§ 1 Cædmon's Hymn is the name given to a poem recorded in Old English in some manuscripts of Bede's Historia Ecclesiastica, an ecclesiastical history of the English people written in the early eighth century at Wearmouth-Jarrow in Northumbria. It has long received the attention of Anglo-Saxonists for many reasons, but particularly because it is (perhaps) the oldest surviving record of poetry in Old English, it is one of the very few poetic texts from Anglo-Saxon England which survives in multiple copies, and its transmission is unusually complex even for an early medieval text. Despite this interest, however, O'Donnell's is the most comprehensive study to date.[1]

§ 2 As O'Donnell's title suggests, his work incorporates a critical edition, scholarly study, and textual archive of the Hymn, and all three components are presented in both print and digital formats; even the full text of the book is included in the accompanying CD-ROM. The work therefore crosses several scholarly boundaries and is relevant to those interested in Anglo-Saxon language and literature, textual transmission, manuscripts, and principles of editing but also digital humanities, hybrid publication, and digital editions.


§ 3 The book opens with two important sections before the study proper: the Preface and Conventions, Symbols and Encoding, where the aims and principles are outlined in some detail. Following these are three chapters of Literary and Historical Introduction. The first considers the Hymn and its relationship to Bede's historical text. The second presents a very full review of all the suggested sources or analogues to Cædmon's poetic inspiration; this is to be read with Note E, a list with sources and summary of some forty-six analogues to the Cædmon Story which can be found in the scholarship. The third introduction contains a detailed analysis of the Hymn in the context of early Germanic poetry and an assessment of its aesthetic and formal innovation.

§ 4 Part B, the Textual and Linguistic Introduction, begins with very useful short descriptions of the twenty-one manuscripts containing the Hymn. In Chapter Five, O'Donnell discusses the filiation and transmission of the Hymn itself and argues for an entirely new recensional development in which the so-called West Saxon eorðan version is most authoritative. The next chapter contains a detailed discussion of the dialectal and orthographic variation in the surviving witnesses. In Chapter Seven, the last of the Introductions, O'Donnell presents a detailed rationale for each of his eight critical editions of the Hymn.

§ 5 Following the seven chapters of introduction are five Notes, each of which addresses a particular scholarly debate about the Hymn. Eight editions of the Hymn then follow: the critical edition of the Hymn itself, proposed archetypes of the different recensions, and three critical editions of scribal performances. These are supported by diplomatic editions of all twenty-one witnesses in the Witness Archive. Back matter includes a glossary of all Old English words in the Hymn, a full bibliography, general and manuscript indices, and system requirements and installation instructions for the CD-ROM.

Content Review

§ 6 The introductions comprise nearly 170 pages in print and form an important study of Cædmon's Hymn. The discussions are very wide-ranging and generally seem well informed, balanced and well argued, with a lot of detail provided and careful and informed use of statistics. The Notes are again useful reviews of key debates, and the careful treatment of numerical and geometric patterns is especially welcome. The discussions are therefore a valuable mine of information even for those who do not accept the conclusions they contain. They also consider the text from many different directions — historical, cultural, manuscript, linguistic and poetic — in a way that most medievalists advocate but few achieve. Inevitably some arguments are more convincing than others, and this reviewer sometimes felt that the conclusion was probably correct but that the arguments were not (and probably could not be) conclusive, but in general the key points, such as the Hymn not being a back-translation from Bede's Latin, seem secure.

§ 7 This reviewer noticed few typographical and formatting errors, and spot-checks of the transcripts show them to be accurate. Although editorial principles are discussed in detail, the diplomatic transcripts could have benefited from more discussion since these also involve making judgements such as the representation of spacing, word-division and allographs.[2] Similarly the sections of manuscript P which are marked as physically damaged in some way are still very easily legible and show only the slightest damage in contrast with, for example, H or M which are not so marked. The transcripts also contain a few bold readings contrary to previous editors, such as scwlun for Wuest's scuilun and Dobbie's sciulun in the first line of manuscript Di.[3] These are very minor quibbles, though, and there will inevitably be disputable details and small inconsistencies when transcribing so many different manuscripts. Much more important is the inclusion of full diplomatic transcripts, facsimiles, and detailed notes for the transcripts, and for each critical edition an extensive and detailed apparatus with each variant classified as significant, substantive, and orthographic, along with the discussion of editorial principles. This openness and quantity of information allows the reader in principle to evaluate every step of the editorial process and to engage with the result in a way which is very unusual when editing from such a relatively large number of witnesses.

