Peter Woetmann Christoffersen
Publishing 15th century music: Open Access and Digital Editing
Paper read at the symposium Digital Editions: Perspectives for Editors and Users at The Department of Arts and Cultural Studies, Section of Musicology, University of Copenhagen, January 19, 2008.
To me digital editing is something that has grown from a field of activity, which I have loved and worked with for nearly forty years. In the early years pencil, ruler and music paper were the tools to use before musical scores went to the professional typesetter. Years later I had to struggle to learn myself typesetting music on a computer, and later again it became a sort of professional career to edit and typeset music of every sort for publication (church, chamber and symphonic music, traditional and avant-garde). During recent years my attention has turned more and more towards publishing on the Internet. My interest in this channel of publishing, which is open to every user and in principle unlimited in scope, has been fuelled by the nature of the sources that I am studying and want to edit. The main reason is of course that digital publishing offers much better possibilities of adequate presentations of the material, but a growing dissatisfaction with the way in which we usually edit the repertories of the 15th and 16th centuries and especially with the complete works of single composers has also influenced my choice of publication strategies.
This means that I’ve arrived at digital editing through traditional music publishing and that my approach to the digital possibilities is mainly graphical in nature. The future potential of music editions generated from data extracted from the sources and stored in databases, and the development of interactive user interfaces seems to me highly interesting, and I’m sure that new techniques in time will change our thinking of digital editing. On the other hand I don’t think that there is any reason to wait for such developments to mature. In many cases standard applications are fully capable of doing the job of presenting musical editions on the Internet. In my projects I basically use only two different file formats: html for communication through browsers and PDF for the music editions. Both formats of course can and will in some cases be expanded with interactive functions and multilayer presentations, but in principle they are static, and their contents must from time to time be updated from the data behind the presentations. The important thing is to begin publishing music (as well as scholarly articles or books) on the Internet according to the standards of open access.
I shall not say much about the principles governing open access publishing and the open access declarations, which still more universities and institutions are signing. The easiest way of getting information on this subject is to consult the very informative web page managed by Søren Dorch at the Royal Library, Copenhagen (http://www.kb.dk/en/kub/open-access). To the musicologist it is of great importance that publishing music in scholarly editions on the Internet makes the effort far more visible – searchable in Google and other search engines – and useful for many more users than if published in expensive volumes, which end up standing shoulder by shoulder on the shelves of a few central libraries. Moreover it is probable that research published in this way will have a higher impact factor, a greater exposure, than if published in library editions. This has been the case in the natural sciences where articles published in open access channels get more citations (by about a factor 2) than articles in journals demanding high fees for admission to either the printed or the digital version.
Even if the political climate in Denmark at the moment is not very favourable to open access publishing – politicians apparently only appreciate heavy volumes published by multinational publishing houses – I’m sure that in the future open access publishing will become unavoidable also in the humanities. As is already happening in other countries, open access will be demanded of all publicly founded research and of projects with support from private foundations.
The discussion following The New Josquin Edition has touched a tender spot in musical editing. Here we can observe how the authority of the monumental edition, the exacting philological standards in research and the wish to present only the absolute truth have made the composer Josquin Desprez fade away in favour of an abstract idealized picture of a composer, which his contemporaries hardly would be able to recognize (cf. further Rob C. Wegman’s article ‘Who Was Josquin?’ in Richard Sherr (ed.): The Josquin Companion. Oxford 2000, pp. 21-50). This edition was meant to be an up-to-date tool for research in the music of an important composer. Instead we have got a monument celebrating a generation of Josquin scholars, their aspirations and their dreams of a genius and the perfect, autonomous musical work. The fame and the mythologizing of Josquin during his own time and the following generations are real elements of his musical influence, but such slippery data are not wanted as aspects of a perfect edition. Substantial parts of what 16th-century musicians heard, sang, admired and were influenced by have simply been edited away as false and not significant for an understanding of the “real Josquin”.
