The Anglo-Saxon Chronicle 
edited and translated by M.J. Swanton.
Dent, 364 pp., £20, June 1996, 0 460 87737 2
Show More
Show More

In some respects The Anglo-Saxon Chronicle is a classic Post-Modern text. For one thing, it does not exist. It is a ‘construct’ of much later historians, obsessed with the discovery/invention/creation of a ‘national Chronicle’ as opposed to ‘merely local annals’, to quote the most influential of them, Charles Plummer, whose 1899 edition of The Anglo-Saxon Chronicle has still, significantly or ominously, not been replaced. Since the Chronicle is a Post-Modern work, even this brief account contains slurrings or inaccuracies, but one could press on by saying that even if it didn’t exist before, it certainly does now. No modern historical work on the period is without its long index entry on Anglo-Saxon Chronicle, while the libraries of the world contain scores, if not thousands of books with that title on their spines, the product of equally large numbers of scholars. So of course ‘The Anglo-Saxon Chronicle’ exists. You could argue that it is a product of later English scholars rather than of Anglo-Saxons, but you could not deny that Anglo-Saxons wrote it, or them, or at least the words out of which it has been made. So, to put it Post-Modernly, what is this Anglo-Saxon Chronicle, and what do we mean by ‘wrote’? How is the complexity which underlies the familiar three-word title to be presented in a manner true enough to be useful and simple enough to follow?

The last problem has not yet been solved and may with present technology still be insoluble. Seven complete manuscripts of what we call The Anglo-Saxon Chronicle currently exist, along with a fragmentary eighth, but none, at least of the five major manuscripts, constitutes any more of a unity than the Chronicle itself. Take the ‘E’ manuscript, the ‘Peterborough Chronicle’. We know the manuscript we have was copied in Peterborough in the year 1121, as it is written out in the same handwriting up to that date. Since it is known that Peterborough Abbey suffered a disastrous fire in 1116, it is usually thought that the copy we have was written to replace a manuscript destroyed in the blaze. However, the 12th-century monk copying away in his scriptorium no doubt saw the fire as a (perhaps literally) heaven-sent opportunity to get the 12th-century Peterborough version of events accepted as part of an authentic-looking seventh-century text. His version of the Chronicle, and his alone, contains a long account of the foundation of the abbey by King Wulfhere in 656, with a firm statement by the King to the effect that within the boundaries given ‘nan man ne haue pær nan ofsting buton seo abbot ond se muneces’ (‘nobody shall have any authority there except the abbot and the monks’), that ‘pær ne be numen of na geld na gaule buton to pa munecan ane’ (‘neither tax nor rent shall be taken from it except for the monks alone’), and that it should be regarded on a par with Rome as a place of pilgrimage. There is no doubt that this is 12th, not seventh-century work: apart from the obvious self-interest, no real Anglo-Saxon would have put a masculine singular definite article in front of a plural noun (‘se muneces’), nor produced the strange conflation of singular and plural, nominative/accusative and dative, seen in ‘to pa munecan ane’. Those words were written at a time when the Old English inflectional system was breaking up, or had broken up, though the copyist made no effort to modernise the archaic forms of his original.

If the date of original composition could so easily be erased every time a manuscript was copied, are any annal numbers or any palaeographical indications in any of the manuscripts trustworthy? The general view, and the very foundation of the ‘national Chronicle’, is that the concept of The Anglo-Saxon Chronicle – it has already become a concept not a text – was arrived at in the last years of King Alfred’s reign, as a record of the victory over the Danes and the beginnings of a true national state. This relies on the fact that the ‘A’ manuscript is written up to the year 891 in one hand. Alternatively, did Alfred issue instructions, say, in the year 890, for a record of his deeds and those of his ancestors to be created for posterity, with continuous circulation of authoritative supplements – ‘bulletins’ is the term Professor Swanton uses – to keep all copies up to date and all in line? Or did one man in Winchester, like the later monk in Peterborough, simply copy out a good deal of pre-existing and perhaps heterogeneous material, and cast it onto the waters of chance?

This gives a glimpse of the problem, but the true situation is almost unimaginably more complex. There are five major manuscripts in Old English. Several Latin histories, however, clearly borrow from manuscripts of the same type, though none of these authors seems to have consulted any of the manuscripts that still exist. Each of the five manuscripts, whatever the number of the scribes who copied them and the unity or otherwise of their handwriting, is the product of many different annalists writing over a period of two to three hundred years. In general outline and particular wording they are very like each other – if they were not no one would ever have thought to call this composite ‘the’ Chronicle – but almost no year annal of any importance is identical to all its other versions.

The ‘A’ manuscript of the annal of 878 says: ‘And that same winter a brother of Ivar and Halfdan was in Wessex in Devonshire with 23 ships, and he was killed there and 800 men with him and 40 men of his war-band.’ The ‘E’ manuscript adds: ‘and there the banner which they called “Raven” was taken.’ In the Annals of St Neots is a long paragraph to the effect that the Raven Banner was woven within one night by the three sisters of Hynguar (or Ivar), and that it predicted victory by flapping its wings, defeat by hanging motionless. ‘B’, ‘C’ and ‘D’, like ‘E’, mention the banner, but Asser’s Life of Alfred and the Chronicon of Æthelweard, an 11th-century alderman, reveal no more about the unnamed brother of Ivar than ‘A’ does. Does that prove that an addition was made to a lost copy of the Chronicle written after ‘A’ and copied into ‘B’, ‘C’, ‘D’, ‘E’ and St Neots? Or was the magic banner there all along but accidentally deleted? Alternatively, was the author of St Neots using a different source: the proto-saga in Old Norse about the descendants of Ragnar Hairy-Breeks, which survives as two late texts as well as a famous poem?

Every fact about the Chronicle seems to promise a revelation about its origins, which then trickles away in a dozen dispersing streams of doubt. Æthelweard’s Chronicon preserves a sentence missed out of all versions of the Chronicle’s annal for 885 – obviously by eye-skip, so Æthelweard must have had a prime original version! No, he mustn’t: he could have had a later copy revised and updated. The ‘B’ manuscript ends with the words ‘then he [King Edgar] died, then Edward, Edgar’s son succeeded, and held’ – the sentence breaks off with room still available on the page. That shows – doesn’t it? – that this text was written after 8 July 975, the day Edgar died, and before Edward was murdered on 18 March 978: the annalist knew Edward was king but didn’t know how long he would live. Not at all, Professor Janet Bately replies: it would be very unusual for an annalist to start reckoning the lifespan of a young king as soon as he’d succeeded. More likely, the text was written after Edward was known to have been murdered, and the annalist left a blank because he had not yet checked the exact duration of his reign.

As the data and the cross-references multiply like snarls in a ball of string, the hypotheses of origin become more complex. Janet Bately’s stemma needs a minimum of eight hypothetical manuscripts to explain the five-plus we in fact have. Audrey Meaney has six and Charles Plummer about eight, though he does not reduce his theories to a diagram. All those stemmas, though different, do at least look like each other and all start off from the Alfredian initiative and dispersion. Cyril Hart, by contrast, offers a much more economical thesis, which drops King Alfred and proposes instead that the Chronicle didn’t begin to be disseminated till more than a century later.

This inability to decode or unscramble the chain of transmission of a well-known and by no means hopelessly over-recorded text does look like a failure of technology. Printed texts destroy information: however elaborate their apparatus, they make it impossible to see where handwriting changes, where an author is evidently squeezing words onto the page or leaving blanks, where elaborate capitals have been inserted or where annal numbers have been written, perhaps by guesswork, into the margin of a continuous text. Moreover, even major research libraries cannot provide clear texts, edited according to identical principles, of all the Chronicle manuscripts. It often seems that the only way to make much sense of the way the manuscripts differ would be to get all of the originals together, lay them out on the floor and crawl from one to the other trying to see what their scribes have done. An alternative research strategy is that of the ‘Collaborative Edition’ of the Chronicle launched by Boydell and Brewer in 1983, and projected to run to 23 volumes, of which six have appeared, with five more in the pipeline.

