Old English Meter and Linguistic Theory 
by Geoffrey Russom.
Cambridge, 178 pp., £25, August 1987, 0 521 33168 4
Show More
Show More

It is hard to know why the English should nowadays take so little interest in their Anglo-Saxon predecessors. Perhaps the main reasons lie in 20th-century history. The Victorian statue of King Alfred of Wessex which stands in the market square of his birth-place, Wantage, testifies to a pride in that great founding father which modern England no longer feels. We have shrunk, and Alfred has shrunk with us. Perhaps, too, the defeat of Nazism left Germanic origins under a cloud – especially in a multi-racial society where many do not share them. For whatever reason, anyway, the Anglo-Saxons strike most contemporaries as dull and blockish creatures. The name Alfred itself, so popular in Victorian times, has come to seem tattered and slightly ridiculous, along with many other Christian names of the period before the Norman Conquest (Edgar, Egbert, Oswald); and most children learn little or nothing about the King himself, perhaps only the (apocryphal) burning of the cakes. There are at present, as it happens, some excellent historians of Anglo-Saxon England – witness a Phaedon book, The Anglo-Saxons, edited by James Campbell (1982) – but the history of the period figures little in most school curricula. So far as the general public is concerned, it is only the archaeologists who have succeeded in striking a spark of interest. It would be hard, after all, to walk past the display of the Sutton Hoo treasure in the British Museum, or look at the pictures in Campbell’s book, without feeling that one’s monochrome image of the age must be somewhat inadequate.

Such reflections will be familiar to anyone who, like the present reviewer, teaches Old English (alias Anglo-Saxon) literature in a university. In those English departments where the subject is still compulsory, the requirement commonly arouses some discussion-and understandably so, not only because some students and colleagues share the general low view of the Anglo-Saxons, but also because other periods and branches of literature make their legitimate and competing demands for a place in the syllabus. Perhaps, so far as the university system as a whole is concerned, it would be best for certain places well-equipped for the subject to make Anglo-Saxon a requirement, while others (perhaps the majority) offer it as an option, or not at all. If such a distinction between English departments were clearly understood in schools and elsewhere, then there should be little question of Anglo-Saxon being forced upon students who had absolutely no capacity or use for it. The tricky question, of course, is why it should be ‘forced’ upon anyone. One answer is that for any university to keep Anglo-Saxonists on its strength, in present circumstances, there will need to be a sufficient, and a sufficiently steady, demand for their teaching: but if all universities make the subject optional and the demand for such an option is spread out over all of them, that condition might not be met – perhaps not anywhere outside Oxbridge.

This argument will not cut much ice with those who think that the rewards of reading Anglo-Saxon are insufficient to justify the effort spent in learning the language – except for students of an antiquarian or philological bent. For literary students, certainly, the prose and especially the poetry are both strange and difficult:

Hwaet we Gardena in geardagum
Theodcyninga thrym gefrunon.

But it ill becomes those who protest at the narrow horizons of Eng Lit to object to strangeness, and one hopes that they might not object to difficulty either; the essential question concerns the quality of Old English literature – and that means above all the poetry of the period: Beowulf, The Dream of the Rood, The Seafarer, and the rest. This is a matter on which no general consensus exists, but one claim can be made without much fear of contradiction from anyone capable of judging the matter: Old English poetry is very mature and assured in its metrical art – clearly more so than most Middle English and much modern English poetry. In that respect at least, it has the quality that one looks for in classics.

I know few better demonstrations of this than Geoffrey Russom’s Old English Meter [sic Cambridge University Press] and Linguistic Theory. All surviving Old English poetry (about 30,000 lines in total) employs essentially the same alliterative line: but no native discussions survive to record how the poets and audiences of the time thought about it. Modern scholars have had to work out its rules and constraints for themselves. In this process of conjecture and refutation, which goes back at least to the pioneering work of Sievers in the late 19th century, Russom’s book will surely be a landmark. It is a work of great intellectual power and elegance, deriving long chains of consequence from four simple initial ‘principles’ or hypotheses to pro duce ‘a coherent rule system for Old English poetry, a system that defines the norm and explains the limits of deviation’. It brings Chomskyan linguistic theory to bear on the problems, effectively and with a minimum of technical fuss.