The old Josquin-edition by Smijers and his successors depended heavily on prints by Petrucci and later 16th century publishers, the sources which to a great extend were responsible for creating Josquin’s reputation, and therefore this edition in spite of its shortcomings is still valuable. Let us have these sources in new scholarly editions – a new cheap edition of the Harmonice musices Odhecaton has recently appeared (ed. by David Fallows et al., Watertown MA, Amherst Early Music Inc. 2005), and in 2003 Motetti De Passione, De Cruce, De Sacramento, De Beata Virgine Et Huiusmodi B from 1503 was edited by Warren Drake in the series Monuments of Renaissance Music (University of Chicago Press); more are probably forthcoming in big volumes. One has to recognize that Josquin was the first really medialized composer.
The solution to the deficiencies of the opera omnia editions may be to direct greater attention towards research in and publication of single sources. Here digital publishing has a lot to offer, first and foremost in the shape of facsimile editions and in the future also music editions, I hope. One of the great benefits of the technological evolution is the easy accessible digital facsimiles of musical sources on the Internet. The Royal Library in Copenhagen was among the first to offer this service. In my area of study the facsimile of The Copenhagen Chansonnier was widely admired and made several colleagues rather jealous. In the same genre I will mention the high quality facsimiles published by the Wolfenbüttel and Sankt Gallen libraries. I will also mention the very impressive work by a single man and his digital camera, the monk Jean-Pierre Voutaz who made nearly all of the manuscripts in the library of the monastery in the Alps, the famous Grand-Saint-Bernard, accessible on the net. And he did send me a CD with two of the manuscripts before their publication. It was a great help, as was all the other web-facsimiles. In the two publishing projects, which I shall describe below, I’ve made links to facsimiles wherever possible (permalinks are a great feature). The second project in fact builds on the user’s simultaneous access to the facsimile as well as to the modern edition.
It is important to analyze and edit sources in their entirety, to look at certain sources as independent works of art in which texts, music, illuminations, the writing and binding as well as the composition and planning of the repertory create an artistic whole. The Copenhagen Chansonnier is such a source. In this field Knud Jeppesen was a sort of pioneer when in 1927 he published the chansonnier complete exactly as he read it on the pages of the MS. In his time there were not strong traditions for this sort of editing. It was primarily found in connection with the big manuscripts of early – mostly anonymous – medieval music. However, at the turn of the 19th century there were certain movements in this direction including music of the 15th and 16th centuries, especially in France. For example Henry Expert’s monumental series Les Maîtres Musiciens de la Renaissance Française did include as volume 5 a complete edition of one of Attaingnant’s printed chansonniers (Paris 1897). Later the publishing of works by single composers became predominant. One could say that the aesthetic of the musical work won the market from the aesthetic of the source.
The wish to publish complete musical sources has returned. They appear as facsimiles and as transcriptions in printed editions, heavy and difficult to handle, and difficult to use if one wish to compare composers or repertories. A good example is the new edition by Rebecca Gerber of the famous manuscript known as Trent 88 from the 1460s (Sacred Music from the Cathedral of Trent. Trent, Museo Provinciale d’arte, Codex 1375 (olim 88). Edited and with an Introduction by Rebecca L. Gerber. (Monuments of Renaissance Music XII) Chicago 2007). It is like a slab of marble. I hope that someday the prestige of open access publishing can keep up with such an edition, that an editor could get the same sort of heavyweight support from foundations for a publication on the Internet. Most copies of this book will never leave the reading rooms of libraries, and therefore it will not really be used for teaching students at universities about the context for the famous masterpieces. It is a shame, because it is an important publication. Not because it contains a new, very good edition of Du Fay’s famous Missa Se la face ay pale or of Ockeghem’s just as famous Caput-mass, but because it includes a wealth of anonymous music, mass sections and complete cycles. Most of its music has only been known to experts with access to the manuscript or to the rare facsimile edition. Now we can study the famous compositions and the anonymous music side by side and in context. But it is not very practical!
Digital edition and open access can change that. Sources can be presented complete, and the user can navigate freely in the repertory, and with many sources made accessible the user can make his own selections of repertory according to genre, to subject matter, to gender etc. – to find his own path through the music in order to map the context. Digital editions may help to bring the study of musical context at level with the work-oriented research.
