Skip to content

Old English Poetry Fifteen Essays

page 11 note 1The Wanderer, ed. Dunning, T. P. and Bliss, A. J. (London, 1969), p. 113.

page 11 note 2The Interpretation of Old English Poems (London and Boston, 1972), pp. 118–19.

page 11 note 3 The word ‘linguistic’ is used in its ancient sense – ‘of language’ – and has no reference to its present-day use by practitioners of a ‘science’ which has hijacked the word and which in many of its aspects will (I believe) prove to be one of the great non-subjects of the twentieth century – though I do not deny that it has valuable techniques in the analysis and teaching of current languages. It is with pleasure and gratitude that I acknowledge my debt to Professor Peter Clemoes for his cogent criticisms of earlier drafts of this paper.

page 12 note 1N&Q 215 (1970), 115.

page 12 note 2‘Lexicography and Literary Criticism: a Caveat’, Philological Essays: Studies in Old and Middle English Language and Literature in Honour of Herbert Dean Meritt, ed. Rosier, J. L. (The Hague, 1970), pp. 99–110.

page 12 note 3 But see below, Postscript, p. 28.

page 13 note 1The Metre of ‘Beowulf’ (Oxford, 1962), pp. 123–7.

page 13 note 2‘Some Syntactical Problems in The Wanderer’, NM69 (1968), 190–1.

page 13 note 3 This conclusion was reached simultaneously and independently by Dunning, and Bliss, (The Wanderer, pp. 21–3) and by Peter, Clemoes, ‘Mens absentia cogitans in The Seafarer and The Wanderer’, Medieval Literature and Civilization: Studies in Memory of G. N. Garmonsway, ed. Pearsall, D. A. and Waldron, R. A. (London, 1969), pp. 74–5.

page 14 note 1‘The Art of the Singer: Three Old English Tellings of the Offering of Isaac’, Old English Poetry: Fifteen Essays, ed. Creed, R. P. (Providence, R.I., 1967), p. 80.

page 14 note 2The Homilies of the Anglo-Saxon Church: the First Part Containing the Sermones Catbolici or Homilies of Ælfric, ed. Benjamin, Thorpe, 2 vols. (London, 1844–1846) (cited henceforth as ‘Thorpe’) 1, 312, line 34.

page 15 note 1The Wanderer, p. 106.

page 15 note 2 Thorpe 1, 290, line 12.

page 15 note 3The Wanderer, p. 106.

page 15 note 4Anglo-Saxon and Norse Poems, ed. Kershaw, N. (Cambridge, 1922), p. 162.

page 15 note 5The Wanderer, p. 108.

page 15 note 6 ‘The Metrical Epilogue to the Old English Version of Gregory's Cura Pastoralis’, NM 70 (1969), 382, n. 4.

page 15 note 7Beowulf and the Fight at Finnsburg, ed. Klaeber, Fr., 3rd ed. (Boston, 1936), p. xciii.

page 16 note 1Ibid. pp. 186 and 221.

page 16 note 2Zandvoort, R. W., ‘Is Aspect an English Verbal Category?’, Contributions to English Syntax and Philology, ed. Behre, F., Gothenburg Stud, in Eng. 14 (1962), 19.

page 16 note 3 I must now say (June 1974) that I am convinced by F. C. Robinson's defence of S. O. Andrew's explanation of bwil dæges in Beowulf 1495b as ‘daytime’; see Old English Studies in Honour of John C. Pope, ed. Burlin, Robert B. and Irving, Edward B.Jr (Toronto, 1974), pp. 121–4. But this does not affect my point; as Robinson rightly says, most critics have explained bwil dæges as either ‘the space of a day’ or ‘the large part of a day’.

page 17 note 1 2nd ed. (Oxford, 1968), § 197.

page 17 note 2First Readings in Old English (Wellington, 1948), § 40.

page 17 note 3Sweet's Anglo-Saxon Primer, 9th ed. (Oxford, 1953), § 92.

page 17 note 4An Old English Grammar, 2nd ed. (London, 1958), § 127.

page 17 note 5An Historical Syntax of the English Language II (Leiden, 1966), § 808.

page 18 note 1 ‘The Old English Wife's Lament: an Interpretation’, NM 71 (1970), 588–9.

page 18 note 2 To economize in space I give only the line references to the numerous illustrative passages from Beowulf which follow. So those accompanying me further will need a copy of the text. I ask their indulgence.

page 18 note 3 On the use of a past indicative to refer to a future perfect, as in The Ruin 9, see my ‘Some Problems of Mood and Tense in Old English’, Neopbilologus 49 (1965), 44–6.

page 19 note 1Ælfrics Grammatik und Glossar, ed. Julius, Zupitza (repr. Berlin, 1966), p. 123, lines 15–16.

page 19 note 2Ibid. p. 124, lines 1–11.

page 20 note 1 I have drawn attention to two specific examples of this fallacy in operation in the sphere of Old English syntax in ‘Two Syntactical Notes on Beowulf’, Neophilologus 52 (1968), 297, and ‘The Narrator of The Wife's Lament’, NM 73 (1972), 224, n. 4.

