by alliteration of consonants or vowels, any vowel alliterating with any other vowel through an emphatic pronunciation of stressed words that causes a sharp release of breath approximating a consonantal sound.

Each half-line has two strong stresses. Alliteration occurs only on stressed syllables. The first stress of the second half-line, called the "head-stave," cannot alliterate with the second stress of that half-line, but it must alliterate with one or both stressed syllables of the first half-line.


J.R.R. Tolkien wrote this in his influential 1936 essay entitled "Beowulf: The Monsters and the Critics":

The very nature of Old English metre is often misjudged. In it there is no single rhythmic pattern progressing from the beginning of a line to the end, and repeated with variation in other lines. The lines do not go according to a tune. They are founded on a balance; an opposition between two halves of roughly equivalent phonetic weight, and significant content, which are more often rhythmically contrasted than similar. They are more like masonry than music.


T.A. Shippey, on pages 45-8 of his 1978 book, also says, with respect to alliteration:

It is accordingly not disrespectful to remark on one of the sorest and least-probed points of Beowulfian style -- its casual way with compound words. Consider the powerful and successful section at the end of the Grendel-fight, lines 809-36. As the monster disappears into death and darkness, the poet remarks of Beowulf:

  Nihtweorce gefeh,
ellenmæþum. Hæfde East-Denum
Geatmecga leod gilp gelæsted...
(827-9)
[He rejoiced in his night's work, the fame his courage won. The man of the Geats had fulfilled his boast to the East-Danes.]

Why 'East-Danes'? The same people were 'North-Danes' forty lines before (783), 'South-Danes' before that (463), and 'West-Danes' on Beowulf's arrival (383). They have been 'Spear-Danes' and'Ring-Danes' and 'Bright-Danes' too, but at least you can be all those things together, while to most people 'east' definitely precludes 'west', as 'north' does 'south'. Of course East has been put into line 828 only to make the necessary alliteration with ellen. That is the only reason for many of the poem's other compound forms, and for many more it is the strongest reason. A few lines above East-Denum the poet says that Grendel knew his end had come, dogera dægrim. This phrase is entirely tautologous, meaning literally 'the day-number of his days'. A more sensible word would be dogorgerim (used at line 2728) but in line 823 that would not scan. Dæg is thrown in, then, to make up the weight. One could balk at line 820 as well, when Grendd has to escape, feorhseoc fleon under fenhleðu [flee life-sick under fen-slopes]. This is a good line, in grim contrast with the futile purpose of the next phrase, 'to seek a joyless dwelling'. But fen-slopes'? A hlið is something steep, cliff or hillside or even wall. Fens, however, are flat. It is hard to resist the thought that the poet wanted a compound word which would scan, imply inaccessibility , and above all alliterate on f; he created fenhleðu, then, by a kind of double analogy with the misthleðu [misty slopes] of line 710 and the fenfreðu [fenfastnesses] of 851. The word sounds all right, but it is not meant to stand close examination.

This is not to say that the Beowulf poet never used words imaginatively. But he did accept an element of redundance and tautology as part of his style (just as he saw no sin in heroes drinking). Compounds like heal-reced and heal-ærm (they both mean 'hall-building' or even 'hall-hall') are then neither tributes to the vividness of his imagination -- though they form part of Professor Brodeur's statistics of originality -- nor signs of intellectual barrenness. They are, like the poet's many indistinguishable synonyms, simply functional; and their function is to create metric or alliterative pattern.


These questions (and several others) have in recent years been almost literally bedevilled by the discovery of a third stylistic feature of Beowulf -- the formula, sometimes called the oral-formula. The facts behind this phrase can be briefly stated. Through the whole of Beowulf about one half line in six will be repeated more or less exactly elsewhere in the poem: examples include hine fyrwyt bræc, already discussed, or in the lines just quoted ellenmærþum and þe hie ær drugon. Very many more will be repeated with slight (or great) changes: thus nihtweorce gefeh (827) is at least rather like secg weorce gefeh (1569), while ond for þreanydum (832) is virtually identical with ac for þreanedlan (2223). Any search for repetitions in Beowulf uncovers vast but not easily organized sets of resemblances; and these once more offer radical challenge to concepts of originality, precision, the poetic mind. Thus to many the epithet nacod niðdraca in line 2273 seems beautifully calculated to evoke the monster's obscenely and unnaturally leathery hide, the famous armour-plating of the longworm. On the other hand the poet alliterates nið and nacod elsewhere, in line 2585, where he is talking about swords, and he has a whole range of similar phrases for the dragon, some of them ('dangerous fire-dragon', 'terrible earth-dragon') no more tightly appropriate than Grendel's fenhleðu. You can, in short, regard the brilliant stroke as one more accidental overlap of two repetitive systems; the formulas turn out as functional as the compounds and the variations, and as artistically neutral. Some critics welcome these implications as a gain in precision. To others the whole line of thought acts as intolerable provocation to their most cherished notions of individual art.