There is a tendency to regard the expansion of research using open access publishing as a technical problem: When the technological development sometime in the future is able to offer a perfect interactive user interface, the user will become able to participate in the process of scholarly editing as an equal partner. With all sources online the user should be able to produce own editions of needed music. This will hardly be the case. As everyone knows, scholarly editing is a highly specialized process only mastered by few. Today the knowledge and experience required for greater editorial ventures is only found in groups of scholars including several different areas of expertise. Open access research is rather an attitude towards research, which in principle is independent of technology, even if the idea of course is based on fast communication – it is difficult to keep a process open using snail mail. It is primarily about creating competent groups of scholars, about pre-publishing results, and incorporating reactions and discussions in final results – about using the technology and the net in an open process.
Now to a short description of two of my own projects concerning digital open access publishing of music:
The first concerns a complete edition of the polyphonic music in Amiens, Bibliothèque Centrale Louis Aragon, MS 162 D, a mixed collection containing music sections as well as two incomplete missals. The music MS can be dated to the years just after 1500, and it was probably executed for a confraternity associated with the great Benedictine Abbey in Corbie near Amiens in Northern France. Originally the contents of the music MS were carefully planned. A booklet of four fascicles contains three- and two-part simple music for funerals or commemoration, which probably were copied from several sources. The texts and some tunes are known from other French monastic sources from the second half of the 15th century. Before long this small manuscript was enlarged with a collection of two-part sequences. Another section was intended for monophonic music including a mass for St. Catherine. With some pieces left unfinished the intended order broke down and music was randomly added on empty spaces and pages. All hands in the MS were trained in copying liturgical books with plainchant – the two original copyists were probably professionals – and a later hand apparently only copied the visual appearance of some pieces of mensural music having no real understanding of the notation.
The repertory of polyphony for funerals and commemoration is especially interesting. It points to a neglected music of the second part of the 15th century, a period when composers became professionals and famous artists – with names as Du Fay, Ockeghem, Busnoys, Obrecht and Josquin as standard bearers for generations of musicians. The very simple polyphony survived beside this explosion of art music. Most often as improvised music, not recorded in notation, as cantus super librum, but in a few happy instances – as in this MS – also written down for singers who did not have sufficient musical education to improvise polyphony, not even the simplest. The MS lets us glimpse the stylistic breadth of simple music with roots far back in time, and it demonstrates the solemn sound of the prayer, to which the “great” sacred music as an important part of its means of expression so often associates. (see the longer version of this description)
It is important to study this repertory, and of course an edition is needed for the purpose. Examples of simple polyphony have been published, usually just as examples in shortened versions. For my project the point is to study the entire range of styles. As this MS does not use any sort of abbreviation of the texts and music – it was written for singers who in fact could not read music, could not improvise, and could not sing their parts from only the text of the following stanzas – everything is written out in full. An edition in score runs to many pages containing nearly identical music, and I’ve never even though about raising the money for a publication in a printed volume.
From the start my plan has been to publish one or two articles on the MS including a few music examples, but also at the same time to refer the reader to the Internet-edition (http://amiens.pwch.dk/). It gives a great freedom in the investigation of the highly interesting changes, which were made on the pages of the MS during use of the music, to be able to refer to a complete edition made without the restrictions caused by considerations of space. The edition also contains all the related settings in other sources from the same sort of milieu, primarily monasteries, which use the same texts or tunes. This also serves to place the MS in its context. As mentioned, the MS contains some additions in mensural notation, small spiritual pieces. It has been fun and important for the discussion of the dating of the additions to have room for publishing different versions of certain pieces.
As examples I will show you first the long three-part setting in simple polyphony of a trope for Libera me, “Juxta corpus spiritus stetit” (Amiens MS 162 D, ff. 18v-28), namely the first two openings of the song as they appear in the MS and the same two stanzas in the edition (see Examples 1-2 and the Edition no. 5). It goes on for ten stanzas in all. We find a related two-part setting of the same text and tune in a MS from the Augustinian monastery Grand-Saint-Bernard, where it is still kept today in Bibliothèque de l’Hospice as MS 6 (pp. 208-213; see the digital facsimile and the edition). This setting contains in all 8 stanzas supplemented with extra 7 responsory verses.
