Juhász, Zoltán (2024) Revealing Footprints of Ancient Sources in Recent Eurasian and American Folk Music Cultures Using PCA of the Culture-Dependent Moment Vectors of Shared Melody Types. MUSIC & SCIENCE, 7. pp. 1-26. ISSN 2059-2043
|
Text
juhasz-2024-revealing.pdf Available under License Creative Commons Attribution Non-commercial. Download (4MB) | Preview |
Abstract
One of the possible objective and universal descriptions of most folk songs can be based on structural musical characteristics such as contour, tone set, tonality, rhythm, meter, and form. Experimental studies in the recent decade supported the universal importance of contour and tonality as the two most important characteristics determining human music cognition and memory. It follows from this statement that a mathematically adequate description of folk songs should be based on both contour and tonality information. We describe a method searching for characteristic groups of universal melody types (MTs) propagating jointly and regularly in several subsets of 59 folk music cultures in Eurasia and America, represented by a database of 59,000 folk songs. The MTs are represented by pairs of contour and degree distribution vectors. We describe the propagation of the MTs by 59-dimensional vectors containing their ?moments? in the 59 cultures studied. We show that principal component analysis (PCA) of these moment vectors reveals assumable ancestor cultures, and we show a method modeling the 59 musical cultures as linear combinations of the musical contents of seven assumable ancestral cultures. The results provide a method and a hypothesis for tracking the footprints of assumable ancient musical ?primary languages? in folk music traditions in Eurasia and America. The assumable musical ancestral cultures presented here show good correspondence with the distribution of certain human genetic characteristics and archaeogenetic results.
Item Type: | Article |
---|---|
Uncontrolled Keywords: | Artificial intelligence, ethnomusicology, principal component analysis, unsupervised learning |
Subjects: | M Music and Books on Music / zene, szövegkönyvek, kották > ML Literature of music / zeneirodalom, zeneművek > ML3918.F65 Folk music / népzene Q Science / természettudomány > QA Mathematics / matematika > QA75 Electronic computers. Computer science / számítástechnika, számítógéptudomány |
SWORD Depositor: | MTMT SWORD |
Depositing User: | MTMT SWORD |
Date Deposited: | 20 Mar 2024 13:41 |
Last Modified: | 20 Mar 2024 13:41 |
URI: | https://real.mtak.hu/id/eprint/190618 |
Actions (login required)
Edit Item |