The problem is ours, not the poet's. Once again the overwhelming temptation for modern readers is to bring to the poem preconceived ideas of beauty and worth, to insist on praising it for what we have already decided is praiseworthy. Beowulf lends itself to these arguments, as does to symbolic interpretations, ironic interpretations, even the neoclassic interpretations of days gone by. That does not make any of them right. Furthermore we must fear that while we pursue sterile controversies of art versus accident and oral versus literate, we are shirking the real problem, which is to see what the poet would himself have found stylistically admirable. In pursuing this objective, probably the most useful thing we can do with compounds, variations and formulas is jettison them; and with them our quasi-autobiographical curiosity about mode of composition. The latter cannot be satisfied. As for the features of style already discussed, they are prominent and they do tell us something, but the most useful thing they tell us is that for the underlying charm and power of the poem -- as distinct from minor local lapses or neatnesses -- we shall have to look elsewhere. The style of Beowulf is clearly functional. But we want to know what aim these functions serve.

Marijane Osborn, in her 1983 book, pages 123-4, has the following information:

     The Old English verse line, usually printed in two separate half-lines in the editions, normally contains four beats or stresses, with two stresses and at least one alliteration in each half-line. The third stress of the four-stress line is called the head stave because it always carries the alliteration, as in these two lines:

1785Glád at heárt, the Géatish prince
 went báck at ónce as the wíse king báde ...

Glad alliterates with Géatish, the third stressed word of the line, and once alliterates with wise. From this we see that the sound is significant, not the alphabetical letter, for o alliterates with w. Went does not carry the metrical alliteration because it is not stressed. In this second line there is additional alliteration on the b's of the first and fourth stresses, a feature that occurs in the original at points of narrative importance, such as the first line of the poem: "we gárdéna in geárdágum." In this line the main alliteration is on g, occurring in the head stave, the third stressed syllable, and there is also secondary alliteration on d. In the Old English verse line the head stave may alliterate with either of the two previous stressed syllables, or with both, but not with the last; in formal Anglo-Saxon versification, to alliterate with the same sound on both head stave and foruth stressed syllable would be a mistake.

In my translation I have not been so rigid; I frequently alliterate not on the head stave but on the last stressed syllable, as throughout the following passage:

2688Then to the attack for the third time
 rushed the fierce fire-drake, intent on his feud,
 charging at that hero when he saw the chance,
 raging with fire, gripping him around
 the neck with his terrible fangs. And now
 Beowulf's life-blood drenched his body.

On the whole, alliteration on the final stressed syllable is a heavy rhetorical technique that is best reserved for intense or exciting passages, like this, where the content can absorb it.

     My principle for representing the metrical form of the poem is to produce a four-stress line with at least one alliteration connecting the first two stresses with the last two. But I have adopted various stratems and emphasized others that are used in Old English in order to tone down the force of this feature for modern ears. For example, I may alliterate on an internal syllable, as on two of the lines above: attack and time, 2688, and raging and around, 2691. Or I may alliterate on a syllable of a relatively lesser stress, as in the second line below:

1776                    ....when the ancient foe
 cáme on his éndless vísits to cáuse me
 immeasurable grief.

Here alliterative came is one of the four stressed syllables in line 1777, but it receives less emphasis than does non-alliterating endless (which is linked through assonance with ancient and immeasurable). Similarly, in line 1759 the alliteration is subdued syntactically:

1758Shield yourself from conflict with sin,
 dear Beowulf, by choosing what is better.

Here the word Beowulf contains the stressed syullable alliterating with better, but is deemphasized by being in a parenthetical clause. In the first line above, 1758, the alliteration is on sh and s; this is not acceptable in Old English prosody, though alliteration between one vowel and any other, strangely enough, is:

758No wise man among the Shieldings
 had ever expected that anyone
 could break that beautiful antlered building.

     In addition to muting the alliteration by such methods as these, I have used other expedients such as breaking the rhythm, usually by relocating the caesura as in lines 2692 and 1759 above, or by proceeding from many syllables in one half line to very few in the following. So long as the four stress rhythm is maintained, the number of syllables in a line is variable.


SydAllan@gmail.com -- https://docs.google.com/leaf?id=0B2H1mtZZCfMDYzIxNzhhOGMtZmE5ZC00YzQyLTgxMWMtNTA5ZjNjYmIwMmMx&hl=en