Example 1, Amiens, MS 162 D ff. 18v-19 “Juxta corpus spiritus stetit” a 3, stanza 1.
Example 2, Amiens, MS 162 D ff. 19v-20 “Juxta corpus spiritus stetit” a 3, stanza 2.
The second example consists in all the versions of the very simple three or four-part Italian song “La grant pena que io sento” (Amiens MS 162 D, f. 1, Washington, Library of Congress, M.2.1. L25 Case (Laborde Chansonnier), ff. 137v-138, Copenhagen, Royal Library, MS Ny kgl. Saml. 1848 2°, p. 403 and p. 411, and St. Gallen, Stiftsbibliothek, Cod. 462, p. 102, see Edition no. 19), which is only known from sources copied in France. The text and the tune is nearly the same in all of them, but the settings are so flexible and free that they could be different notations of an orally transmitted song. The song shares this condition with a lot of settings of popular tunes and with the repertory of simple sacred music.
The edition is completely traditional. It will be available as a digital music edition in one volume with indices, description of the source and editorial apparatus (as a single PDF file) – there will be more than a hundred pages of music, but during the work I’ve stopped counting the pages. But you can also download all pieces individually. They will be accessible from a page also containing links to facsimile editions of some of the sources. The use of digital editing and open access publishing gives us a chance to study music, which never had any ambition of being art or a “work”, and to let it contribute to a more balanced description of 15th century music.
The second project is a complete edition of the French chansonnier in The Royal Library, Copenhagen, MS Thott 291 8° (The Copenhagen Chansonnier – around 1470) with full editions of the concordances in the related sources, the so-called “Loire Valley” chansonniers:
It is a small vellum manuscript bound in mauve velvet, which contains a repertory of the most sophisticated secular songs, 33 in number. All songs are anonymous in the MS, but many can be identified as composition by Busnoys, Ockeghem, Morton, Hayne van Ghizeghem, Delahaye and others. This article of luxury is adorned with subtle illuminations and it is a work of art carefully designed for viewing, reading as well as for hearing. At the same time it is part of a network of relations and meanings involving four similar chansonniers from the 1460s and 1470s. Knud Jeppesen described this repertory as Burgundian, but recent research has shown that at least the illuminations in the chansonniers were executed in the workshops of the Loire Valley with Tours as the main centre, and that the manuscripts all belong to a central French courtly culture. I use the following abbreviations for the manuscripts:
Copenhagen – Copenhagen, The Royal Library, MS Thott 291 8° (Copenhagen Chansonnier),
Dijon – Dijon, Bibliothèque Municipale, Ms. 517 (Dijon Chansonnier),
Nivelle – Paris, Bibliothèque nationale, Rés. Vmc. ms. 57 (Chansonnier Nivelle de la Chaussée),
Laborde – Washington D.C., Library of Congress, MS M2.1 L25 Case (Laborde Chansonnier), and
Wolfenbüttel – Wolfenbüttel, Herzog August Bibliothek, Codex Guelf. 287 Extravag. (Wolfenbüttel Chansonnier).
An extraordinary aspect of this group of sources is that most of the Dijon chansonnier, all of Copenhagen and sections of the Laborde chansonnier were copied by the same scribe, usually named the Dijon scribe.