From a metrical point of view, Old English poetry is not so much early English as late Germanic (one reason, some would say, for excluding it from an English syllabus). It differs from the poetry of Chaucer, Pope or Tennyson in three fundamental ways. It has no fixed number of syllables in the line (and so is not ‘octosyllabic’, ‘decasyllabic’ etc); it has no single pattern for the foot (and so is not ‘iambic’, ‘trochaic’ etc); and it does not rhyme (hence no ‘couplets’, ‘quatrains’ etc). The scribes of the manuscripts in which the poems are preserved write them out, as we would say, ‘as prose’: but everyone agrees that they fall into shortish units (minimum four syllables) linked together in pairs by alliteration. Older editors printed these as lines, which therefore formed what one might call alliterating couplets, but the modern practice is to print each pair as a single line composed of two ‘half-lines’:

Hwaet we Gardena in geardagum
Theodcyninga thrym gefrunon.

So much is agreed. And in practice it is not too difficult to get the rhythms of the lines more or less right, provided one knows which words are which, and what they are saying. Since there is no prior metrical norm such as the repeated foot pattern of the iambic pentameter, one can only try to read each half-line ‘naturally’ – that is, according to habits of stressing words and sentences which are evidently so deeply embedded in our language that they have changed little over the centuries. The alliteration helps with this, if help is needed, because there will regularly be one or two syllables in each first half-line alliterating with one (and only one) in the second, and these syllables always carry some degree of stress. Such alliteration needs stress because it is an essential part of the verse-structure, holding its parts together just as rhyme does in later times. It is not an optional expressive extra: it belongs to the metre, not to the rhetoric.

Reading the half-lines ‘naturally’ in this way produces results which, to ears accustomed to Chaucer, Pope or Tennyson, sound rhythmically quite unregulated – more like prose than verse. Each half-line seems to be a law unto itself, conforming to no regular pattern and usually not even matching the pattern of its other half. What principles of selection, if any, are at work here? This is the problem to which Russom mainly addresses himself. As he says, of Beowulf: ‘Sievers and others have provided detailed arguments showing that the poet avoided a number of imaginable verse [i.e. half-line] patterns. At present, however, no single set of rules distinguishing metrical patterns from unmetrical patterns has achieved consensus.’ His own proposed set of rules depends most upon the first of his four principles: ‘Foot patterns correspond to native Old English word patterns.’ This turns out to mean that, if one takes each half-line to consist of two metrical feet, divided in every case at a word boundary, then each foot will have the rhythmical shape of a word – that is, it will conform to one of the nine patterns of stressed and/or unstressed syllables to which Old English words and compounds themselves conform. The commonest of these patterns consists of a long stressed syllable followed by an unstressed one: so two words of that shape form what Russom regards as a norm for the half-line (e.g. furthur feran). Yet the verse can certainly not be called simply trochaic. Indeed, Russom’s analysis of the Beowulf-poet’s practice comes up with no less than 25 different half-line patterns, diverging more or less far from the trochaic norm (as against the five main types distinguished by Sievers).

Russom states the assumptions at work here as follows: ‘I assume that the native speaker of Old English would have possessed, as part of an internalised grammar, one set of rules specifying the word patterns attested in the language and another set determining the position of stress in a word with a given pattern. I also assume that a native speaker introduced to poetry in the normal way could identify metrical rules as analogues of linguistic rules already learned. Once the native speaker grasped the relation between language and general principles of verse construction, many corollaries that must be made explicit for a speaker of Modern English would have followed as a matter of course.’ The extraordinary thing is how many corollaries Russom convincingly derives from these assumptions. Thus he manages to explain certain specific and previously unexplained constraints on the half-line, especially in the matter of unstressed syllables (patterns, that is, that do not occur); and he shows that the distribution of alliterating syllables in the whole line can be explained as mimicking the same ‘internalised grammar’. His theory also supports the opinion expressed by J.C. Pope in another fine study, The Rhythm of Beowulf, that not all half-lines will have two stresses. Here and in general Russom’s arguments tell against what one might call thumping four-stress readings. Although his theory becomes quite elaborate in places, it is no more so than many theories required to explain ordinary speech-behaviour – upon which, indeed, the whole metrical system, as Russom understands it, rests. His book may be criticised for confining itself largely to the evidence of Beowulf, and perhaps for questionable handling of ‘extrametrical syllables’. Yet his ‘coherent rule system’ is just that, and it will certainly have to be taken fully into account by all future studies of the subject. It represents a remarkable attempt to uncover and display, even down to very delicate detail, the inner logic which governed the metrical practice of the Old English scop.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN


Please include name, address, and a telephone number.

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences