Abstract
Variation in music can be caused by different phenomena: conscious, creative manipulation of musical ideas; but also unconscious variation during music recall. It is the latter phenomenon that we wish to study: variation which occurs in oral transmission, in which a melody is taught without the help of written music from one singer to another, or by singing together in a group of singers.
The Dutch Song Database 1 is a large collection of more than 150’000 Dutch folk songs, and contains more than 7000 transcriptions from orally transmitted melodies, which lend themselves to the investigation of varia- tion through oral transmission. The songs in the collection are categorized into tune families. Songs belonging to the same tune family presumably stem from the same ancestor melody, and can be considered as variants of each other.
Strikingly, some melodic patterns within a tune family vary a lot, while other melodic patterns remain relatively unchanged. We call the latter phenomenon stability in oral transmission: a melody’s resistance to change. We propose a number of hypotheses for possible causes of stability in oral transmission: our hypotheses predict that those patterns will be easiest to remember which are highly expected within their harmonic, metric, or melodic context. These hypotheses are tested in an iterative process using musical pattern matching, which gives an indication of how often a specific melodic pattern recurs within a tune family.
Moreover, we present our research on cadences, conclusive musical formulae which appear at the end of phrases. According to folk song classification systems and folk song theories, these cadences are stable in oral transmission. Using a rule-mining algorithm, we investigate the musical features underlying musical cadences in folk songs.
The Dutch Song Database 1 is a large collection of more than 150’000 Dutch folk songs, and contains more than 7000 transcriptions from orally transmitted melodies, which lend themselves to the investigation of varia- tion through oral transmission. The songs in the collection are categorized into tune families. Songs belonging to the same tune family presumably stem from the same ancestor melody, and can be considered as variants of each other.
Strikingly, some melodic patterns within a tune family vary a lot, while other melodic patterns remain relatively unchanged. We call the latter phenomenon stability in oral transmission: a melody’s resistance to change. We propose a number of hypotheses for possible causes of stability in oral transmission: our hypotheses predict that those patterns will be easiest to remember which are highly expected within their harmonic, metric, or melodic context. These hypotheses are tested in an iterative process using musical pattern matching, which gives an indication of how often a specific melodic pattern recurs within a tune family.
Moreover, we present our research on cadences, conclusive musical formulae which appear at the end of phrases. According to folk song classification systems and folk song theories, these cadences are stable in oral transmission. Using a rule-mining algorithm, we investigate the musical features underlying musical cadences in folk songs.
Original language | English |
---|---|
Publication status | Published - 12 Jun 2014 |
Event | Digital Humanities Benelux Conference 2014 - Huygens Instituut & Koninklijke Bibliotheek, The Hague, Netherlands Duration: 12 Jun 2014 → 13 Jun 2014 http://www.dhbenelux.org/?ar=2014 |
Conference
Conference | Digital Humanities Benelux Conference 2014 |
---|---|
Abbreviated title | Digital Humanities Benelux Conference 2014 |
Country/Territory | Netherlands |
City | The Hague |
Period | 12/06/2014 → 13/06/2014 |
Internet address |