Steve Watson

Sievers-Bliss

As part of my effort to create a performance-oriented version of Beowulf, I had to look into the proposed theories of metre for Old English. I thought I might as well put the summaries of what I found here. 

From a study of old Germanic and Scandinavian poetic metres, Eduard Sievers[1] derived a podic system in which each verse was required to instantiate one of a limited number of patterns of stress. With later modifications by Alan Bliss,[2] this system has gained wide acceptance.

According to this system the stressed and unstressed syllables of a verse must fall into one of just 5 patterns.

Type A:  / x (x x x x) / x
Type B:   (x x x x) x / x (x) /
Type C:   (x x x x x) x / / x
Type D:  / (x x x) / \ x
Type E:   / \ x (x) /

The parentheses indicate the number of unstressed syllables which optionally occur at that position.

The verses in a line may be any combination of verse types.

Notes:

1. Anacrusis may occur in lines of type A or D, thus:

Type A⁺: (x x) | / x (x x x x) / x
Type D⁺: (x x) | / (x x x) / \ x

2. The stress pattern of a verse may occasionally be augmented by what seems to be half of another Type added to its end. Such half-lines are called hypermetric verses and often come in groups of three.

Just because this system creates such a complete typology of line forms and is almost universally used even by those who are not committed to it as a metrical theory we shall include here a description of the metrical sub-types that the system recognises. Note that for brevity the optional unstressed syllables are left out of the descriptions of each type and subtype

A             / x / x
A1           Alliteration on the 1^st lift or (for the on-verse) on both
A2          \ replaces x in either foot or both; often has double alliteration
A2a         / \ / x     \ in 1^st dip, which may be a resolved syllable
A2b         / x / \     \ in 2^nd dip, which may be a resolved syllable
A2ab       / \ / \     \ in both dips, which may be resolved syllables
A3           Alliteration only on 2^nd lift
A3b         / x / \     \ in 2^nd dip, which may be a resolved syllable     

B             x / x /
B1           2^nd dip has one syllable
B2          2^nd dip has two syllables
B3          Alliteration only on 2^nd lift

C            x / / x
C1          1^st lift is not resolved
C2         1^st lift is resolved
C3         2^nd lift is short

D            / / \ x     Double alliteration is required and so this is always an on-verse
D1          \ is in 3^rd position and long
D2          \ is in 3^rd position and short
D3          2^nd lift is short and \ is long
D4          / / x \    
D*          Called Expanded as it has an extra syllable or two after the 1^st lift
D*1        \ is in 3^rd (equivalent) position and long
D*2        \ is in 3^rd (equivalent) position and short
D*4        / / x \    

E             / \ x /     Sievers’ sub-types here are no longer generally distinguished as such

Critics attack this system for being more complex than the source material would seem to justify and also for not giving any explanation for why the verses have to go into just those 5 patterns. Moreover, it is observed that there are grammatically and semantically correct lines of OE poetry which do not fit the scheme. The typical response to this is to assume that such lines are the result of scribal errors and to seek to correct them, but this is felt by some critics to be presumptuous: a theory should be made to fit the data, not the other way around.

[1] Sievers, Eduard (1893) Altgermanische Metrik. Sammlung Kurzer Grammatiken Germanischer Dialekte, Ergänzungsreihe, 2. Halle: Niemeyer

[2] Bliss, Alan Joseph (1958) The Metre of ‘Beowulf.’ Oxford: Blackwell; (1962) Introduction to Old English Metre. Oxford: Blackwell