My first example from the Copenhagen Chansonnier is a rondeau by the little known composer Delahaye, maybe a Johannes or Jean who lived in Tours in the Loire Valley in the middle of the 15th century – nothing is sure about the identity of the composer. The chanson is a rather anxious declaration of love. The poet asks about his standing in the heart of the chosen lady “Comment suis je dans vostre cueur”. The song is found in three of the related manuscripts, in Nivelle Chansonnier, in Dijon, and it opens the Copenhagen Chansonnier. In Copenhagen it is beautifully written and adorned with grotesque illuminated letters of high quality, which act as entertaining identifiers (making it easy to find a certain song again) and maybe as oblique comments to the text and the genre. In the superius we see a figure hold forward its heart, rather accommodating, but the figure itself is some sort of hybrid between a human, a fish and a snail. The person at the tenor looks like a thief with a big sack rising up from a flower – has he stolen the heart? –, and the contratenor is a hand playing a wind instrument. There are rich opportunities for interpretation of the illuminations (see the facsimile and the following page).
I present the rondeau in score with text laid under all voices and the repetition of the half-stanzas of the rondeau fully written out. This repetition presents in many cases a problem during performances. It often has several possible solutions, which need to be considered in the edition. My edition is in a way a close reading of the source, an analysis of the relations between text and music, and a performance in score – but that is a different discussion. Indications of incipits, voice ranges, ligatures, coloration, etc. along with the usual editorial remarks are a matter of course. The edition unites a critical source edition with a practical edition. Everything needed according to my experience for a performance is present, and it is possible to reconstruct the source’s notation from the text and the notes added on the pages. This is rather uncomplicated to implement as the edition in principle only relates to one source. In this case, however, we have three sources for the chanson, and all three are edited in exactly the same way and accessible from the chanson’s main page on the site, which also contains discussions of style and expression and of the relations between the sources.
This song is a bit special because the results of the three editions of the individual sources become nearly identical. Most users will not be able to see much difference between them, and fewer will hear any differences in performances of the different versions. Why then publish all three versions? This question has several answers. Here I will only try to give one of them.
First take a look of how alike the beginnings of the three versions are: Copenhagen, Dijon and Nivelle (click the start of the three files in http://chansonniers.pwch.dk/CH/CH001all.pdf). Then let us concentrate on what in fact are the main differences between them, namely the copyists’ use of key signatures: Copenhagen has a flat only in the tenor, Dijon a flat only in the contratenor, and Nivelle has B-flats in both tenor and contratenor, and a flat before the high f” in the superius. The differences in the use of key signatures could very well be caused by an uncertainty in the perception of a characteristic passage by the Dijon scribe, who copied two of the versions: first in Dijon Chansonnier, and later in the Copenhagen Chansonnier as a more careful and concentrated piece of work.
The passage in question is bars 14-19 at the start of the second section of the rondeau. Here we find in the superius and tenor duet two movements to cadences on A (at b. 16.3 and b. 19.1), in which tenor and superius the second time exchange cadential functions. The edition in score leaves no doubt that these cadence movements must be Phrygian and involve B-flats; just look at the repeated f’s in the contratenor and the repeated melodic curve from f’’ to a’ in the superius (Example 3). But if you sing one of the upper parts from a single voice part – that is the condition when performing from the original notation – using the version in the Dijon Chansonnier, you could begin to think that this passage possibly should function as a contrasting line of music with cadences on A involving B-naturals and F- and G-sharps, especially as the two semibreves sounding B-flat in the cadences provoke normally forbidden diminished fifths with the contratenor. If the lines are performed that way the fifths become perfect (marked with blue arrows, see Example 3), but you will run into a lot of other problems involving melodic tritones from F to B-natural. The Dijon version is not the best starting point for a performance, but the singers will presumably reach the same solution as the editor after repeated tryouts.
Example 3, Dijon Chansonnier, no. 61, ff. 71v-73 “La plus bruiant”, bb. 15-19
In the Copenhagen version, which does not show any writing errors at all, the same scribe has changed his mind. Now the key signature of one flat has moved to the tenor and clearly indicates what is expected in most cases. We don’t know which signatures his model for copying had, but it is certainly possible that the scribe decided on this correction himself – maybe after trying to sing or think through the chanson. It would not be a problem to omit the B-flat in the contratenor, as nobody in their right mind would sing B-naturals here.
