This Old English agreement between King Alfred of Wessex (871–899) and Guthrum, the viking king of east Anglia (d. 889/90), cannot be dated with certainty. It established the boundary between their kingdoms and regulated relations between the English and Danish subjects of the two kings in criminal matters and procedures and warranty.

It was translated into Latin c. 1100: see Quadripartitus .



by Thom Gobbitt


The written record of a late ninth-century Anglo-Danish peace treaty – the Peace of Ælfred and Guthrum, henceforth the Peace – survives in two English copies and in early-twelfth-century Latin translation as part of the Quadripartitus collection. The two English versions of the Peace, different from each other in their total scope and the voice in which the terms of the treaty are written, both survive in the same early-twelfth-century manuscript, Cambridge, Corpus Christi College MS 383 (henceforth CCCC 383). The Peace has drawn a significant amount of attention in the scholarship on Anglo-Saxon law and history, with particular attention having been given to the boundaries of the Danelaw and the Ælfredian politics of English identity, and to trying to identify the Peace with the various truces referred to in the Anglo-Saxon Chronicle (See Dumville, 1992: 13). Patrick Wormald notes that the tone of the Peace is concise and forward looking, in that it more closely represents the legislative idiom of Edward the Elder, Æthelstan and Edmund, and contrasts with the traditional tone of Ælfred’s Domboc. As he further notes, it demonstrates the kind of legislation Ælfred could produce when circumstances required it (1999: 286).


Manuscript Contexts

CCCC 383 is a collection of Anglo-Saxon laws and related texts produced in the early years of the twelfth century, possibly at or for St. Paul’s Cathedral, London. The manuscript is copied throughout in a late vernacular minuscule by a single scribal hand, with rubrics, emendations and notes added by a separate scribal hand of the first half of the twelfth century, and another set of additions, including two further items in English on the final folio of the manuscript, added by a third hand dating to the second quarter of the twelfth century. The manuscript in its current form it comprises seven twelfth century quires in two blocks with at least two quires now missing, and interspersed with two further quires added in the sixteenth century. The order of quires in the first block is a little muddled, in that what is now the first quire should originally have followed after the end of what is now the third quire. This disorder of the quires dates back to at least the sixteenth century, as it Archbishop Parker’s red-crayon pagination reflects the current arrangement of the medieval parts of the manuscript. That the two medieval blocks were originally intended to be part of the same book seems apparent from the overall similarity in dimensions and mise-en-page, with folios measuring (after some trimming associated with the binding of the manuscript in the sixteenth-century) approximately 112-120 mm x 185-188 mm. A full description of the manuscript and its contents can be found on the Production and Use of English Manuscripts, 1060 to 1220 project website (Gobbitt, 2010).

An important feature of the manuscript is that the exemplars from which the scribe copied it appear to have been a series of ‘mini-collections’ of Anglo-Saxon laws and related texts. Patrick Wormald identifies these on the basis of groups of laws that can be seen to be transmitted together in multiple manuscripts, including the Textus Roffensis and some of the copies of the Quadripartitus (1999: 228-53). Wormald’s speculation regarding these mini-collections has been confirmed on independent, palaeographic grounds with Carole Hough having shown that the Textus Roffensis scribe copied many of the graph forms and orthographic choices accurately, thereby allowing the original exemplars to be identified (2001: 57-79). The same methodology applied to CCCC 383 also reveals the underling mini-collections, although here they are harder to discern as the strategy of the Corpus scribe is to update and normalise the palaeographic forms to the use of ð rather than þ, and caroline s instead of insular s. As the scribe’s consistency in updating the forms increases steadily as the copying of the manuscript progresses, the variant graph forms of the exemplar must sometimes be identified in the small proportion of graphs that were copied directly rather than updated (Gobbitt, 2012: 3-23).

