Inventio

Inventio in rhetoric is the first step for preparing an oration or a written statement. It consists of drawing up an inventory of essential information to invent arguments. In previous research (starting with my doctoral dissertation in 2018), I gathered information about the employment of signal tones. I developed hypotheses about the schemas that define their interrelations with other musical, textual and rhetorical schemas defining ‘European medieval liturgical chant’. So which information is essential I have found out in the meantime.

Schema theory is a neuropsychological concept that has proven useful describing text-music relations represented by signal tones. For an excellent introduction to what schema theory might imply for the study of this field, see: Ghosh, Vanessa E., and Asaf Gilboa. “What Is a Memory Schema? A Historical Perspective on Current Neuroscience Literature.” Neuropsychologia 53 (2014): 104–14.

To substantiate my hypotheses about signal tones, I organise information from the internet and from sources in libraries. Basic material often can be found in CANTUS . Fortunately, CANTUS provides indexes for both the Dendermonde manuscript and the Riesencodex. A selection of the fields in the CANTUS indexes for the two manuscripts I imported into Filemaker, database software I have used since 2018 for the inventio of data related to my research. I always set up one layout in which I organize all incoming information, that subsequently is analyzed in other layouts. The ‘General Input’ layout under construction looks like this:

Top left, the metadata of the chant per manuscript provided by CANTUS. Underneath top-down, links to the image of the chant analysed in the digitised sources, the Latin text plus its translation, followed by the centrepiece of the database, a table with words containing signal tones, the signs employed and their rhetorical categorisation. Finally some tables with quantitative summaries about the signal tones registered per chant and in the collection selected (this can be filtered for any subcategories in the database, for instance per genre, per source, etc.).

The right hand part of this layout shows a number of buttons that open other websites I frequently use for gathering the basic stuff for this database and, if relevant, the page or subject links found for quick reference. Below these fields, audio files. I should add information from the discographies available and the chant-specific links but, as said, this is a layout under construction. The Latin text is copied into online text processors, a syllabifier and a hyphenator. The former is used to calculate relative signal tone densities per chant, the latter is the basis for transcription in neoscript by Gregorio software.

The relative signal tone density is calculated by dividing the number of signal tones by the number of syllables per chant; it enables quantitative comparisons of signal tone employments between chants, composers, genres and whatever filter you may think of. Whether it makes sense from a methodological point of view remains to be seen; I do not bother too much because the software does the tedious job for me.

Gregorio software I use to transcribe the chants as notated in the manuscript into material that can be used by choirs. For instance, the following composition by Radbod of Utrecht (899-917), as notated in NL-Uu 406 results in the subsequent notation:

For encoding information about signal tones found in the manuscript, I keep two code tables at hand.

In other layouts in the same Filemaker database for this project, I assemble my notes from secondary literature. Separate layouts will combine the manuscript image with my findings per chant and/or filtered data from the respective manuscripts, comparisons etc.

Which signs in Hildegard’s compositions are signal tones I will sow in the next blog.