The independent source, the Nivelle Chansonnier, gives us flats in both low voices, and the copyist repeats the flat before the high b’ in the tenor in bar 15; there can be no doubt that the flats in the tenor are intended. He also puts a flat before the high f’’ in the signature of the superius, which clearly indicates the use of the high, fictive hexachord on c’’ – and no use of E-flats! The question is then, if this signature could convince a singer to sing B-naturals in this passage and hereby cause a dramatic shift in the harmony – we don’t know the answer. This ambiguity is however part of the aesthetics and charm of the 15th century chanson.
I could chose to offer only one PDF-file containing all three versions of the chanson by exploring the PDF-format’s possibility of storing different layers. Either three layers each containing a complete version of the chanson, which the user could chose between using internal links, or only one complete version with the variants in small overlays. In the last case I would have to select one version as the most accurate or original, as the one “nearest the intentions of the composer”. In this repertory I would soon get into serious trouble – of the kind which mars the New Josquin Edition. The choice of independent, possibly nearly identical editions of all versions reflects my wish to present an integral source (the Copenhagen Chansonnier) without any aggregations of foreign data and to take its musical testimony at face value. Also I want the separate editions for a discussion of the relations to the other sources – keeping an option open to build a presentation of one or more of the related sources some day.
My final example is the refined anonymous bergerette “La plus bruiant, celle qui toutes passe”, which only appears in the Dijon and Copenhagen Chansonniers, that is, copied by the same scribe within a few years. It is a courtly chanson, which uses terms from contemporary musical theory to put the traditional lover’s complaint into words. The poet praises the most perfect lady ”in a fictional hexachord lowering the natural B-quadratum” (muant nature en becarré la basse) and moves his “scale to another range”, joining his high hexachord “to a foreign one”. The music is consciously made to illustrate the theoretical terms of the text. Alone the key signatures give some hints. Like the first chanson the superius has a flat before f’’ along with a normal B-flat, tenor has one flat, and contratenor two flats. The flat before f’’ makes one again expect the sound of a high fictive hexachord on c’’ including the major third (mi = e’’); but in fact most E’s in the song’s first part has to be lowered to E-flat – the singer changes his high hexachord into a foreign musica falsa, a hexachord on b-flat’.
The version in Copenhagen Chansonnier has only the usual editorial problems (they are difficult enough). The slightly older version in Dijon Chansonnier shows several difficulties. The copyist encountered big problems in fitting the two couplets, which characterize the form of the bergerette, with text. The two lines in the poem apparently seemed too short, because he did not realize that the tenor defines the text declamation in this section. He repeated some words and added a short line “a ma chante pleure” stretching the text so much that it could be placed also below the short bridge passage back to the first part of the song, which is copied without text in the Copenhagen Chansonnier (see p. 3b in the Dijon version).
The edition with text in all three voices shows that the Dijon scribe’s attempt to revise the couplets was unnecessary, which he as well realized in the version he made in the Copenhagen Chansonnier. But the words “A ma chante pleure” are intriguing, and I have kept them as a possible text for the bridge, which by its triplets precisely indicates the return to the correct tempo for the semibreves in tempus perfectum in the first part of the bergerette.
The use of musical terms brings up associations to the poems of Charles d’Orléans who often used such terminology. His mother, Valentina Visconti, widow of duke Louis who was murdered in 1407, took as her emblem a picture of a chantepleure, a sort of watering can pouring out big tears. It is a picture of desolation, well known and admired in contemporary society – and absolutely fitting for the chanson, even if it was added by a scribe in want of a longer text. On the other hand, Chantepleure alsowas a dance song in popular plays, wild and happy but ending in tears and sorrow, illustrating the course of any damnable but joyous vice.
Again the possibility of publishing both versions of the song in their entirety and being able to view them alternately by a click on the computer makes it much easier for the editor to discuss the music and I hope easier for the user to follow. It also underscores the view that in the 15th century every writing down of a piece of music represents a performance, an interpretation of the sounding reality of the music notated according to contemporary needs and conventions in the scribe’s surroundings.
I probably don’t need to mention that this chanson edition is an open project, and that I hope to find interested parties to remedy its limitations – or best of all to go on with the other chansonniers in the group.
PWCH January 2008