Both versions of the Peace are copied in the text-block of CCCC 383 as single blocks of continuous prose, and the sub-clauses that Liebermann divides the text into have no manuscript authority for the vernacular versions at least. However, the sub-divisions of the text are reflected in some of the manuscripts of the Latin translation of the Peace. In both the earliest surviving copy of the Quadripartitus, Manchester, John Rylands Library, Lat. 420 (fol. 70v, l. 14 - fol. 71r, l. 14), dating to the mid-twelfth century and the slightly later London, British Library MS Additional 49366 (henceforth BL Add. 49366), fol. 86r, l. 17 - fol. 86v, l. 17, dating to the third quarter of the twelfth century, the Peace is written as a single block, but includes red pen-drawn initials at the start of each sub-clause. BL Add. 49366 also has some of the clauses numbered in roman numerals in the outer margins adjacent to the respective part, with the list of clauses (existing only in the Quadriaprtitus text, not the vernacular versions) labelled as ‘i’, the prologue and boundary clauses omitted, and the main details of the treaty numbered ‘v’ to ‘viii’. While this is clearly an innovation in the layout of the laws, facilitating reader cross-reference to specific legal content, it is not universally present in the Latin copies, and the turn of the thirteenth-century, London, British Library, Titus MS A. xxvii (fol. 133v, ll. 3-24) has the Peace copied in a single block without differentiating the clauses.

The two copies of the Peace in CCCC 383 comprise version 1 (edited by Liebermann as B2) on fol. 12v, ll. 1-26, and version 2 (edited by Liebermann as B) on fol. 57r, l. 17 - fol. 57v, l. 23. Taking into account the compilation of the manuscript from mini-collections of Anglo-Saxon law, these two Old English versions of the Peace were copied from different exemplars. If nothing else, this reflects the greater circulation of the text during the later Anglo-Saxon period than is otherwise attested in the other surviving manuscripts containing Anglo-Saxon laws. The replication of the Peace in CCCC 383, however, is an occurrence that Wormald dismisses as the oversight of a scribe unfamiliar with the materials he (or she) was copying and not realising that the text had already been copied some few quires previously (1999: 232). It should be noted, however, that while the second version was left without rubrics and received no additions or emendations prior to the sixteenth century, this is also true of most of the other adjacent texts copied in the same quire. Nevertheless, at least one of the other twelfth-century users of the manuscript seems not to have deemed the second copy to be superfluous, as the miniator supplied a pen-drawn initial Ð at the start of the item (fol. 57r, ll. 16-18). It should be noted here that the miniator did not passively add the decoration to the various laws, as elsewhere in the manuscript he or she used the initials to emend the presentation of various laws and, presumably deeming its content to be beyond the thematic remit of the manuscript, struck out the charm against cattle theft that had been copied two pages later, on fol. 59r, ll. 6-20 (Gobbitt, 2010; Gobbitt, 2012).

The first version of the Peace in CCCC 383 (fol. 12v, ll. 1-26) is the shorter, copied as a single block of text onto the left page of the central opening of a quire which, unusually, has six folios rather than eight. Considering the six lines originally left blank following the end of the preceding text on the recto of the folio the positioning of version 1 of the Peaceappears to have been carefully if subtly staged. This staging is made more apparent as the usual tendency in the manuscript is not to differentiate between the laws beyond an pen-work initial and perhaps a rubric, giving new texts no more attention than new clauses within the laws receive. Furthermore, it would appear that the reduction of the quire from eight folios to six may have also have been part of the staging of the text, as by removing the central bifolium it positions the Peace in the centrefold of the quire, the point at which the quire most naturally opens. Directly opposite it on the facing folio is the start of Pseudo-Eadward and Guthrum, a follow up law-code drafted by archbishop Wulfstan, with textual echoes that let it follow on from the Peace. Where the Peace begins, “This is the truce between King Alfred and King Guthrum [...]”, Pseudo Edward and Gurthrum continues it by beginning “And this is the law also of King Alfred and King Guthrum, and after of King Edward and King Guthrum”. The manuscript contexts, then echo and support the claims of the written texts and, within the plain framework of CCCC 383, give the Peaceand its purported continuation an additional degree of emphasis (Gobbitt, 2013).

