Canadian Journal for Traditional Music (1988)

"M'en, revenant de la Joli'Rochelle"::
A song from c/ 1500 in the current French-Canadian repertoire

Jay Rahn

There is a song that is currently sung in French Canada, the verses of which begin with "M'en revenant de la joli' Rochelle" (see Example 1, following Notes at end of article). Over the years, more than ninety variants of this song have been written down or recorded on cylinders, discs, or tapes in French Canada. A few variants have also been found in the northeastern United States and France.1 One of the Continental French variants (see Ex. 2) is especially interesting. Unlike any of the others that have surfaced over the years, this one does not date from the twentieth century or even the nineteenth; rather, it is found in a manuscript that has been assigned the approximate date 1498~1502.2 Accordingly, one is dealing in this instance with a song tradition that appears to span more than four centuries and an entire ocean. How could this situation have arisen?

In this study, I will argue that the body of French songs among which the Renaissance variant is found is somewhat special. In the cultural context of its own time (ca. 1500),3 this body of songs stands out in contrast with other French songs of the period. Indeed, in certain respects, this special repertoire, which consists of monophonic (i.e., unaccompanied) French songs ca. 1500, has, in certain respects, greater affinities with later French and French-Canadian folk song than it has with other, polyphonic French songs of its own time. Moreover, some of its special features, features that render it comparable with, later French and French-Canadian folk song, appear to have been recognized by French writers around 1500. And there are further stylistic affinities — not directly acknowledged ca. 1500 — between monophonic French songs around 1500 on the one hand and latter-day folk songs of France and French Canada on the other; these affinities help one to understand how the survival of "M'en revenant" might have occurred. In sum, then, I will argue that what we know of French monophonic song ca. 1500 indicates that the repertoire in which the archaic variant of "M'en revenant" appears is (a) culturally and stylistically distinguishable from other French song ca. 1500 and (b) culturally and stylistically continuous with later French and French-Canadian folk song.

The discussion of these points, which takes up the first section of this study, provides a context in which relations between the archaic variant and more recent versions found in French Canada can be understood. Accordingly, in the second principal section I examine relations between the old French variant of "M'en revenant" and its apparent descendants in the New World. Thereupon, I conclude with a few observations on how, in the light of this study, historical folk-song research of this sort might be carried out in the future.

Provenance and Transmission.

The repertoire in which "M'en revenant" is first found consists of about 500 monophonic songs written down between about 1490 and 1520. Of these songs, about 250 are secular and were notated in a form that preserves both their texts and tunes. Approximately 70 others are secular songs that were written down without their tunes (i.e., only their texts were notated).4 And finally, the overwhelming majority of the remaining songs are religious contrafacts, for the most part Christmas carols, the texts of which were notated and provided in many instances with indications of the tunes to which they were to be sung (i.e., by means of a phrase such as "Surle chant(=cantus = melody) de"; i.e., To the tune of' ').5

These monophonic French songs from around 1500 differ from other French songs of the time in several ways. First, and most obviously, they were sung unaccompanied rather than being set in elaborate polyphonic arrangements.6 Secondly, though they were in all likelihood sung in aristocratic circles (like their polyphonic counterparts), they were also performed in open-air public theatres and seem also to have been sung by amateurs in their private devotions as well as by professional street singers (batelleurs).7 Unlike their polyphonic counterparts, they were frequently published — generally with their texts only — in small, inexpensive chapbooks rather than in costly musical prints or manuscripts.8 When two or more copies of such monophonic songs survive, one almost invariably finds differences in the texts and tunes of the variants. These variations are generally much greater than those that are found between variants of French polyphonic songs of the same period and seem to indicate that the transmission of the monophonic songs was largely oral rather than written.9 Finally, the monophonic songs differ from their polyphonic counterparts in that, almost without exception, they are anonymous in origin. 10

Each of these contrasts points to a relatively "popular" provenance for the monophonic repertoire as compared with the corpus of French polyphonic song of the time.11 Such a popular provenance might account for the survival of an individual song (such as "M'en revenant") in French and French-Canadian oral tradition of later centuries. And the songs' apparently popular provenance might also account for the survival of certain stylistic traits across an ocean and over more than four centuries.

