Or sus vous dormes trop
A new transmission of the famous chanson “Or sus vous dormes trop” on Vienna Ars Nova-Fragment A-Wn Cod 3917 (photographed as part of the “Musical Sources”-catalogue) was introduced and preliminarily transcribed in an earlier post to this blog site. The fragment is glued to the back cover of the host codex and the notation can only just be made out shining through from the other side of the folio. Since this new source still harbours information which cannot be gleaned even from the provided high-resolution image of the bleed-through notation, the “Musical Life”-project had requested for the fragment to be removed from the binding of the codex. However, because there was a certain danger for the fragment to be damaged in the process of removal or that the procedure might have to be aborted if the glue should prove to be too strong, Friedrich Simader from the Österreichische Nationalbibliothek had decided to try another, gentler and noninvasive method. The resulting UV-image gives a much clearer impression of the notation and the underlying text. Even if not every detail can be discerned with absolute certainty the image is a much safer alternative, which also allows for a completion and emendation of the preliminary transcription. We would like to thank Friedrich Simader for his initiative and this revealing new picture of the fragment.
The stave lines for “Or sus vous dormes trop” remain invisible (the visible but empty staves in the picture are on the other side of the notation), but most of the minim stems can now be seen. Also, there seems to be some sort of heading to the piece, hidden under the writings on the external side of the page. Future studies with a removed fragment might be able to solve this remaining enigma.
First, I would like to give a transcription of the underlying text:
It may be puzzling to find that the text for the A-part ends about a verse too early and breaks off in the middle of a word. The two B-parts have the complete text. Further analysis will show why the text of the A-part is incomplete.
The revised transcription of the fragment could in most cases confirm the “educated guesses” of the preliminary transcription for places which—with the traditional photography—had remained vague or indecipherable. In a few cases emendations could be made, especially in regard to rhythmic values, and the custos at the end of every musical line could be identified. (All these emendations and additions are marked in dark red in the following transcription.) The text underlay, surprisingly, turned out to allow interesting observations:
The text for the A-part does not simply break off one line short of its completion—it breaks off at the point where space runs out and where the music of the A-part ends. Thus it seems that the music at some point “overtook” its text, or—more precisely—where the text could not keep up with the music. It is clear that the musical notation for this source was written before the text underlay was made, and apparently this procedure was not a rare phenomenon (see for instance Lawrence Earp: “Texting in 15th-Century French Chansons: A Look Ahead from the 14th Century”, in: Early Music, Vol. 19, No. 2 (May, 1991), pp. 195-210). Especially in those syllabic passages where birdsong is imitated and which contain a multitude of musical and textual repetitions, the musical notation is much more efficient with its use of the available space than the underlying text—even when, as is the case here, abbreviations are used to save space. The layout of the text suggests that the writer was well aware of being “behind” the music in many places. Examples for this are: a) the ample use of abbreviations such as “est“, and “que”, b) overshooting of the musical line, such as in staves 3 and 5, and c) the space-efficient notation of the repeated “titon”, where the second syllable is “typographically” raised.
As the transcription shows, the writer was more successful with the underlay of the B-part, where more of these devices were employed. Maybe the scribe was alert after the failed texting of the A-part. A closer look, however, reveals that for the larger part of the piece the text underlay is actually fairly accurate. Only when the repeated text-patterns appear in the A-part does the underlay get behind the notation. It is here, in the midst of the confusing ostinato-patterns, where the error occurs which results in the fragmented text. In the second line, the music runs ahead of the text, quickly and uncatchable:
Since—even by falling behind the music—bits of text (albeit from an earlier sequence) end up correctly under their assigned musical figures, the scribe might have been led into a false sense of security. Apparently the scribe was also aware of a “key landmark” for the continuation of the proper text to set in after a section full of repeated phrases (“que dist dieu” and “ilh est jour”): a breve b-natural, clearly sticking out of the ostinato patterns. However, the melody features two such breves b-natural, of which the scribe chose the second one… the wrong one:
In addition, the following edition shows that a total value of two perfect semibreves is missing in the notation of the cantus line—a fact, which was only evident for one of the two lacunas before the UV-image was taken (bar 29), while the notation was too faint to verify it also for the second one (bar 72). The latter is an understandable mistake, since the repeating patterns in this part always come in groups of four, while the last one has five repetitions, the final one of which was missed out in text and music.
Furthermore there is one place where a minim is written instead of a semibreve and another one where minim stems seem to be missing. They could, however, also be too faint to shine through even in the UV-image. Apart from these mistakes the surviving notation appears to be correct and at the same time presents interesting deviations in small details from concordant transmissions of the chanson. I undertook a traditional texting for the edition, which probably was intended by the scribe. The second text for the A-part, necessary for a complete virelai-form, could have stood on the now lost facing page of an hypothetical original opening. This page would have also contained the tenor voice for the B-part and might have featured a contratenor voice as well.
Simader provided us with more information about the host codex (A-Wn Cod 3917): Apparently, the codex was not listed in the Vienna catalogue from 1433 and can only be traced back via its title label to the Domkapitelbibliothek of Salzburg around 1500. The binding seems to be the original one from the end of the 14th century when the codex was written. Entries on the last folio, however, already point to the Salzburg diocese. As was shown in the first blog post to this fragment, scribbles on the external side of the fragment also point to the Austrian region in the 15th century. For further information see the online catalogue.
Taking into account that Martin Staehelin had found another fragment with a textless version of “Or sus vous dormes trop” that was used as binding material for a codex from Constance (c1500), it appears increasingly likely that this chanson (and other famous ones like it?) must also have been popular in German speaking lands in the 15th century, thus strengthening the case of A-Wn Cod 3917. The Council of Constance readily comes to mind as a possible occasion for exchanges of such “international” Ars Nova compositions.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
 Staehelin, Martin: „Das Fragment einer französischen Chanson um 1400 in Stuttgart“, in: Staehelin, Martin (ed.): Kleinüberlieferung mehrstimmiger Musik vor 1550 in deutschem Sprachgebiet, Berlin (Walter de Gruyter) 2011 (Neue Quellen des Spätmittelalters aus Deutschland und der Schweiz, vol. IX), pp. 37-40.