Scratched-out notes, erased pieces, and other lacunae in the chansonnier Nivelle de la Chaussee.
In this article, I will challenge the prevailing assumption that the musical notation in books such as the Loire Valley chansonniers could not be used and understood by their owners as a means to understanding why Nivelle contains so many small-scale erasures. I will also identify a few erasures that may have been made post-copying, or in the course of use. Finally, I will offer a hypothesis for Nivelle's large-scale erasures while seeking to restore the reputation of the composer Johannes Sohier (dit Fede). Although contemporary accounts seem to have held Fede in high esteem, the erasure or otherwise incomplete state of his only extant secular works (three chansons unique to Nivelle) has prompted David Fallows to suggest that Fede must have disgraced himself in some way to spur the removal of his works from the manuscript. (3)
The five songbooks comprising the Loire Valley group have most often been studied collectively; recent research has focused on the manuscripts' relative dating and evidence of artistic exchange between the French and Burgundian courts as reflected in their repertoire. (4) The Chansonnier Nivelle de la Chaussee, or the Nivelle Chansonnier, stands apart from the other books in the group, however, in several ways. (5) Physically the largest of the Loire Valley chansonniers, Nivelle's songs are copied onto pages lined with eight staves, rather than the usual seven. In addition, the hand of the principal scribe (known as scribe A) and the style of the illuminations of Nivelle are unique in the Loire Valley group--facts that argue against it being produced in close proximity to the others. The origin of the manuscript also remains contested. Paula Higgins has suggested that Nivelle originated in Bourges, in part, because the semi-erased inscription on folio i, "de palacio bit ..." suggests the Latin name for Bourges (Bituricum), although Jane Alden has recently argued convincingly that Nivelle's origins may lie in Tours. (6) Finally, Nivelle has many small-scale erasures (compared with the immaculate pages of the Dijon scribe) as well as ten erased folios.
The idea that chansonniers participate in the late-medieval period's culture of small books has recently been gaining widespread acceptance. Similar to books of hours, chansonniers are thought to have functioned as small, lavish gifts to be enjoyed in semi-private environments, for the exclusive pleasure of their owners. A recent article by Jane Alden contextualizes chansonniers within late-medieval book culture, and art historians such as Adam S. Cohen have also focused on the meaning and value of the opening, which has particular relevance to the chansonnier, since, an entire three- or four-voice song can usually be seen and thus "heard" in viewing a single opening. (7) By the mid-fifteenth century, chansonniers' presentation was as highly prized as their musical content: the aesthetic value of an opening of a chansonnier was elevated to the extent that scribes apparently attached greater importance to beautiful, evenly-spaced musical notation, than to careful alignment of the text with music--a fact which continues to confound today's editors who wish for greater accuracy in text underlay. (8)
Although chansonniers are often our only surviving sources for the fifteenth-century song repertory, the idea that they could have served as performing documents has frequently been met with opposition by scholars. First of all, many chansonniers are very small, with pages ranging in size from 3" x 5" to 4" x 7" such that it is questionable whether two or even three individuals could read from the book simultaneously. (9) The lack of underlaid texts in lower parts or for the cantus's subsequent stanzas of text in a rondeau have also been taken as indicators that these small books could not have been used in performance. Scholars such as Craig Wright have also cited uncorrected errors, and a general lack of physical evidence of wear (including marginalia or other insertions) as reinforcement for theories that no one actually sang from these books. (10) In addition, few scholars have supported the view that fifteenth-century nobles were musically literate (apart from a few exceptions, such as Henry VIII). Finally, the few examples of iconography that could plausibly be said to demonstrate chanson performance show servants or nobles singing from a rotulus (an oblong sheet) or scroll, and not a small book. (11) Rather than serve as a performing document, it is thought that the chansonnier functioned more like an aide-memoire such that a single opening evoked a visual and aural image of a chanson in performance. (12) As such, small errors might not affect the private reader's use considering his or her supposedly limited knowledge of the musical notation, nor would it lessen appreciation for a volume of songs he or she probably already knew by heart.
There is reason to believe, however, that chansonniers' owners often had a working knowledge of musical notation. Some song collections were preceded by basic treatises on mensural notation (such as those in the Porto and Dijon chansonniers), while the Dijon Chansonnier's introductory treatise is particularly notable for being in the vernacular. (13) I would argue that we also overestimate the relative difficulty of the notation in these sources; the chanson repertoire from the second half of the fifteenth century is comparatively easy to read and understand compared with music from the fourteenth and first half of the fifteenth centuries. With the increasing popularity of imperfect tempus and prolation, performers or readers of secular music were forced to confront far fewer questions of imperfection or alteration, little to no hemiola or triplet coloration, and almost no proportions beyond the occasional use of proportio to duplex. As a result, the reader of ca. 1470 only rarely had to make determinations about the values of notes based on context within a perfection. The user-friendly device, the custos (a sign placed on the staff at the end of one line of music to indicate the first pitch of the next), also implies use rather than casual reading or appreciation; earlier repertoires preserved in presentation manuscripts such as the Chantilly Codex (Musee Conde MS 564) do not use them. (14) Use of ligatures also became increasingly simple in the song books from the 1460s onward: ligatures in Nivelle and contemporaneous chansonniers are rarely more than two or three notes long, with c.o.p. ligatures (cum opposita proprietate, or "having opposite propriety": a ligature written with an ascending stem to the left) being by far the most common. The notation treatise that precedes the Dijon Chansonnier reflects this limited notational vocabulary. It is short and provides only the necessities for the user of the book. It does not address perfect prolation at all, and deals with only simple ligature combinations. These facts strongly suggest that a book like Nivelle was meant to be enjoyed through use, and that it is conceivable that it could be read and understood by its owner (s). (15)
The strikingly high number of small-scale corrections in the main layer of Nivelle, be they crossed-out or erased notes, also call attention to the issue of musical literacy and the book's user. If the Nivelle Chansonnier was not meant to be "used" as such, what was its owner to think of the crossed-out notes or messy scrapings in practically every third piece of his or her presumably expensive presentation manuscript meant to anthologize many of the owner's favorite songs? (16) If we consider that a piece found on a single opening was meant to be more or less "evocative" of the musical experience rather than a literal representation of it, then why make corrections?