But maybe print is just not the medium for this pre-Modern/Post-Modern Platonic idea of a document? Maybe instead the Chronicle has always been crying out (silently) to be a hypertext? Certainly, even more attractive than the crawl-round-the-floor method, would be to have every manuscript edited electronically, with all the suppressed information of printed texts available on sidebars, and the ability to ‘toggle through’ from one text to another, setting annal against annal and phrase against phrase at the touch of a key. Nice idea. But would it work? And would it be the empowering and liberating experience which technophiles promise? Some say that hypertext is an even more insidiously imprisoning technology than print: with it, you are under the control of the designers of the software and what they decide to be relevant, and are even less likely to notice it than you are to notice the results of editorial decisions in printed texts. Be that as it may, there is, as yet, no hypertext Chronicle.

The more important reason for taking an interest in The Anglo-Saxon Chronicle is that it is, to quote Swanton’s Introduction to this new Everyman translation, ‘the first continuous national history of any Western people in their own language’. It is the major source of information about pre-Conquest England; it is fascinating linguistically and stylistically; and it is different. Unlike the Latin Frankish chronicles, with their everlasting victories of the supradictus praeclarus rex against the Saxons (so often belied by the fact that he has to go back and repeat them next year), the late ninth-century Chronicle tells a grim, laconic, credible story of misunderstanding, disobedience and determination muddling through to a messy draw: an account that seems closer to Dark Age realities than any other surviving text. The shire aldermen do not obey their orders; vital forts are left ungarrisoned; relieving armies turn up too late; besieging armies bloody-mindedly raise the siege and go home because – and this is just the point where the original handwriting of ‘A’ stops – ‘they had completed their call-up, and had used up their food.’ At the same time, some things work. The English develop an ‘army in being’ strategy: don’t fight the Vikings, hang off them in force so they daren’t leave their ships unguarded. They try a ‘scorched earth’ policy and river-blocking; they try to challenge Viking supremacy at sea by building bigger ships to offset less competent sailors. The Vikings respond by changing their strategies, too: ‘they went through the forest in gangs and mounted groups, on whichever edge was without an army, and almost every day they were sought by other groups both from the army and also from the strongholds, either by day or by night.’ ‘B’, ‘C’, and ‘D’ take the second ‘groups’ in that sentence to mean ‘peoples’ or ‘armies’, a sensible statement but one which fails to get the immediate tactical point of (in modern terms) small-scale patrolling to dominate no man’s land. The ‘A’ manuscript more relevantly and less predictably reads floccum not folcum, ‘flocks’ not ‘folks’, or ‘platoons’ rather than ‘battalions’. On the other hand, both ‘A’ and ‘B’, unlike ‘C’ and ‘D’, have missed out the words ‘either by day’, though they are demanded by the context. Even the earliest copyists, it seems, had trouble with a text so little like any other and so unconditioned by Latin convention.

It is not surprising, therefore, that modern histories of the period often offer little more than a glossed text of the Chronicle. It tells a story we can understand. But perhaps this is a product of deliberate art rather than naive recording? Few suggestions have caused more historical fury in recent years than the argument that the Alfredian ninth-century annals are propaganda, written under the King’s direct control if not by the King himself, to set out as established fact the aims of the Wessex dynasty. No other English king has any legitimacy (the Mercian king is just a ‘foolish king’s thegn’ and a Viking puppet). No other member of the Wessex dynasty has any legitimacy (Alfred’s elder brother’s son Æthelwold is a traitor; Alfred’s own son Edward gets no credit from the Chronicle for the victory at Farnham till his father is dead, only to be written back into the story years later by Æthelweard, a descendant of the traitor; Alfred’s daughter Æthelflæd is mentioned by ‘A’ only when she dies, though ‘C’ and ‘D’ make it clear that she fully merited her title ‘Lady of the Mercians’). Alfred’s statesmanlike diplomacy is completely successful, except for one outburst of indignation from the ‘A’ chronicler which gives the game away. These queries make the Chronicle more interesting and potentially more valuable for historians, the ripples on the water hinting at powerful cross-currents underneath. Manuscript variation can expose contradictory opinions and Anglo-Saxon party politics, especially as the Chronicle moves on to the tenth and 11th centuries, as Scandinavian factions develop to match the English ones; as partisan chroniclers write irreconcilable accounts of the same event; and as one chronicler at least gives up in despair and writes an account of the years of Æthelræd’s reign which has for ever confirmed his nickname of Unræd, ‘Unready’ or, more accurately, ‘No-Idea’.