Digital Review

§ 8 The work is a model of hybrid print-digital publication, with the same SGML files used to produce both the digital output and the camera-ready copy supplied to the publisher (O'Donnell 2005, copyright page and §S.2).[4] The core content is provided as HTML for display (XHTML 1.0 Transitional); this includes the entire text of the book augmented at a minimum by use of colour and by extensive hyperlinks, and the editions and transcriptions are developed more fully. Each of the editions can be visualised with different types of critical apparatus depending on the type of information required, where the precise options depend on the editorial principles used in each case. Similarly the transcripts can be viewed as diplomatic, semi-diplomatic, diplomatic alongside an image of the manuscript, or a facsimile of the entire page containing the Hymn. Furthermore, the editions and transcriptions also include a useful list of links to related information. Separate stylesheets are provided for screen and print visualisation, and the few cases of invalid markup still displayed correctly in the browsers. Javascript is minimal and <noscript> alternatives are provided, and the CSS stylesheets largely pass W3C validation. Special characters are represented as entities with Unicode code-points, and the Junicode font is provided on the CD-ROM to ensure that all necessary characters are implemented.[5]

§ 9 As well as the HTML, an indexed display text is provided which again shows the HTML display but allows full-text searching using the Greenstone Digital Library program; unfortunately this is only available for Windows systems.[6]

§ 10 The full content is also provided in SGML along with stylesheets for the (proprietary) Multidoc/Panorama SGML browsers. The SGML is conformant to TEI P4 with some minor extensions which are fully documented (O'Donnell 2005, §ii.9–11), Unfortunately the markup uses features of SGML such as unclosed tags and case-insensitivity which would make conversion to XML very time-consuming if the encoded text was to be reused for further study and analysis.[7] Furthermore, the markup is very dense and relies on a relatively large text database of entities meaning that it cannot easily be read without processing. The marked-up text is therefore less useful than it could have been but is still valuable for preservation and its inclusion is very much welcomed.

§ 11 The digital images of the manuscripts are in PNG format and each comes in low and high resolutions (96 and 150 dpi). The quality of the images vary significantly (as explained by O'Donnell 2005, §7.11) but they are still invaluable for considering the Hymn in its manuscript context, and the inclusion of full pages as well as details is to be commended.


§ 12 O'Donnell is undoubtedly helped by the brevity of his text, as the eight lines of verse in the Hymn allows much closer analysis than is possible in a longer work. Nevertheless, if even some of the principles employed here were used in other editions, particularly the wide interdisciplinary approach on the one hand and the intelligent combination of print and digital output on the other, then we would have a much more secure basis on which to work. In his preface, O'Donnell expresses the hope that his work will prove a useful addition to Anglo-Saxon studies and harness the particular strengths of the two media [print and electronic] to produce a work that is more useful than either of its parts (§1.6). It is the opinion of the present reviewer that he has succeeded admirably in both goals.


[1]. Other important studies include O'Keefe 1990, Schwab 1972, and Dobbie 1937, but for a full bibliography see O'Donnell 2005 itself.

[2]. Spacing is always difficult at best since the range of medieval spaces cannot be reproduced adequately in either print or digital format (Saenger 1997). Dotted and undotted i are distinguished in the transcripts presumably to indicate ambiguous minims but this is not explained and no other allograph (a licensed and recognized variation in the representation, Davis 2007, 254–5) is so distinguished.

[3]. This is despite O'Donnell transcribing almost exactly the same sequence of strokes as two letters without comment (e.g. dumgeard and firum, both in line 6), and the form of his proposed w in scwlun would be very unusual in a manuscript of this date. Another example of debatable minim-resolution is in line 4 of SanM, wnndra for wundra.

[4]. This section draws loosely on the principles of evaluating digital work proposed by Rockwell 2005 and Bodard and Garcés 2009. For further references see Muri 2009.

[5]. Junicode is a Unicode font developed for medievalists by Peter Baker and is available at <>.

[6]. The Greenstone software is open-source and covered by the GPL (Greenstone Digital Library Software <>). The Greenstone website also lists versions for Linux and MacOS 10.5 but they were not provided as pre-packaged collections on the Cædmon's Hymn CD-ROM and there was no obvious way of running the standalone versions with data from the Hymn.

[7]. The @teiform attribute has sometimes been used to provide the case-sensitive form of the element but this is done inconsistently.

Works cited

Bodard, Gabriel, and Juan Garcés. 2009. Open source critical editions: a rationale. Text editing, print and the digital world. M. Deegan and K. Sunderland. 83-98. Aldershot: Ashgate.

Davis, Tom 2007. The practice of handwriting identification The Library (7th series) 8: 251-76

Dobbie, E. V. K. 1937. The manuscripts of Cædmon's Hymn and Bede's Death Song with a critical text of the Epistola Cuthberti de obitu Bedæ. Columbia Studies in English and Comparative Literature, 128. New York: Columbia University Press

Muri, Allison. 2009. Disciplinary standards in digital humanities. Humanist 22.575 (2). <>

O'Donnell, Daniel Paul. 2005. Cædmon's Hymn: a multimedia study, edition and archive. Woodbridge: D.S. Brewer.

O'Keeffe, Katherine O'Brien. 1990. Visible song: transitional literacy in Old English verse. Cambridge Studies in Anglo-Saxon England, 4. Cambridge: Cambridge University Press

Rockwell, Geoffrey. 2008. Short guide to evaluation of digital work. <>

Saenger, Paul Henry. 1997. Space between words: the origin of silent reading. Figurae: Reading Medieval Culture. Stanford, CA: Stanford University Press

Schwab, U. 1972. Cædmon. Testi e Studi: Pubblicazioni dell'Istituto di Lingue e Letterature Germanische. Messina: Peloritana Editrice