Certainly, some of the corrected errors in Nivelle were clearly made in the course of copying. Albert d'Haenens's study of the iconography of writing in the Middle Ages finds significant evidence that scribes held the quill in the right hand, and a penknife in the left, "at the ready for erasing or holding the parchment steady." (17) And in his study of thirteenth-century musicography in the Manuscrit du roi, John Haines reminds us that "in the middle ages, to write, then, meant also to erase." (18) Yet, we find no evidence of similar corrections in works copied by the Dijon scribe, who was responsible for copying the entire Dijon Chansonnier (the largest of the Loire Valley chansonniers), the Copenhagen Chansonnier, and parts of Laborde. (19) Is it possible that Nivelle's scribe A's methods for copying from examplars was simply less systematic than that of the Dijon scribe and thus prone to inconsistencies or mistakes? Or rather, does the high number of corrections imply that scribe A was not so familiar with his sources, or with copying music generally? Or perhaps lingering evidence of erasures signals scribe A's departures from his exemplars?
In his seminal work on medieval scribal practice, Les manuscrits, Alphonse Dain summarized four stages in the copying process: the scribe read the exemplar, remembered the text, recited (or, in the case of music, sang) it internally, and then copied it down. (20) From his examination of troubadour manuscripts, John Haines has estimated that "the scribe looked up on the average every one to six--and more often than not, fewer than three--notes." (21) The numerous single-note erasures of Nivelle's scribe A testify to a copying practice similar to the one described by Haines.
Scribe A appears to have worked semi-simultaneously in copying both text and musical notation in Nivelle, checking his work against the exemplar every three or fewer notes, which allowed him to catch mistakes quickly. Most of his copying errors were errors of omission (where the eye leaps ahead in the exemplar to the same note, skipping parts of the passage), or repetition, and occasionally errors of transposition (up or down a step or a third). An examination of scribe A's copy-editing process also shows that he typically made a diagonal Crosshatch through errors (from the upper right-hand corner of a note-head to bottom left) before erasing them. For instance, figure 1, a detail from the cantus of Pour entretenir, demonstrates that the scribe looked up every two or so notes, and marked errors to be erased with a Crosshatch through the note-head from upper right to lower left. Similarly, erasures in the cantus of Ockeghem's Ma bouche ril and the contra of the anonymous rondeau. Flora, flora (figures 2 and 3, respectively), indicate that errors of repetition were marked with similar diagonal crosshatches and subsequently erased. An erasure in the contra of the verse of Ockeghem's Tant fuz gentement, however, implies a different rationale. As one can see in figure 4, an oblong extension of the second ligature on the second stave of the contra was erased and replaced by two free-standing breves whose pitches and values are identical to the erased material. In that case, scribe A did not make a mistake of repetition, per se. It would seem that the exemplar from which he copied probably contained the six-note ligature, but he made a conscious decision--that is, an editorial decision--to shorten the ligature he had copied out exactly by erasing the oblong portion and copying out the final two breves separately. The converse procedure, which assumes that scribe A read the ligature, remembered it, sang it, and wrote out a six-note ligature only to look up again at the exemplar and realize that the exemplar had left the last two breves freestanding and not ligated (prompting the resulting "correction"), suggests that scribe A understood musical notation well enough to translate what he sang into an extended ligature, but that he was obstinate in his adherence to the exemplar since he reverted to its reading despite the fact that it was redundant. (22) I find this idea less convincing, however, since it implies that scribe A was remembering and copying at least six notes or more before referencing the exemplar (which is not supported by his quickly-corrected errors of repetition or omission elsewhere).
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
While it is clear that some erasures are clean, and occurred as part of a copy-editing process either during or immediately following copying, a few corrections seem to have been executed later. For instance, in figure 5, one can see that the second stave of the contra of the unique Se vous me voulez contains three erasures. The first erased character appears to have been completed cleanly and quickly. The fact that it is followed by a minim of the same pitch suggests that it may have been an error of repetition on the part of the scribe. The next two erasures, however--apparently a B[??] semibreve, and c minim--must be read to complete the counterpoint of the song. They look as though they were executed roughly, and the vestigial Crosshatch across the c minim runs opposite from scribe A's normal practice, suggesting that these emendations were made sometime later, and by another person. (23) The unique virelai attributed to Ockeghem, Tant fuz gentement (fols. [lvi.sup.v]-lviii), also has many erasures in the tenor and contra parts of both the refrain and the verse. While some of these appear to be errors of repetition or transposition, figure 6a-b details erasures in the tenor and contra in the second phrase of the refrain of Tant fuz gentement correct dissonances and parallel octaves as demonstrated in figure 7a-c. The resulting uneven spacing, and darker ink of the insertion at the end of the first stave of the contra even suggest that the dot of addition on the c.o.p. ligature and minim were fit in later to correct the counterpoint.