The Chronicle is, then, a vital and compelling text, made more so by its composite nature. But how is this to be conveyed? Plummer edited Two of the Saxon Chronicles Parallel with Supplementary Extracts from the Others, basically ‘A’ and ‘E’ on facing pages. A similar if more complex layout was used by Dorothy Whitelock in her English Historical Documents of 1954-5; her and Susie Tucker’s Chronicle translation from that work was revised and reissued separately in 1962. It has long been unavailable, and Garmonsway’s Everyman translation of 1953 simply follows Plummer, from the last century. A new translation which made the work more accessible and took account of modern discussion would therefore fill a gap, at least till the hypertext revolution comes: this is what Swanton has tried to provide.

The layout remains the familiar one of facing pages with inserts. On pages 60-1, for instance, we have ‘A’ facing ‘E’ for the years 817-29; the two chronicles are almost identical. Pages 90-1 are almost entirely from the long and detailed account of 896 in ‘A’, with mere inserts recording Alfred’s death in 899 from ‘D’ and ‘E’. Thirty pages later, the ‘Death of Edgar’ poem from ‘A’ faces what may well be a retrospective account written years after 975 by Archbishop Wulfstan from ‘D’, with a simpler version (the original or an abbreviation?) in ‘E’. Another 30 pages on, Æthelræd’s son Edmund Ironside is fighting King Canute in ‘D’, ‘E’ and ‘F’, and 30 pages later still, the contending chroniclers of ‘C’, ‘D’, ‘E’ and ‘F’ nevertheless agree in recording the refusal of loyal and rebel Englishmen alike to ‘fight against men of their own race’. One can see that the internal variety of the Chronicle is getting fair treatment, and the prose translation is clear and consistent, the latter a valuable quality when one is dealing with variant texts. It may seem needlessly fussy always to translate here as ‘raiding-army’, even when it obviously indicates the East Anglian defence forces only. But then ‘F’ uses here while ‘E’ sticks to fyrd: it may mean nothing, or be a simple mistake, but trivialities in the Chronicle often turn out to be relevant.

The translator remarks wrily in his dedication to ‘the saints and demons of the word processor’ that ‘while to err may be human, to really louse things up it takes a computer.’ I can understand what he means: it would not be too difficult a business to hand-write a multiple translation with the approximate layout of one’s final printed text, but given the ability of word processors invisibly to shift you onto another page while you think you’re still writing a footnote, or to change the pagination at the speed of light (but not your internal references to it), I cannot imagine how a text like this could be written on a word processor, or successfully transferred from that to print. And though it has been done, there is a price. In particular, the bibliography – full though it is of both up-to-date and highly recondite material, especially on archaeology – contains no mention of several of the major discussions of Chronicle origins. Bately’s Texts and Textual Relationships is there, but not her ‘Manuscript Layout’. Hart’s two volumes of Early Charters are there, but not his cat-among-the-pigeons ‘B Text’. No work by Audrey Meaney is mentioned, nor is Swanton’s major predecessor, the Whitelock and Tucker translation of 1962. One can probably see how this happened, for the bibliography ‘is confined to references used in footnotes’, and does not take in material used in the Introduction. Was Swanton using, to build his bibliography, some kind of tagging program for footnote references, deciding eventually not to interleave his Introduction references? Did he perhaps then have to cut down his Introduction, so that vital discussions – the first and most obvious things one would normally think of including in a bibliography – wound up falling between the editorial or electronic cracks? Some such explanation seems the likeliest, and if true it provides some excuse for omissions (and errors) which shrink the value of what will probably be for decades the most accessible modern version of the most important early vernacular European historical text.

In contrast to manuscript, print standardises and so destroys information; in contrast to both print and manuscript, the word processor eases and disguises revision. As one of Swanton’s many interesting plates shows, any fool can see when a manuscript chronicler is writing in an afterthought. But a word processor smoothes it over ‘swa heo no wære’ (‘as if it had never been’), even faster than the Peterborough chronicler. So mistakes are made which no one would make with simpler technology. In the dark all cats are grey, and all floppy disks look the same, whatever their contents. Scholars need to learn new protocols of care for Post-Modernity.

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