Prosodic Features

Monophonic songs from around 1500 also contrast with their courtly, polyphonic counterparts in matters of style, and they do this in ways that are reminiscent of what one finds in modern French and French Canadian oral tradition. A number of these contrasts have to do with prosody. First, one should note that in courtly French poetry ca. 1500, and in polyphonic songs based thereon, assonance, half-rhyme, and non-rhyme are forbidden in theory and avoided in practice. On the other hand, partial rhyme and even non-rhyme are fairly common in French monophonic songs around 1500 and are found in modern French and French-Canadian folk songs with rather great frequency. 12

Second, unlike courtly chanson texts ca. 1500 and like modern French and French-Canadian folk songs, French monophonic songs around 1500 frequently feature a treatment of the unaccented e (as in faire and cela) that is erratic from the point of view of "official" scansion. Often an unaccented e has to be elided where it would be pronounced in "educated" poetry or song, and often it has to be pronounced where it would be elided in "standard" French prosody.13

Third, in courtly song ca. 1500, syllable counts are considered sacred in theory and treated consistently in practice. If, for example, a given line of a given stanza has eight syllables, the corresponding line of other stanzas has eight syllables too. In monophonic song of the same period and in French and French-Canadian folk song of more recent times, syllable counts are frequently found to vary from stanza to stanza; from time to time, the number of syllables in a line increases or decreases at a given point in the rhyme scheme.14

Fourth, in "standard," courtly prosody of the Renaissance, ten-syllable lines are consistently divided by a caesura into four-plus-six syllables, whereas in the texts of a number of monophonic French songs of the period as in some modern French and French-Canadian folk songs, ten-syllable lines are found to be divided not only into four-plus-six but also five-plus-five syllables. The distinguishing feature here is the five-plus-five division. 15

Fifth, in courtly verse ca. 1500, one never finds the laisse form of rhyme scheme. In such a rhyme scheme, there is a long series of verses consisting of (approximately) isosyllabic lines all ending in the same syllable. Such a scheme is fairly common in French monophonic song of the period and in later French oral tradition.16

Sixth, there is a special prosodic procedure that is found in modern French and French-Canadian folk song and in French monophonic song ca. 1500 that never seems to be encountered in "elite" song of the Renaissance. This procedure, which I will refer to as imbrication, consists of repeating the last half of one verse as the first half of the next. 17

Seventh, refrains of French and French-Canadian folk songs and French monophonic songs ca. 1500 are often constructed in a parallel manner, such that the first half of the refrain's text is repeated in a modified way to form the second half. This results in a type of redundancy that in courtly prosody around 1500 was discouraged in theory and avoided in practice. Generally, too, the first and second halves have contrasting endings; one is masculine and the other feminine. 18

These seven features are illustrated in Examples 3 to 5. In order to save space, single examples of French monophonic song ca. 1500, modern French folk song, and modern French-Canadian folk song have been selected, respectively. Each example illustrates some, though not necessarily all, of the seven features listed above. Since these features are stylistically characteristic rather than definitive (see below), a different selection of songs would have provided examples of different combinations of the seven features. For the modern French and French-Canadian examples, I have chosen pieces that are readily available on disc recordings: accordingly, readers can actually hear the assonances, elisions, etc. that are alluded to in the commentary; in written versions of such songs, the presence or absence of these features is sometimes in doubt, either because of carelessness or a desire to "clean up" texts on the part of compilers and editors.

In Example 3, a piece of French monophonic song ca. 1500, one finds the verses arranged in laisse form with the rhyme syllable -on uniting the ends of verse lines. The last half of each verse is repeated as the first half of the next verse in imbricative fashion. Both the refrain and the verse lines are cast in ten-syllable units divided by a caesura into 5 + 5 syllables (f stands for the feminine ending on an unaccented e at the end of the first half of each line). In the first line of the first verse there is an exception to this overall pattern: either the syllable count is 5f + 6 (this is more probable to judge from the text underlay implied by the music in the source), or the e of bonne must be elided to retain the 5f + 5 metre. In many such cases where one is dealing with a written rather than an acoustical source — whatever the century — one cannot decide whether there is a discrepant syllable count or a "non-standard" elision.

In Example 4, the rhyme syllable -a links the verse lines in a laisse form. Assonance (compare and cheval) is found in verses 3 and 4. Between verses 2 and 3, 3 and 4, etc., imbrication is present. The syllable count for individual lines of the verses varies among 5(f) + 6 (where (f) stands for an elided feminine ending in the first and third verse lines), 5(f) + 5 (in the next four lines that are sung as well as the last), 5 + 5 (at the end of verse 4 and the beginning of verse 5), and 5 + 6 (in the remaining lines, including the second). "Non-standard" elisions permeate the text and are marked as follows in the example: _'. Finally, the second half of the refrain constitutes a reworking of the first half in such a way that there is a masculine/feminine contrast between the halves.

