Boel Lindberg

Digitisation and On-Line Publishing of Medieval Ballads

This project aims to make available for research a unique collection of manuscripts discovered in the municipal library of Växjö in 2005. The manuscripts contain recordings of medieval ballads collected by George Stephens and Gunnar Olof Hyltén-Cavallius in Sweden during the years 1835-1855. We will digitise the texts and melodies of the ballads and publish them on the web site. Our target group is mainly scholars in the Nordic countries in disciplines such as comparative literature, linguistics and musicology who are researching medieval ballads. However, the sources are also of interest to historians and ethnologists. Another target group is musicians and music bands that have medieval ballads on their repertoire. An aim for the future is that the project develops a platform on the web from which it is possible to reach many different sources that are relevant within the research field of Nordic medieval ballads.

Final report

Boel Lindberg, Linneaeus University

The aim of the project

We have in princip reached the goal we set up in the project: to make an unique collection of handwritten recordings of medieval popular ballads (George Stephens Manuscript Collection - GSMS) that was found in 2005 in Växjö, Sweden, available in a researchable form through digitising and publishing it on the web. Manuscripts from this collection that represent new knowledge compared to the scholarly editions of medieval ballads that existed in 2007 are now published in a database (Visdatabas) which can be reached at the address http://Folkvisa.se. At present the database contains 40 ballads and 40 others will be added early in 2012. Gradually other popular ballads and folk songs from the collection will be added to the database.

Results

The publishing of 80 recordings of medieval popular ballads from the George Stephens Manuscript Collection may seem modest compared to the around 380 recordings that were mentioned in the application to Riksbankens Jubileumsfond in January 2007. The estimation of how many recordings that ought to be published was at that time based on a preliminary examination of the collection carried out in 2006. Unfortunately, this examination was not done enough methodically to meet the requirements for the kind of scholarly editing of manuscripts that our project aimed at. The manuscripts therefore had to be scrutinized carefully once more and compared to other manuscripts that were related to the GSMS. The new examination of the records resulted in a new registry over the manuscripts in the collection that contain texts and melodies related to the ballad genre. 80 of those manuscripts were considered interesting enough to be published. The rest of the manuscripts (around 320 mss) were copies of recordings of medieval ballads that already are published in the scholarly national edition Sveriges Medeltida Ballader (Stockholm, 1983-2001).

The technical solutions that we developed for the digitising of the texts and melodies of the ballads are those we described in our application. The ballad texts are XML-encoded which includes a lemmatising and morpho-syntactic analysis of every word. The texts are prepared to help researchers within linguistics, literature, history, musicology and ethnology to use them in advanced studies. The project has in this undertaking adopted methods developed within the Nordic project Menota (Medieval Nordic Text Archive - cf. http://www.menota.org).

The digitising of the ballad melodies includes transcription of the handwritten musical notation in the notation software Sibelius. The document thus created is further processed into a jpeg-format for presentation on the website and to a music-XML format. The latter will make it possible to search the melodies for similarities. We have made special efforts to develop a search function for the melodies but have so far not found a solution that works in an open access-milieu.

The methods used for digitising the ballad texts and melodies have been documented in manuals that are now used in the continuation of the project that is running from January 1, 2011. Routines have also been developed for the collection of various metadata that is presented for each record in the database. Those metadata comprise provenance, the names of the informant and the recorder; time when the recording was done as well as precise facts on sources to the recording. They are presented in a format that link up with how metadata are presented in Sveriges Medeltida Ballader. With every record in the database come facsimiles of the sources that have been transcribed. There is also an ambition to add a scholarly annotation with bibliographic references to every ballad that is presented in the database and by now approximately half of the published ballads have such annotations.