[FIGURE 5 OMITTED]
[FIGURE 6a OMITTED]
[FIGURE 6b OMITTED]
[FIGURE 7 OMITTED]
In addition to small-scale erasures and crossed-out notes, the Nivelle Chansonnier is unique in the Loire Valley group for its large-scale erasures. In its current state, the manuscript has ten erased folios, or most of four songs, three of which are unique to Nivelle. (24) Two of these four songs occur in the first gathering of the manuscript, while the others are found contiguously in the seventh gathering, and are both attributed to Fede. To describe these pages as completely erased, however, is not entirely accurate. The illuminated letters were, without exception, retained, as were the staff lines (to the extent possible due to the scratching). In some cases, clefs and/or mensuration signs, and part identifiers ([T]enor or [C]ontra) were left intact. (25) The fact that whoever erased the pages in Nivelle left illuminated letters, clefs, and staves intact strongly suggests the intention to replace the erased works with other songs. (26)
While textual palimpsests are not unusual, musical palimpsests are quite exceptional. Besides the erased pages in Nivelle, I know of no other instances of fully erased pieces in a polyphonic manuscript. Elizabeth Aubrey has mentioned a single musical palimpsest--the mono-phonic song Ay tan gen vens (Bibliotheque national de France, fonds fr. 22543, fol. 43)--as "the only one of its kind in the troubadour manuscripts ... where an entire melody was written and then erased so that a completely different one could be entered." (27)
Unfortunately, all three of Fede's pieces in the current state of Nivelle are incomplete, a fact that, as noted earlier, prompted David Fallows to hypothesize that Fede must have done something to disgrace himself in the eyes of the owner of the book that would spur his works to be effectively "rubbcd-out." If one considers, however, that only one of Fede's three works in Nivelle was completely erased (A la longue) then the picture begins to change. Furthermore, if we acknowledge that the other two erased works are or may be by famous composers represented elsewhere in the manuscript (Ma plus amee, and Quant ce vendra), then the assertion that Fede and his works were singled out for removal becomes suspect.
Nivelle's first erased piece--the unique, anonymous virelai, Ma plus amee de re monde (fols. [ii.sup.v]-[iv])--bears strong resemblance to early works by Ockeghem, including Fors settlement l'attenle, which follows Ma. plus in the manuscript. (28) Stylistically, Ma plus is very similar to another unique virelai attributed to Ockeghem found later in the manuscript, Tant fuz gentement (fols. [lvi.sup.v]-lviii). (29) Both refrain and verse of the virelai are in imperfect tempus, both have virtually homophonic openings of their A and B sections, and imitation between cantus and tenor is limited to a single phrase in both works. In addition, both Tant fuz and Ma plus amee feature the rhythmic signature of dotted semibreve followed by two semi-minims, and have very limited syncopation. Ma plus is also closely related to Ockeghem's Fors seulement. As Clemens Goldberg has noted, Ma plus contains several musical quotes from Fors settlement. (30) In fact, as one can see from Nivelle's inventory (appendix A), Fors seulement follows Ma plus in Nivelle, and I would like to offer the possibility that Ma plus amee might also be by Ockeghem. Barring that, one could plausibly argue that Ma plus is a tribute to Fors seulement by another (perhaps) major composer. The other erased work in the first gathering is Quant ce vendra, a famous piece that occurs in nine sources besides Nivelle. Of its ten sources, it is attributed twice to Busnois (in the Dijon Chansonnier, and in a later hand in the Laborde Chansonnier), and once to Ockeghem (in the Escorial Chansonnier, though perhaps this refers to authorship of the si placet contratenor) .(31)
Fede's only fully erased work is the virelai A la longue (fols. [xlix.sup.v]-li).(32) While A la longue is interesting for its melodic sequences, it remains somewhat old-fashioned compared with other works in the manuscript. In addition, having reconstructed the piece from its erased state, it clearly contains a variety of compositional or contrapuntal errors including accented dissonances, a second-inversion sonority, and poorly masked parallel octaves. Fede's Tout a sa dame precedes A la longue in Nivelle, though only its cantus was erased (fol. [xlvii.sup.v]); its tenor and contra remain intact. Fede's third chanson, Mon cuer et may, the last piece in the original layer of the manuscript, is incomplete only because what would have been the tenth gathering is missing; it was not rubbed out, per se. (33) The facsimile of this piece clearly demonstrates that the music and text for the first system of the cantus was erased, and then consciously recopied or repaired. The recopied text is difficult to decipher, however, and the repaired staff lines are a little too wide; perhaps they were executed hurriedly since they appear to have been drawn without a rule. As a result, the redrawn note-heads are large and awkward.
Nivelle's large-scale erasures beg two questions: why were they erased, and when? Attempting to answer these questions has required me to make several assumptions. The first and largest assumption I have made is that the large-scale erasures were executed at the same time. It also seems reasonable to assume that these erasures took place after the book was illuminated, foliated/assembled, and bound rather than before these steps had been completed. In addition. I have assumed that the pages were erased at the time that the first additions to the manuscript were made and not sometime later. That may seem like the largest assumption of all, but I have to wonder what scribe B would have thought when he copied N'aray je jamais onto the first opening of the manuscript if the second opening and following were already blank from full erasure. No doubt, it would have been quite strange to be copying new music into a book that had already effectively been defaced.
Instead, I would like to offer the suggestion that scribe B, charged with adding several newer, popular songs to the existing book at the request of a new owner, erased the pieces himself with the intention of copying new pieces in their place. He planned the pieces to insert (all popular, well-known works slightly post-dating those in the main layer of Nivelle), made space for them (by erasing the existing works efficiently with a pumice stone instead of a knife while he retained clefs or key signatures as needed), but then found his task complicated by the fact that he would have to replace the erased songs with others whose titles corresponded to the extant illuminated letters. (In fact, none of the titles of the songs he added correspond to the extant initials on the erased pages.)
It is interesting, too, that apart from the erased works attributed to Fede in the seventh gathering, the large-scale erasures are concentrated in the first gathering. Jane Alden, Clemens Goldberg, and Paula Higgins, among others, have all commented on the apparent significance of initial chansons and/or on the importance and interrelatedness of pieces selected near the beginning of a chansonnier. (34) Perhaps the erasure of three openings in the first gathering, with the intent to replace them with other songs, could have been an attempt to remake the book to reflect the tastes of a new owner.