In addition to the antiquarian comments, the first version of the Peace also receives a significant degree of attention by twelfth-century scribal hands, in the form of a rubric, and numerous comments and alterations in the margins and interlinear spaces, including glosses to update words and corrections of orthography. Many of these additions modify the contents of the shorter text, by supplying information from or correcting it against the second version of the truce. The second version of the Truce (fol. 57r, ll. 19 - fol. 57v, l. 23) is also written as a single block of text, divided over the two sides of the folio, and with a majuscule in the hand of the main scribe marking the beginning of the two sub-clauses. Only the first of these, marking the beginning of the boundary clauses is represented with a majuscule in the corresponding place in version 1. While the second version of the Peacedid not receive many alterations or additions by readers during the twelfth century, then, it clearly was being read and used. The staging of the second version is not so pronounced as the first, although it should be noted that it is one of the few laws in CCCC 383 where line space was left blank, perhaps for a rubric, immediately preceding the new text (fol. 57r, ll. 14-15). As noted previously, the rubric was not subsequently supplied (Gobbitt, 2013). It remains an interesting irony that modern scholarship has preferred the latter text to the former (note that Liebermann edits it as ‘B’ and gives the version that appears first in the manuscript the secondary identifier of version ‘B2’), while the evidence on the page indicates that in the the twelfth-century it was the first version that garnered the most attention.


The Treaty

The real-world truce between King Ælfred (r. after 15 April 871–26 October 899) and King Guthrum (d. 890) for the Danes has traditionally been dated in the scholarship to between 886 CE and 890 CE, although David Dumville has suggested a slightly earlier date of 876 CE to 878 CE and certainly not after 886 CE (1992: 2, 13-14); overall, a more general date of 880–890 is now favoured, although Wormald notes that an overall reassessment in light of Dumville’s arguments is required (Kerhsaw, 2000: 43; Wormald, 1999: 285-86). Exactly when the written version of the Peace was first produced is uncertain. Presumably the written version was produced contemporary to or soon after the real-world peace between the two kings was negotiated, especially in light of arguments that the written form of many early medieval legal texts were initially produced as notes made by interested, literate parties present at the gatherings in which laws were orally promulgated and confirmed (Keynes, 1990; Wormald, 1977).

The two versions of the Peace are subtly different, with the second version being longer, including a number of sub-clauses and details that are omitted from the first version, and with the concluding sentences have been written in an easier to understand format. The voice changes between the two versions as well, with the second version being in the first person and giving the impression, at the very least, of being legislation in the royal voice. The first version, being written in the third person, increases the distance between the writer (and, subsequently, the reader also) and gives the impression of an onlooker to the formation of the treaty rather than a major participant. Wormald suggests that this change in voice is the result of a later redaction adapting the laws and re-framing them to a more general perspective, and in the process perhaps accidentally omitting some of the details regarding the values of goods, compensation, and warrantors in business relating to men, horses or oxen (1999: 285). Both treaties reflect the English perspective on the truce, or rather the perspective from Wessex, as despite claiming to be a truce for England, the northern border between the Danes and the Mercians is left undefined, essentially leaving Mercia to fend for itself (Dumville, 1992: 22).

The first clauses, edited by Liebermann as the Einleitung (that is, prologue), here edited as clause 1 in both versions, are broadly identical in both versions, each recounting the participants in the truce - the two kings as the leaders of the respective armies, as well as referring to the councillors and people of the English nation. While there is some variation in the wording, version 2 uses ‘cweden’ to describe the speaking of oaths, while version 1 originally uses ‘sworen’ which is then glossed in the interlinear space as ‘cweþ’, most probably by the hand who supplied the rubrics throughout the manuscript. Another difference between the two is that version 1 presents a shorter version of whom the oaths were made for and by, the people and their ‘of spryng’, kin, while version 2 gives more detail naming their descendants both born and unborn. The second version also connects the keeping of the truce to royal and divine favour, stating that the terms are to be kept by all who want the mercy of king and God.

