Classic work on Cain vs Abel

Why does Norton publish two translations of Beowulf - E. Talbot Donaldson's 1966 prose version and Seamus Heaney's 1999 verse version? The answer is because each has appeared in Norton's Anthology of English Literature , the leviathan of English textbooks, which since 1962 has dominated the US college market. In the seventh edition, published in 2000, Heaney's Beowulf replaced Donaldson's, just as in 1975 Donaldson's superseded Charles W. Kennedy's.

These volumes include selected criticism following the annotated text - a formula that meets a student's modular needs. While scholars favour texts unaccompanied by critical essays, students find such essays useful. But pre-packaged criticism can be biased. The bias is palpable in the seventh edition of the Norton Anthology , which has been "turned" to new historicism.

In US universities, "great books" are read in translation. Is half a loaf better than no bread? All translations are partial, and a translation in a set textbook tends to replace its original. While both Norton Beowulfs are accompanied by a retinue of critical essays and miscellaneous scholarly appendices - Donaldson's has 160 pages of these, Heaney's 210 pages - it is on the translations that these volumes deserve to stand or fall.

Donaldson translated Beowulf into plain and flexible prose, faithful to the sense of the original if not to its poetic style. Like many scholars, Donaldson preferred modest, accurate prose to risky poetic ambition. Accuracy was the first aim, the second readable modern English. Beowulf resists such processing, for its verse combines complex syntax with elevated diction and a riddling style. Donaldson translates the opening " Hwœt " as "Yes"; his Beowulf is true to its aim of an understated informal style. Yet I miss the fullness of nuance realised in the best prose translation - that by G. N. Garmonsway in Beowulf and its Analogues (1968). Garmonsway's version is absent from the bibliographies of both Nortons.

In the mid-1980s, Norton commissioned a translation from Heaney, then at Harvard. I, as a verse translator of Beowulf for Penguin Classics, reviewed Heaney's translation favourably, as a good poetic version - adventurous, stripped for action. Heaney and I make different stylistic choices: where he finds "forthright delivery", I find ruminative reserve; and my imitative metre is stricter. But from a scholarly point of view, the obvious difference between Heaney's Beowulf and all other versions is that Heaney put Irish references into the poem. His engaging introduction ends with a strong apologia for these Irish elements. Heaney chose what his Norton editor calls "a rural Ulster dialect". Those as interested in Heaney as they are in Beowulf will be guided by the poet's account of why he rendered " Hwœt " as "So". For Heaney, this monosyllable recalls the dignity of speech of some uncles of his, named Scullion. He also employs "graith", "hirpling", "wean" and "scaresomely" (Ulster locutions), and kesh and brehon (Irish words). All these are glossed in the Norton edition. Heaney refers to Hrothgar's hall as a "bawn", which he glosses in the Faber edition as "a fort for cattle". A "Translator's note" specifies further:

"Fortified outwork of a court or castle. The word was used by English planters in Ulster to describe fortified dwellings they erected on lands confiscated from the Irish."

Heaney concludes his introduction with an explicit defence of "putting a bawn into Beowulf ". As a poetic text, his version works well, but this translator's note stakes a claim. Ezra Pound put a "frigidaire patent" into his "Homage to Sextus Propertius" to show that he did not intend literal translation. Heaney puts a bawn in Beowulf , asserting a parallel where there is no connection. Beowulf is not about the English in Ireland but about whether mankind takes after Cain or after Abel. Insofar as it is historical, Beowulf is concerned with kingly conduct among the ancestors of English kings.

The preface by editor Daniel Donoghue argues that "all dialects have an equal claim on the remote origins of English because they all have a parallel history. The voice of the Scullions is as adequate to the task as received pronunciation or any other variety of English." Here Donoghue creates three confusions. Dialects may be linguistically equal but, being less generally understood than a standard language, themselves require glosses. Received pronunciation is an accent not a dialect. And not all stylistic registers are suited to render the style of all poems.

Michael Alexander is professor of English literature, University of St Andrews, and author of Beowulf: A Verse Translation .