I recognize, however, that this hypothesis leaves open the question of why Fede's virelai A la longue was erased in the seventh gathering. Perhaps A la longue was deemed outmoded by scribe B or by the book's new owner, or perhaps its contrapuntal errors marked it as flawed, and therefore removable. Alternatively, Fede's virelai could simply have been seen as the most expendable--that is, it was convenient to choose a virelai by Fede for replacement. By erasing a virelai, the book loses only one song, but two consecutive openings are freed up to interpolate two rondeanx. Furthermore, Fede's A la longue is surrounded by virelais and ron-deaux by Busnois and Ockeghem in gatherings six and seven, which, being better-known or by famous composers, may have seemed more essential or whose inclusion added more value to the anthology as a whole. (35) After all, chansonniers were essentially "preservation" texts that derived their cultural value from the established popularity of the songs they contained (and perhaps secondarily through their pieces' attributions to known composers). By contrast, unique works or those with few concordances in other manuscripts served to fill out the manuscript and are generally thought to be by local composers, or perhaps by the scribes themselves. (36) Also, if we consider Fede as a "local" composer relative to Nivelle's initial planning and copying, then his work's subsequent erasure presuming it coincided with the addition of new works to refashion the book for a new owner) might also imply a geographical relocation from Tours to Bourges, for instance, that could be supported by the later (now partially erased) inscription on Nivelle's folio i, "de palacio bit. ..."
While there is no firm evidence that scribe B erased the pages, it is an interesting coincidence that the contents of a total of five openings were rubbed out in Nivelle, while scribe B added rondeaux on six openings." (37) The first opening was always blank when scribe B sat down to make his additions, and he would have needed to create five more blank openings to complete six pieces. Ultimately, scribe B contributed six chansons to Nivelle, adding his first song, N'aray je jamais, to the (always blank) initial opening of the first gathering, and attaching the last five in a separate gathering at the end of the book. Scribe B left the last two openings of that new final gathering ruled but blank, where two different scribes, C and D, finally copied the last two pieces of the existing collection. Tart aura by Molinet, and Ha qu'il m'enuye (variously attributed to Fresneau and Agricola). Whether or not scribe B was the one to erase the folios to provide places for new chansons to be inserted, it is clear that Fede was not singled out for removal from Nivelle; the famous piece Quant ce vendra (probably by Busnois), and the unique virelai related to Ockeghem's Fors settlement (Ma plus amee de ce monde) suffered a similar fate.
Fede appears to have been a musician of some stature since he was mentioned alongside Ockeghem, Dufay, and Binchois in Simon Greban's poem Complainte de la mort de Jacques Millet (1466), in addition to being cited in several other lists of prominent musicians of the period. (38) Significant documentation supports a long, varied career in which Fede may have worked alongside many of the most famous musicians from the third quarter of the fifteenth century. Documentation of his career ranges from the Ferrarese court, Cambrai Cathedral, and Sainte Chapelle in Paris in the late 1440s, to the court of Charles d'Orleans at Blois in the early 1450s, from the chapel of Marie d'Anjou (1462-63), to San Pietro in Rome in 1466, and from Sainte Chapelle in Bourges (1472-73), to the household chapel of Louis XI (1473-74). (39) In fact, there is so much information relating to Jean Sohier, dit Fede, that Jane Alden has recently suggested that there could be more than one individual of the same name. (40) In particular, she has queried whether the two Magnificat antiphons attributed to Fede in the Ferrarese court choir-book (Modena, Biblioteca Estense, [alpha].X.1.11 [ModB])--the three-void' O lumen ecclesie, and fauxbourdon setting of Magne pater sancte dominice--could be by the same man that authored the chansons in Nivelle. (41)
Despite the fact that Fede's chansons in Nivelle and the antiphons in ModB may have been composed as much as fifteen to twenty years apart and notwithstanding that they were created to serve very different functions (the superius of the antiphons have their origins in chant paraphrase), I would like to suggest that Fede's compositional style is sufficiently idiosyncratic that they may be traced to a common author. His sacred and secular works are most commonly in three voices, with occasional duo texture, and the lower parts occupy the same range; none of his works use a low contratenor. Double leading-tone cadences predominate, V-I cadences in the contra are used sparingly, and the octave-leap variety does not occur. Conjunct motion generally predominates in the superius and tenor, while Fede appears to favor a disjunct contra rife with syncopation. As one can see from the reconstructions of Tout a sa dame and A la tongue, his contras also frequently span or outline wide intervals, notably sevenths, and there is little to no imitation between voices. (42)
We can observe these conventional traits as well as Fede's more progressive tendencies by examining the reconstruction of his virelai, A la longue. His experimentation with the virelai links him to the French court circles that included Busnois and Ockeghem, who led the vogue for this form in the 1460s and 1470s. (43) Providing a contrast to his normal predilection for rhythmic counterpoint and syncopation, Fede uses virtually homophonic settings for the penultimate phrases of the refrain of A la longue, and in his rondeau, Tout a sa dame. In addition, the extended sequence that closes A la longue's refrain is quite remarkable. Typically, the verse of a virelai may have a contrasting mensuration or musical texture from its refrain, as Fede's does. Fede shifts from O in the refrain to [cents] in the verse, and from what can be made out from the ultraviolet scan of the opening, cantus and tenor are set virtually homophonically in parallel sixths. It is interesting that Fede took a more homophonic approach to the verse as it echoes Ockeghem's style in the verse of Tant fuz gentement. The first phrase of the verse also adopts melodic sequence as its organizing principle, picking up this trend from the last phrase of the refrain. Ultraviolet exposure of fol. li also makes clear that no contra voice was ever copied for the refrain of the verse. (41) However, since it is completely atypical for a three-voice chanson to suddenly shift to two voices, I have reconstructed a contra for the verse. Apparently the illuminator was also confused by the lack of text or music for the contra of A la longue's verse, and added a letter where that part should appear (though notably not a C for contra, rather a V that matched the text of the cantus Vous estes si belle).
Unfortunately, Fede's Man cuer el may is incomplete because the penultimate gathering of the manuscript is missing. Only the cantus part of the refrain survives, and thus reconstruction of the lower voices is more speculative with respect to style. I include my reconstruction for the refrain of Mon cuer et moy in appendix B, where I set equal-range lower voices including a disjunct contra, avoided imitation, and took cues from the rhythmically restrained approach to imperfect tempus that Fede showed in the verse of A la longue.
