Abstract
An attractive goal in the study of Gregorian chant melodies is reconstructing unobserved melodies as they may have been transmitted along their history, especially as early chant notation does not capture pitch exactly. We propose doing this computationally using Ancestral State Reconstruction over phylogenetic trees. Bayesian phylogenetic trees have shown promise as a tool to study the evolution of chant melodies, by inferring a plausible topology of chant transmission. However, the inferred trees cannot be used as Ancestral State Reconstruction inputs directly, because they are undirected, and their branch lengths conflate time and evolutionary rate. We therefore first apply Divergence Time Estimation to separate them and represent the tree in a directed form on the time dimension. Using Ancestral State Reconstruction, we then obtain reconstructions of melodies for each of the ancestral nodes, in addition to their distribution in time obtained from Divergence Time Estimation, and thus recover a phylogeny of chant melody with a music-historical interpretation. We applied this method to the Christmas Vespers dataset, and compare the results against musicological knowledge and melodies reconstructed at Solesmes using methods of contemporary philology, which shows potential for reconstructing cultural transmission through time.
Cite
CITATION STYLE
Ballen, G. A., Hedvika Mühlová, K., & Hajič, J. (2026). What did the dove sing to Pope Gregory? Ancestral melody reconstruction in Gregorian chant using Bayesian phylogenetics. PLOS ONE, 21(5 May). https://doi.org/10.1371/journal.pone.0350014
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.