Unexpected technical and methodological problems

The project was initially carried out in close cooperation with the university library at Växjö University. The Chief Librarian Mats Herder supported long into 2009 a solution on how to publish the ballads that involved a presentation of them within the program for digital preservation that the Swedish Royal Library (Kungliga biblioteket, KB) was developing. This solution included access to the ballads in the Swedish National Library System (Libris). Unfortunately the good cooperation with the University Library in Växjö and with KB in Stockholm took an end when, from Fall 2009, all Herder's time was needed in the process to fuse Växjö University with the university college in Kalmar. When Herder in April 2010 left his position at the new university (that is the Linnæus University) it became difficult to upheld good cooperation with KB. What has worked, however, is the fine contributions by librarian Erik Åberg in providing the project with routines for XML-encoding, setting up style-sheets for how the database and the records are presented on the web, and also more generally by apt IT-support.

The fusion that resulted in the new Linnæus University brought with it several reorganizations that had consequences for how the project could cooperate with the IT-department at the university. It has been difficult to achieve a solution as to how the database that we have created could be made public via the official webpage for the Linnæeus University. This meant that we long into 2011 only could use a provisional web address. From December this year the webpage Folkvisa.se is functioning and it is here that the database "Visdatabas" now is found.

Another unforeseen problem appeared in the planned cooperation with Svenskt Visarkiv (SVA, The Swedish Folk Song Archive) in Stockholm. Right from the beginning of the project we saw it as natural that the digitising and web publishing of the ballads in GSMS should be connected to the scholarly work Sveriges Medeltida Ballader (SMB, 1983-2001) that SVA had been responsible for. In our project we have had frequent contacts both with the editors of SMB (all of them have by now retired) and new members of the staff at SVA. Our discussions first aimed at the publishing of the GSMS ballads as a digital supplementary volume to SMB. This volume should also, with contribution from SVA, contain a long planned register for the content of the five volumes that SMB consist of as well as scholarly annotations to the ballads in them.

Unfortunately our discussions led nowhere and in February 2010 we finally let go the idea that our publishing should be done in cooperation with SVA. A contributory cause to the collapse of the cooperation was the great uncertainty that prevailed from around 2008 about the future of SVA, The archive as well as other institutions within the field of Music were at this time the subject of a governmental investigation and suggestions to shut down SVA were long in the air. This strongly affected our efforts to cooperate with SVA. The archive has now been incorporated in the new governmental agency - Statens musikverk - that has been set up from May 1, 2011.

Our solution as to how to publish the ballads contained in GSMS has from February 2010 been aimed at creating a more general database (Visdatabas) for popular ballads and folk songs in which also medieval ballads can be presented. From January 1, 2011 the continued work of updating and maintaining this database is done within a recently formed collaboration between the Linnæus University and Smålands Musikarkiv in Växjö. The collaboration is regulated in an agreement running for two years.

The results of the project as well as research that have been generated within the project will be presented in an anthology that will be published in 2012 with me and docent Per Stille (one of the researchers tied to the project from 2011) as editors. The content of the anthology is:
Magnus Gustafsson, "Hur 'gamle wijsor' blev ballader. Begreppsförskjutningar I nordisk visforskning."
Boel Lindberg, "Medietransformeringar av genren medeltida ballader från 1500-talet till nutid."
Karin Eriksson, "Visuppteckningar från Västanland. Om en grupp tidigare ej kända visor som återfunnits I George Stephens manuskriptsamling."
Gunilla Byrman, "Rapp och ballad - en jämförande studie."
Boel Lindberg, "Kvinna vet din plats." Balladen Tiggargubbens brud som patriakalt lärostycke.
Boel Lindberg och Eleonor Andersson, "Reviderad registrant over balladuppteckningar I George Stephens manuskriptsamling."
Lennart Carlsson, "Ballader på webben. Om projektet Digitalisering och nätpublicering av medeltida ballader."

Deadline for submission of the texts to the anthology is January 31, 2012. The texts have been discussed at several seminars arranged during 2011. Contact has been established with Gidlunds förlag, which will publish the anthology if enough grants can be secured. Otherwise we will be content with a digital publishing in Open access-format (DIVA).

Grant administrator
Linneaeus University, Växjö
Reference number
In2007-1039:1-IK
Amount
SEK 1,420,000
Funding
RJ Infrastructure for research
Subject
Unspecified
Year
2007