Apart from the two Magnificat antiphons in ModB and the three attributed chansons in Nivelle, Fede was also credited with having composed L'homme banny in the chansonnier Firenze, Biblioteca nazionale centrale, Magl.XIX.176. The stylistic analysis offered by reconstructions of Fede's chansons from Nivelle, however, makes clear that L'homme banny is not the work of Fede. Instead, they further confirm L'homme bunny's attribution to Barbingant in the Mellon Chansonnier (which is also corroborated by writings of Tinctoris, Gaffurius, and Giovanni del Lago). (45) Conflicting attributions, however, may tell us less about who definitively composed the piece, so much as they can show affiliations between composers such as those drawn by David Fallows between Barbingant and Ockeghem as evidenced by conflicting attributions for Au travail suis, or between Barbingant and Compere. (46) Similarly, Gerald Montagna has convincingly argued that conflicting attributions for works by Firminius Caron in Italian manuscripts that name Busnois and Morton, among others, as their composer, do less to support theories of Caron's career in Italy than they solidify his association (and thus confusion) with other well-known Burgundian composers. (47) In this way, confusion between Barbingant and Fede in Italian manuscripts can perhaps be understood as representing a close association between the composers. "Recovery" of Fede's secular works from Nivelle's erased pages may help to properly situate his style and potential influence among his more famous colleagues, just as deconstructing Nivelle's small-scale post-copying erasures may help us to understand how and by whom chansonniers were used, and ultimately modified. Ultimately, extensive paleographic work has yet to be completed for Nivelle, which may enhance our understanding of when and why the large-scale erasures were made. Indeed, I hope the ideas put forth in this paper will spur further discussion of the role that chansonniers played in the lives and leisures of their users.
Inventory of the Chansonnier Nivelle de la Chaussee
 missing from MS
() composer or poet attribution in concordant sources
No. Composer Incipit Scribe Folios Gathering 1. (Morton) N'aray je B [i.sup.v] - jamais [ii] miculx * 2. Ma plus amee A il-iii de ce monde Car je rous iii-[iv] ay este loyal 3. (Ockeghem) [Fors A [[iv.sup.v] - 1 seulement v] Pattente] Qu'il n'est [[v.sup.v]] - doulcur vi 4. (Busnois/ Quant ce A [vi.sup.v] -vii Ockeghem) pendra * 5. Se je demeure A [[vii.sup.v]] - / vii Puisque dieu [[v.sup.v]] - a voulu ix]]-xi 6. Pour le mal A [ix.sup.v] - x qu'on vous fait porter * 7. Pour les A [x.sup.v] - xi biens qu'on vous je parcoy * 8. Se vous me A [xi.sup.v] - 2 voules estre xii bonne 9. Quant de mon A [xii.sup.v] - cuer vous xiii feray part 10. S'il vous A [xiii.sup.v] - plaist bien/ xiv Car que je [xiv.sup.v] - sceusse xv vivre 11. Dufay Puisque vous A [xv.sup.v] - estez xvi campieur 12. Puisqu'a vous A [xvi.sup.v] - servir xvii 13. Le tres des A [xvii.sup.v] - plus eureux xviii 14. (Busnois) Quant vous me A [xviii.sup.v] - ferez plus de xix bien 15. (Pullois) Se ung peu A [xix.sup.v] - d'esperance xx * 16. Flora, flora A [xx.sup.v] - tres noble et xxi digne * 17. A quoy tient A [xxi.sup.v] - 3 il (Fredet) xxii * 18. Querez A [xxii.sup.v] - ailleurs xxiii paille en estrain 19. Pour avenir a A [xxiii.sup.v] - mon atainte xxiv 20. (Fede/ L'omme banny A [xxiv.sup.v] - Barbingant) de sa xxv plaisance 21. La fiance que A [xxv.sup.v] - j'ay xxvi 22. Binchois/ Tout a par A [xxvi.sup.v] - (Frye) moy xxvii 23. Busnois Lessez A [xxvii.sup.v] - Dangier/ Amours m'a xxviii fait par Bel Acucil [xxviii.sup.v] - * 24. Delahaye Puis qu' A [xxx.sup.v] - 4 aultrement ne xxxi puis avoir * 25. Delahaye Pour les A [xxx.sup.v] regretz que -xxxi j'ay 26. Delahaye Puis qu'il A [xxxi.sup.v] - convient xxxii 27. Delahaye Comment suits [xxxii.sup.v] - ju de vostre xxxiii cuer 28. Busnois C'est vous en A [xxxiii.sup.v] - qui j'ay xxxiv esperance/ Ma maistresse [xxxiv.sup.v] - et mon tout xxxv seul 29. Busnois Soudainement A [xxxv.sub.v] - mon cuer/ xxxvi Car il ne [xxxvi.sup.v] - 5 puet rien [xxxvii 30. Busnois Est mercy de A [xxxvii.sub.v] - quoy xxxviii 31. D'une belle A [xxxviii.sup.v] jeune File/ xxxix Coquille bobille 32. Delahaye O Dieu A [xxxixsup.v] - d'Amours xl 33. Busnois Ma damnisclle A [xl.sup.v] - ma xli 34. Bl.STV.1S Cest bien A [xli.sup.v] - malbecur xlii 35. Charge de dueil/ Ha se A xlii'-xliii mon xliii'-xliv ennemy mortel 36. (Businois) En tous les A [xiiv.sup.v] - 6 lieux/ Que xlv xlv'-xlvi pietist a dieu 37. Par dieu A [xlvi,sup.v] - madame/ Que [xlvil.sup.v] puis je xlvii'-xlviii avoir * 38 Fede T out a sa A xlliil-xlix dame A*39. Fede A la longue A [xlix.sup.v] - jay bien l cagnu/ I ous estes gente belle of bonne 40. Binchois/ Je ne vis A [li.sup.v] - (Dufay) onques la lii pareille 41. Okeghem Ma bouchc A [lii.sup.v] - rit/ Iiii Ha cuer liii' - liv pervers 42. Busnois Le corps s'en A [liv.sup.v] - va iv 43. Ockeghem S'elle A [lv.sup.v] - mamera/ lvi Petite * 44. Ockeghem Tant