page 20 note 2An Historical Syntax 11, § 810.

page 21 note 1 We may note here Davis's observation (Sweet's Primer, §92) that ‘even the form with bæfde Sometimes has the sense of a simple past’.

page 21 note 2 (i) In principal clauses (15): 205, 665, 743, 825, 828, 883, 893, 1294, 2321, 2333, 2381, 2397, 2844, 2952 and 3046; (ii) in ac clauses (2): 694 and 804; in an ond clause (I): 2707; after nealles (I): 2145; (iii) in parenthesis (I): ?2403; (iv) in adverb clauses of time (6): 106, 220, 1472, 2104, 2630 and 3147; (v) in other subordinate clauses (6): 117, 1599, 2301, 2726, 3074 and 3165; (vi) expressing impossibility in the past (I): 1550. Total 33.

page 21 note 3Ælfrics Grammatik, ed. Zupitza, , p. 124, line 9.

page 21 note 4 Thorpe 1, 148, line 10.

page 22 note 1 The evidence which leads me to this belief is based on a long and complicated argument which is intended to form part of my Old English Syntax, now in progress.

page 24 note 1 See The Wanderer, pp. 112–13 and Interpretation, pp. 118–19.

page 25 note 1Essays and Poems presented to Lord David Cecil, ed. Robson, W. W. (London, 1970), p. 33. The line numbers are those of Dame Helen's translation.

page 25 note 2 ‘Old English pæt an, “only”’, NM 68 (1967), 286.

page 25 note 3 ‘Some Syntactical Problems’, pp. 182–7.

page 25 note 4The Wanderer, ed. Leslie, R. F. (Manchester, 1966), p. 74.

page 26 note 1Interpretation, pp. 118–19. (Perhaps I may say that the logic of the Dunning–Bliss argument about the unlikelihood of the wanderer burying a second dead lord eludes me.)

page 26 note 2 See my ‘The Narrator’, pp. 224–6.

page 27 note 1 See ibid.

page 27 note 2 ‘Two Non-Cruces in Beowulf’, Tennessee Stud, in Lit. 11 (1966), 151–5.

Received wisdom has it that the Beowulf poet put together his poem halfline by halfline (“verse” by “verse”). My work on the poem over the past fifty years has led me to think that we can begin to understand how the poet composed his tale, clause by clause, only if we turn our attention to the whole lines in which he told the story.

The poet built each four-measure-line—and each of the rare five- and six-measure lines—around the alliteration of the root syllables of stressed words. His tradition seems to have provided him with many alliterating word pairs that encapsulate culturally significant ideas. For example, the poet built five lines around the pair dom (achievement) and deað (death)—dom before death, at least seven lines around the pair eorl (nobleman) and ellen (brave action), and nine around the pair soþ (truth) and secgan (say). This does not mean, however, that the poet was constrained to frame each clause within the confines of a single alliteration: rather, he composed many passages with suppleness and flexibility simply by beginning a new clause in the middle of the line. This expedient left him free to develop the clause around different alliterations.

The rhythm of the poem is based on the stress patterns of the poet’s language; most Old English words, like ellen or secgan, begin with a heavier stress and end with a lighter stress. This pattern translates to a downbeat followed by an upbeat, the simplest kind of rhythm. Each measure of the poem repeats this rhythm. Yet there is no question of monotony: though all measures are identical in rhythm and theoretically identical in the length of time it takes to speak them, successive measures are likely to contain very different combinations of stressed and unstressed syllables—and even precisely timed rests. Thus there is a great variety created both by the material that fills each measure and by the succession of different types of measures. The material within the measures makes possible only seven different types of measure. But the various combinations of stressed and unstressed syllables, along with measure-initial and measure-final rests, produce about fifty different subtypes. The rich variety of these subtypes is the source of the complexity of the poet’s prosody. [End Page 214]

What was, perhaps, of greatest use to the poet as he composed was the knowledge that many, though not all, of the subtypes manifest themselves as single words, compounds, and even short (measure-length) phrases. These words and phrases were the ready-made—and readily made—building blocks of the poet’s composition. So the poet did not have to rely on his ability to recall halfline or whole-line formulas—or even the few formulas that wrap around two lines. He had in mind what he needed: a prosody shaped by his lexicon, with which he could tell his story not only rapidly and with flexibility but even at times—often in this poem!—with virtuosity.

Robert Payson Creed

University of Massachusetts, Amherst

Robert Payson Creed

Robert Payson Creed is Emeritus Professor of English and Comparative Literature at the University of Massachusetts at Amherst. He has published widely on Beowulf and comparative oral traditions, including Old English Poetry: Fifteen Essays (1967) and Reconstructing the Rhythm of Beowulf (1990).


Creed 1959. Creed 1959
Robert Payson Creed. “The Making of an Anglo-Saxon Poem.” English Literary History, 26:445–54.

Google Scholar

Copyright © 2003 its author