Conference Agenda

Historically characterised as bishops’ books, there has often been debate on how pontifical manuscripts should be defined and studied, as variations in content greatly affect our understanding of who interacted with these manuscripts. In the continued exploration of pontifical content, musical notation has been a largely neglected factor in examining the relationship between content, transmission, and use. In this paper, I discuss the use of digital tools for establishing musical relationships between English pontifical manuscripts produced up to the thirteenth century. Specifically, I address the methodological challenges of establishing concordance when assessing manuscript relationships.

In the early stages of the project, I indexed over 1200 individual instances of notation across roughly twenty-five pontifical manuscripts and fragments. This corpus contains a vast but inconsistent repertoire of notated chants, situated within a network of manuscripts demonstrating a wide variety of liturgical content. Because of this, manuscripts demonstrating textual or liturgical compatibility might have a more ambiguous relationship musically, in which a chant is present in multiple manuscripts, but not all instances are notated. I approach these manuscripts through a tripartite model for establishing concordance, assessing manuscript relationships through overlaps in text, melody, and liturgical context. This presents a multi-layered approach in which manuscripts might demonstrate different levels of concordance based on different configurations (i.e. text-melody, text-liturgy, text-melody-liturgy, etc.). I determine the quantifiable ‘weight’ of manuscript relationships based on these concordances and identify significant textual and melodic variants within this repertoire.

Through an overview of the chant data collected on these manuscripts, I present findings such as the most significant manuscript pairings, most and least common melodies, unique notational occurrences, and any potential geographical trends. I explore best practices for utilising existing online chant databases to assess textual and melodic concordance, as well as methods for assessing melodic similarity across notational systems with differing levels of pitch specificity. Considering these figures, I discuss the methodological challenges in quantifying similarity in early medieval chant scholarship more broadly, and explore how network analysis tools such as Gephi can be used to visualise these musical relationships, thus improving understanding of melodic transmission between the English pontificals.

The Parliament of Birds is an anonymous, 120-line macaronic poem in French and Middle English, which survives uniquely in the early fifteenth-century English manuscript Cambridge, University Library MS Gg.4.27. The poem portrays an avian debate, in which sixteen species of bird weigh the delights and disappointments of love. While the majority of this debate is in Middle English, the final lines in thirteen of the poem’s fifteen stanzas are in French. These French lines are linked to a musical context by the turtledove, who opens the dispute by ‘syngen…In frensch ho so the rondele knewe’ (‘singing…In French, as she knew the roundel’, lines 14–15).

Scholarship (Conlee 1991; Butterfield 2009; Doyle 2021) has found concordances for the song thrush’s line ‘Qui bien ayme tard oublye’ (line 32) in the well-circulated refrain Qui bien aime, Guillaume de Machaut’s Lay de plour (c.1349–1350), and a ballade by Eustache Deschamps. The twelve remaining French lines have not been discussed. Do these lines have similar concordances in French-language poetic works with musical settings? What might such concordances reveal about the performance and dissemination of music in late-medieval England?

This paper proposes concordances for three more of the Parliament’s French-texted lines, and suggests that the nine remaining lines have more general resonances with a broader corpus of courtly song. The network of musical texts collected in the Parliament enables the remembrance and facilitates the performance of these works, transmitting this music both to a fifteenth-century audience, and to readers in the present day.

Comparative study of medieval Byzantine and Latin chant melodies suggests an alternate approach to the study of "tune families" in folk music and other traditions of orally-transmitted melodies. The "tune family" concept in folk music studies has been useful as a way of classifying multiple related melodies, but it has not proved useful at establishing "family tree" relationships, such as "ancestor/descendent" or "sibling/cousin." Medieval chant studies, in which melodic history can sometimes be traced over many centuries, point in a different direction. Instead of focusing on the note-to-note melodic "foreground," we should consider that the datum of transmission was not the melody, but a more abstract "background" or deep structure. Parallel examples from medieval chant and more recent folk and popular music will be used to illustrate four types of background structures that can be found in orally transmitted vocal song.

(1) Modal matrix, which is not the same thing as modal assignment, refers to the region of the overall gamut that a melody inhabits. Multiple melodies can share the same modal matrix, or overlapping matrices, even when they are officially assigned to different modes.

(2) Formal congruence occurs when different melodies, set to the same text, also share the same form, such as A A' B C A, even though the sequence of musical pitches is completely different. The sense that a musical section should be the same or different was passed on even when the "foreground" melody was not.

(3) Sometimes the unit of transmission seems to be a melodic contour, which is elaborated differently in each performance. Evidence from the Music Cognition field can support this, and it is relevant that the earliest medieval neumes functioned as a "contour" notation rather than a pitch notation.

(4) "Cross references," or deliberate quotations from other songs, have been ignored since the discrediting of "centonate" as a term for formulaic chant. But they actually exist.

The way forward in "tune family" studies, then, would be to focus more on "background" structures like these than on the surface, note-to-note "foreground" tune.