fuz A lvi-lvii Ivii' - gentement lviii resjouy/Si haultement avez party 45. Busnois/ Vous marchez A lviii-lix (Isaac) du bout du pie * 46. Pour A [lix.sup.v] - entretenir lx I'allianee 47. Dufay I.es doleurs A [lix.sup.v] - dont me sens lxi 48. Jamais si A [lxi.sup.v]' - 8 bien. lxii 49. Mun cueur a A [Ixii.sup.v] - demy se lxiii depart 50. Boubert L'omme A [Ixiii.sup.v] - enrage lxiv 51. Dufay Ma plus A [Ixiv.sup.v] - mignonne de lxvi mon caeur * 52. Delahaye Tout au long/II A [lxv.sup.v] - n'est lxvi Jacobin/Jennette des Coqueles 53. Ockeghem/ D'ung aultre A [lxvi.sup.v] - (Barbingant) amer lxvii 54. Delahaye Mort j'apelle de A [lxvii.sup.v] - 9 ra riguer lxviii (Francois Villon) 55. Barbingant Esperani que mon A [lxviii.sup.v] bien vendra - lxvi 56. Ockeghem/ Au travail suis A [lxix.sup.v] (Barbingant) -lxx * 57. En tout honneur A [lxx.sup.v] - lxxii 58. J'ay pris amours A [lxxi.sup.v] - lxxii * 59. Fede Mon cuer ct mov A [lxxii.sup.v] avons (Blank) [73.sup.v] 60. Tant est mignonne B [73.sup.v] - 74 ma pensee 61. (Agricola) J'ars de desir si B [74.sup.v] - 75 enflame d'amer 62. Plus voy mon B [75.sup.v] - 76 10 mignon 63. (Convert) Se mieulx ne B [76.sup.v] - 77 vient d'amours 64. (Michelet) S'il advient que B [77.sup.v] - 78 mon dueil 65. (Molinet) Tart aura mon C [78.sup.v] - 79 cuer 66. Ha qu'il D [79.sup.v] - 80 (Fresneau/Agricola) m'enuye/J'ay des ennuis
(1.) Thanks to Jane Alden for sharing her forthcoming study of the Loire Valley chansonniers with me.
(2.) Craig Wright, "Voices and Instruments in the Art Music of Northern France during the Fifteenth Century: A Conspectus." in International Musicological Society, Report of the Twelfth Congress, Berkeley, 1977, ed. Daniel Heartz and Bonnie Wade (Kassel: Barenreiter; Philadelphia: American Musicological Society, 1981), 644.
(3.) David Fallows, "Fede, Johannes [Sohier, Jean]," Grove Music Online, http://www.oxfordmusiconline.com/public/ (accessed 28 May 2009).
(4.) Duff James Kennedy compared the chansonniers in his dissertation, "Six Chansonniers francais: The Central Sources of the Franco-Burgundian Chanson" (Ph.D. diss., University of California at Santa Barbara, 1987). Individual studies include The Copenhagen Chansonnier: The Thott 2918 Manuscript in the Royal Library of Copenhagen, ed. Knud Jeppesen (New York: Broude Brothers, 1965; reprint with new remarks of the edition Copenhagen: Levin & Munksgaard, 1927): Charles Edward Barret, "A Critical Edition of the Dijon Chansonnier: Dijon. Bibliotheque de la ville, MS 517 (ancien 295)" (Ph.D. diss., George Peabody College for Teachers of Vanderbilt University, 1981); Der Wolfenbutteler Chansonnier: Hezog August Bibliothek, Wolfenbuttel, Codex Guelf. 287 Extrav., ed. Martella Gutierrez-Denhoff, Musikalische Denkinaler, 10 (Mainz: Schott, 1988); Clemens Goldberg, Das Chansonnier Laborde: Studien zur Intertextualitat einer Liederhandschrift des 15. Jahrhunderts, Quellen und Studien zur Musikgeschichte von dei Antike bis in die Gegenwart, 36 (Frankfurt; New York: Lang, 1997); Jane Alden, "Makers of a Songbook: The Scribes of the Laborde Chansonnier" (Ph.D. diss., University of North Carolina, 1999).
(5.) For a more complete description of Nivelle, and a review of its recent history, see the introduction to the facsimile edition Chansonnier Nivelle de la Chaussee: Bibliotheque nationale, Paris, Res. Vmc. ms. 57, ca. 1460, introd. by Paula Higgins (Geneva: Minkoff, 1984).
(6.) Ibid., ix. Nivelle conspicuously lacks, however, works by composers closely associated with Bourges (such as Philippe Basiron). Instead, Alden's arguments in a forthcoming study in favor of Tours as the origin for the planning and copying of Nivelle are based on repertorial considerations including local composers, and the style of Nivelle's illuminations.
(7.) Jane Alden, "Reading the Loire Valley Chansonniers," Acta Musicologica 79, no, 1 (2007): 1-31; Adam S. Cohen, The Open Book: Reflections on Medieval Art and Experience (forthcoming).
(8.) Issues associated with ambiguous text underlay in mid-fifteenth-century sources have been treated by Graeme Boone, Patterns in Play: A Model for Text Setting in the Early French Songs of Guillaume Dufay, American Musicological Society Monographs (Lincoln: University of Nebraska Press, 1999); Leeman L. Perkins, "Toward a Rational Approach to Text Placement in the Secular Music of Dufay's Time," in Papers Read at the Dufay Quincentenary Conference, Brooklyn College, December 6-7, 1974, ed. Alan W. Atlas, 102-14 (New York: Brooklyn College, 1974); Leeman L. Perkins, "Text and Music in the Chansons of Busnoys: The Editorial Dilemma," in L 'edizione critica tra testo musicale e testo letterario: Atti del convegno international, Cremona, 4-8 ottobre 1992, ed. Renato Borghi and Pietro Zappala, 165-80, Studi e testi musicali; nuova ser., 3 (Lucca: Libreria Musicale Italiana Editrice, 1995). See also Louise Litterick, "Performing Franco-Netherlandish Secular Music of the Late-Fifteenth Century: Texted and Untexted Parts in the Sources," Early Music 8, no. 4 (October 1980): 474-85.
