2.5.3. Present and future possibilities of the electronic edition

The central data type of the database before the publication of SkP VII (the first volume to appear) was the stanza. Most of the information in the database is associated with stanzas: manuscripts, notes, editions, as well as the text and translation of the stanzas. The principle is the same as that of the print edition, where the bulk of the information is associated with individual stanzas: titles and introductory material form a smaller proportion of the printed volumes. The stanza, prose order and translation were originally encoded as XML fragments, where the structure of the text was marked up electronically as lines, words and kennings.

However, possibilities began to emerge where it seemed desirable to have the words of the corpus as part of the database structure. After SkP VII was exported and sent to the publisher in mid-2007, Wills took the opportunity created by the temporary lull in data entry activity to alter the underlying format of the edition. This involved removing the XML structure of the verse text and translations and transferring it to a relational database model, where individual words are linked to lines and stanzas and ordered for prose interpretation and translation. The new structure means that more information can be linked to each word in the corpus.

The first major extension to the edition has been to link words in the corpus to dictionary headwords. This provides a meaningful way of searching for words independently of their grammatical form and normalisation period. This capability has been extended to grouping each word by sense, so that lexicon entries can be developed. Compounds are also indexed by their constituent parts, and major variant readings are also included. The kennings of the corpus were from the beginning marked up to represent the basic structures of simple and complex kennings. When entered into the database, the kennings are automatically indexed by referent and the kenning index has been searchable online since the stanzas were first entered. The project aims to produce a new reference work for the skaldic kennings system (replacing Meissner), which will involve indexing the constituent parts of kennings and categorising both their parts and referents.

The structure of the database means that more possibilities for indexing and analysis are available: a comprehensive labelling of the metrical forms of the corpus, for example, would allow synchronic and generic analysis of this feature. Any type of information in the database can be related to any other: the date, skald, manuscript, manuscript date, metre, metrical line type, lexical distribution, kenning types, kenning frequency, even word order complexity. Overall, the database provides a map of the corpus: by providing a way of incorporating many different features of the ‘terrain’, we can build a much more complete picture of the corpus as a whole and the relationship between any feature or part of it. This ‘map’ will ultimately allow for the refinement of ideas that were once speculative, such as the dating of certain stanzas. Ultimately, it will lead to more complex and subtle analysis of the skaldic corpus.

References

1. Bibliography
2. Meissner = Meissner, Rudolf. 1921. Die Kenningar der Skalden: Ein Beitrag zur skaldischen Poetik. Rheinische Beiträge und Hülfsbücher zur germanischen Philologie und Volkskunde 1. Bonn and Leipzig: Schroeder. Rpt. 1984. Hildesheim etc.: Olms.
3. SkP VII = Poetry on Christian Subjects. Ed. Margaret Clunies Ross. 2007.