Following the prologue, the next section of the truce (clause 2) outlines the boundary between Wessex and the Danes, using broadly similar terms. Some variation can be found in the prepositions used to describe the course of the boundary, such as version 1 beginning ‘andlang’, along, the Thames, while version 2 goes ‘upp on’, upon the river. As noted previously the course of the boundary as described outlines a division between the Danes and Wessex bit, despite claiming to be a peace treaty for the English, stops when it reaches Watling Street and makes no reference to the interests of the Mercians. The interpretation of the boundary has also been problematised in the scholarship, with R. H. C. Davis having argued for a re-interpretation of the boundary, claiming that parts of Mercia were in fact under the control of Wessex in the late 880s and early 890s, (1982: 805-06). This reassessment of the boundary, however, does not appear to have been widely accepted in the scholarship, and Dumville writes a detailed counter-argument to the claims (1992: 2-13), resulting in the more literal interpretation of the boundary clauses as they appear in the texts being favoured. The frontier defined by the treaty was at best short-lived and the clear boundary line dividing the Danelaw from the English kingdoms as described in the Peace was probably less rigid in practice (Dumville, 1992; Kershaw, 2000: 46).

The remaining part of the Peace - subdivided into clauses based on legal content by Liebermann and in some of the Latin versions of the Quadripartitus as discussed above - is copied as a discrete unit in both English versions in the manuscript, hence being edited here as a single piece under the numeration of clause 2a. This part contains all of the legal details pertaining to the truce, and legislate for a variety of situations including killings, compurgation for accusations, weregilds and the securing of trade between the two peoples. As noted before there are some differences between the two versions of the Peace, with the fuller version two including the sub-clauses mentioned above that expand on the subjects of the values of goods, compensation, and warrantors in business relating to men, horses or oxen.

The laws set the weregild of both Dane and English at the same value, eight half marks of refined gold, and with churls on taxable land and ‘their’, i.e. the Danish, freemen set at the same, lower value of 200 shillings. Witnesses in support of oaths to clear oneself of guilt are likewise set the in two tiers, with a king’s thegn, that is a noble, accused of murder seeking to clear himself requiring the support of twelve other thegns, while someone of lower standing requires eleven of his own rank and one king’s thegn. The Peace prohibits men from joining war-bands to attack the others without permission, and lays out details far exchanging hostages and securing pledges of peace when trade between the two groups was required. The longer, version two adds to the details on trade a concluding remark that it should be declared and known, in a delightful euphemism for honesty, that the one going to trade has ‘a clean back’.



Notes on the Editions

Editions and translations of both of the Old English copies of the Peace of Alfred and Guthrum are presented here, numbered as version 1 (Liebermann’s B2) and version 2 (Liebermann’s B) in order of their appearance in their manuscript contexts. In the editions the orthography, capitalisation and clause division follow the respective manuscript witnesses, but abbreviations have been expanded and word division and other punctuation has been normalised to facilitate reading. Images of the respective copies of the Peace in their manuscript contexts can, of course, be found on the website.

Clause division in both versions is minimal, with both copies having been written as a single piece of continuous prose in the text-block of the manuscript. However, some internal divisions can be noted, in that the Corpus scribe copied both versions in a late vernacular minuscule, and made only rare use of majuscule letters. Both versions begin with a red, pen-work initial ‘Ð’, indented into text-block at the beginning of the Peace. The second version follows this miniatured initial with a row of majuscules, then change to minuscules following the majuscule ‘Æ’ of Ælfred. While emphasising the beginning of the treaty this does not denote any sub-clauses. Both versions of the Peace, however, begin the description of the bounds with a majuscule, and the second version uses a further majuscule following the boundary clauses to introduce the beginning of the legal terms themselves. For the ease of the reader, the two versions of the laws have been divided at these points and numbers as 1) the prologue, 2) the boundary clauses and 2a) the legal terms. For convenience and cross-reference, a concordance with Liebermann’s edition (and some of the later versions in the Quadripartitus) is given in the translation, by noting the start of each new sub-clause and its numbering.