(9.) Wright, "Voices and Instruments in Northern France," 644.
(12.) Alden, "Reading the Loire Valley Chansonniers," 19-20.
(13.) Facsimile editions of both Porto and Dijon are available. See Dijon Bibliotheque publique Ms. 516, introd, by Dragan Plamenac, Publications of Mediaeval Musical Manuscripts, 12 (Brooklyn: Institute of Mediaeval Music, 1971); Porto 714: Um manusrito precioso, introd. by Manuel Pedro Ferreira, Campo da musica. 5 (Porto: Campo das Letras, 2001).
(14.) See the recent facsimile edition with commentary, Codex Chantilly Bibliotheque du chateau de Chantilly Ms. 564, ed, Yolanda Plumely and Anne Stone, 2 vols., Epitome musical (Turnhaut: Brepols, 2008).
(15.) Perhaps it was also one of the book's owners or users who added text in a different (perhaps nonprofessional) hand on several pages. For instance, fol. [xxxii.sup.v] (Delahaye's Comment suis je de vostre cuer) has additional text in an unknown hand reading, "[ ]mant suis je de vostre cueur / Demener si grant rigueur a son loyal / serviteur plain de douleur qui mon." This fragmented text uses some of the same words as the text of the poem, and attempts to use the same rhyming sounds, though the line lengths do not match the rondeau. The song Querez ailleurs (fol. [xxii.sup.v]) also has added text (perhaps in the same hand as on fol. [xxxii.sup.v]. There the first stanza of the rondeau is written out again underneath the music of the cantus, with a slight alteration to the grammar of the first line (nota bene: it is not the same as how the text appears in the Rohan Chansonnier). In addition, a latin inscription, "deus meus ad te de luce vigilo non tradas tradas." appears in a different (more stilted) hand on the same page. If this is a pen trial, perhaps these psalm fragments were added by a distracted user.
(16.) Nineteen out of fifty-eight pieces copied by scribe A contain small-scale erasures.
(17.) Albert d'Haenens. "Ecrire, un couteau dans la main gauche: Un aspect de la physiologic de l'ecriture occidentale aux XIe et XIIe siecles," in Clio et son regard: Melanges d'histoire, d'histoire de l'art et d'archeologie offerts a Jacques Stiennon a l'occasion de ses vingt-cinq ans d'enseignement a l'Universite de Liege, ed. Rita Lejeune and Joseph Deckers (Liege: Mardaga, 1982), 135.
(18.) John Haines, "Erasures in Thirteenth-Century Music," in Music and Medieval Manuscripts: Paleography and Performance, ed. John Haines and Randall Rosenfeld (Aldershot, Hants, Eng.; Burlington, VT: Ashgate Press, 2004), 66.
(19.) The number of errors in his work apparently decreased through the course of copying these three manuscripts. Jane Alden has been revising the currently accepted relative dating of the copying of the Loire Valley chansonniers for release in a forthcoming publication, and I thank her for sharing her work with me.
(20.) Alphonse Dain, Les manuscrits, 3d ed., Collection d'etudes anciennes (Paris: Belles Lettres, 1975), 40-46.
(21.) Haines, "Erasures," 71.
(22.) Unfortunately, since Ockeghem's Tant fuz gentement is unique to Nivelle, we cannot compare the transmission of details such as ligatures in other sources.
(23.) Incidentally, these erasures persist alongside uncorrected errors of transposition in the final phrase of Se vous me voulez's contra.
(24.) These are fols. [ii.sup.v]-[iii.sup.v], [vi.sup.v]-vii, [xlviii.sup.v], and [xlix.sup.v]-li. See appendix A for a full inventory of Nivelle.
(25.) Clefs were retained for the cantus, tenor, and contra on fols. ii-iii (the refrain of Ma plus amee de a monde). In addition, the first stave of the cantus on fol. iiv has an extant mensuration sign (C). Most Clefs and key signatures (one flat) for the cantus of Quant ce vendra (fol. [vi.sup.v]) were also not erased. In A la longue (fol. 1). clefs and mensuration signs (O) For the tenor and contra of the refrain were retained.
(26.) However, as John Haines has noted ("Erasures," 63--64), "once the medieval scribe had scratched out a note, the spot where the parchment had been roughed up could usually not be used since the ink would blot, and in most cases, the scribe needed to continue writing as soon as be had erased." this assertion admittedly raises the question whether new pieces could Feasibly have been copied on the erased pages. It is possible, however, that some kind of soluble ink was used. In Fact, some of the small-scale erasures are completed so cleanly) (appearing almost to have been Faded out) that such a suggestion seems plausible. The nature of the erasure of Quant ce vendra also suggests some means of erasure besides scratching,
(27.) Elizabeth Aubrey, The Music of the Troubadours (Bloomington: Indiana University Press. 1996), 47.
(28.) See edition of Ma plus amee de ce monde in appendix B.
(29.) In Johannes Ockeghem, Collected Works, ed. Richard Wexler with Dragan Plamenac. 3 vols. (Philadelphia: American Musicological Society, 1947-92), 3:91.
(30.) Clemens Goldberg, "Werke, Quelle, Analyse: Betrachtungen zum Chansonnier Nivelle de la Chaussee," in (Quellenstudium und musikalische Analyse: Festschrift Martin just zum 70. Geburtstag, ed. Peter Niedermuller, Cristina Urchueguia, and Oliver Wiener (Wurzburg: Ergon, 2001). 9-10.
(31.) David Fallows, A Catalogue of Polyphonic Songs, 1415-1480 (Oxford; New York: Oxford University Press, 1999). 335. Fallows cites The Mellon Chansonnier, ed. Leeman L. Perkins and Howard Garey, 2 vols. (New Haven: Yale University Press. 1979). 2:250. This subject is also broached by Perkins in "Conflicting Attributions and Anonymous Chansons in the 'Busnoys' Sources of die Fifteenth Century," in Antoine Busnoys: Method, Meaning, and Context in Late. Medieval Music, ed. Paula Higgins (Oxford: New York: Oxford University) Press. 1999). 319-20.
(32.) See edition in appendix B.
(33.) Strangely enough, it was damaged, and 1 do not have an explanation for why. An ultraviolet image of this page might be able to tell us more.
(34.) Jane Alden. "Reading the Loire Valley Chansonniers." 16-18. See also Clemens Goldberg, "Reading Laborde: The Significance of Johannes Ockeghem's Chansons in the Chansonnier Laborde," in Johannes Ockeghem: Actes du XLe colloque international d'etudes humanistes, Tours, 2-8 fevrier 1997. ed. Philippe Vendrix, 253-70. Epitome musical,  (Tours: Klincksieck, 1998). Paula Higgins has also suggested that chanson manuscripts that open with a Marian motet could signal those books' female patronage in "The 'Other Minervas': Creative Women at the Court of Margaret of Scotland," in Rediscovering the Muses: Women's Musical Traditions, ed. Kimberly Marshall (Boston: Northeastern University Press, 1993), 181-82.
(35.) Leeman Perkins has hypothesized (in "Conflicting Attributions," 324) that Nivelle's sixth gathering may constitute a Busnois fascicle where two virelais (or bergerettes) attributed to Busnois--C'est bien malheur, and En tous les lieux (attributed to Busnois--C'est bien malheur, and En tous les lieux (attributed to Busnois in Dijon)--surround other anonymous works.
(36.) Jane Alden has noted (in Makers of a Songbook, 223) that "unknown works appearing alongside such 'classics' as Je ne vis onques or Le souvenir de vous benefited from their proximity to more familiar works. This is supported by the visual dimension: the similarity of artistic decoration on consecutive pages further contributes to the perceived equal status of the works contained."
(37.) My total of five openings counts two openings for Ma plus amee, one for Quant ce vendra, and two for A la longue. I have not included Tout a sa dame in this list since only its cantus was erased--the whole opening did not await replacement by another song.
(38.) A. Piaget, "Simon Greban et Jacques Millet," Romania 22 (1893): 230-43. Fede's name also appears in Cretin's Deploration sur la trepas de Jean Ockeghem (1497) and Eloy d'Amerval's Livre de la deablerie (after 1500), although the appearance of Fede's name on these lists significantly postdates his presumed death (Fallows, "Fede, Johannes").
(39.) Ibid. See also Leeman I., Perkins, "Musical Patronage at the Royal Court of France under Charles VII and Louis XI (1422-83)," Journal of the American Musicological Society 37, no. 3 (Autumn 1984): 507-66.
(40.) Thanks to Jane Alden for sharing her forthecoming study of the Loire Valley chansonniers with me.
(41.) O lumen ecclesie and Magne pater sancte dominice were edited by Masakata Kanazawa, "Polyphonic Music for Vespers in the Fifteenth Century," 2 vols. (Ph.D. diss., Harvard University, 1966), 2:159-64.
(42.) See editions in appendix B.
(43.) Perkins, "Conflicting Attributions," 323.
(44.) Chansonnier Nivelle de la Chaussee, introd. Higgins, fol. li.
(45.) Fallows, "Fede, Johannes."
(46.) David Fallows, "Johannes Ockeghem: The Changing Image, the Songs, and a New Source," Early Music 12, no. 2 (May 1984): 218-30.
(47.) Gerald Montagna, "Caron, Hayne, Compere: A Transmission Reassessment," Early Music History: Studies in Medieval and Early Modern Music 7 (1987): 107-57.
While many scholars still express doubts about whether the owners of chansonniers were musically literate, the idea that chansonniers were read, understood/appreciated, and enjoyed by their owners has steadily been gaining ground. Though the books obviously function as a repository of songs, factors such as their diminutive size and persistent, uncorrected errors argue against chansonniers' use as performing documents. The Chansonnier Nivelle de la Chaussee (Bibliotheque nationale de France, Ms. Res. Vmc MS 57), however, contains many small-scale corrections and crossed-out notes. By analyzing the nature of individual corrections we can learn about the copying process of Nivelle's principal scribe, as well as how chansonniers may have been used and understood by their owners.
The Nivelle Chansonnier also contains several instances of large-scale erasures that appear to have taken place post-binding where entire voice parts were erased. Notably, two of the four erased pieces in Nivelle are attributed to Fede (alias Jean Sohier). Though Fede's career is well documented, his extant works are limited to two Magnificat antiphons from the Ferrarese court choirbook (Biblioteca Estense, Ms. [alpha].X.1.11), and three chansons unique to Nivelle (all of which are incomplete due to large-scale erasures and the loss of Nivelle's penultimate gathering). This article surveys Nivelle's small-scale corrections, as well as offers a hypothesis for Nivelle's large-scale erasures. While David Fallows has suggested that the erasure of the songs by Fede signaled some sort of disgrace for this composer in the eyes of the original owner of the manuscript, this article advances an alternate theory. In addition, it offers an evaluation of Fede's secular compositional style based on reconstructions of his incomplete and erased works.
Debra Nagy is a lecturer in early music at Case Western Reserve University, where she has taught courses in mensural notation and directs the Collegium Musicum. An accomplished performer on historical oboes, she performs with early-music ensembles around the country and as a member of Ciaramella, a group devoted to fifteenth-century music.
|Printer friendly Cite/link Email Feedback|
|Date:||Sep 1, 2009|
|Previous Article:||Music Library Association table of contents June 30, 2008.|
|Next Article:||Written vs. sounding pitch.|