In Example 5, there is a similar reworking of the first half of the refrain in its second half and a corresponding masculine/feminine contrast in the endings of the two halves (obscured in this case by the elision of the final e of jolie). The rhymes of the verse lines form a laisse that is based on the vowel e. This overall scheme is interrupted by the appearance of the non-rhyming word enfant in verses 2 and 3. Syllable counts vary among 4 + 4, 4(f) + 4, and 5(f) + 4, though the laisse lines are essentially isosyllabic. "Unconventional" elisions appear throughout and are indicated as in Example 4.

The seven19 features of prosody listed above can be termed "typical accidents." They are virtually never found in courtly song Ca. 1500, but they are found quite often in monophonic songs around 1500 and in modern French and French-Canadian folk songs. One should note that these features are not invariably found in the latter repertoires, but when they appear they are characteristic. Such typical accidents of prosody can be considered to represent an approach to versification that has persisted in French oral tradition over the centuries. I would go so far as to maintain that one can discern around 1500 the first large-scale signs of a rift in the prosodic practice of French song, a rift between courtly versification in the "official" or "standard" style, and the largely anonymous, aurally-based prosody of the monophonic repertoire. (Remember that all the French and French-Canadian songs I have referred to are monophonic.)

Around 1500, this rift appears not to have gone unrecognized; certain prosodic theorists of the time called attention to phenomena that one can observe time and again in the monophonic repertoire but not in courtly productions, and they often drew attention to these phenomena. For example, some courtly prosodic theorists ca. 1500 describe assonance as "rime rurale," and one, Henri de Croy writes as follows: "Laisse les bregiers (sic; recte: bergiers) user de leur rétorique rural" (let shepherds employ their rural prosody). He groups "regime rurale" with other prosodic "vices." These include the so-called "paltry ways of rhyming" (menues tailles) such as rime en goret (lit., pig rhyme),20 which is another form of assonance that the prosodic theorist Pierre Fabri says "only receives approbation among rural and ignorant people who make such song texts to go to the mustard" (n'est approuvé que entre rural et ignorans qui en font les dictz pour aller a la moustarde.)21 Significantly, prosodic theorists ca. 1500 also refer to songs in strophic form as "chansons rurales":22 the formes fixes of courtly song are non-strophic, whereas virtually the entire repertoire of monophonic French song around 1500, like the body of later French-language folk song, is strophic 23

Literary Themes

Although such prosodic evidence is quite striking, one should not lose sight of the fact that there are other stylistic links between French monophonic song ca. 1500 and more recent French and French-Canadian folk song. First, there are the literary themes of the respective repertoires. Apart from love songs — no particular stream of the French song tradition has a thematic monopoly on this type — monophonic songs ca. 1500 are of the following kinds (according to Theodore Gérold):24 chansons narratives (ballads), de mal manes et malmariees (about unhappy marriages), d'aventuriers (about adventurers), politiques (political), historiques (historical), pastorales (about shepherds or shepherdesses), rustiques (about country life), satiriques (satirical), grivoises (licentious), sottes (silly), and bachiques (for drinking).24 One can compare this list with headings found in modern French folk-song collections: complaintes (ballads), de manage (concerning marriage), chevaleresques (chivalric), patriotiques (patriotic), historiques (historical), humoristiques (funny), grivoises (licentious), badines (trifling), bachiques (for drinking), de yin (concerning wine), and épicuréennes (about food).25 And if one turns to modern French-Canadian folk song collections, one finds such headings as the following: contes tragiques (tragic stories), maumaries (concerning unhappy marriages), maria ges tragiques (tragic marriages), la vie du soldat (the soldier's life), aventures galantes (chivalrous adventures), histoire (history), pastourelles (concerning shepherds and shepherdesses), vantardises (boasting), mensonges (falsehoods), grivoiseries (licentious), beuveries (drinking), and ripailles (feasting) .26

An obvious drawback of such headings is that they are somewhat imprecise, and a given song could well fit under more than one heading. Nevertheless, these headings do serve a useful function in that they indicate that much the same subjects and genres have been used in the French monophonic tradition over vast reaches of time and space. Moreover, the very variety of their subject matters serves to distinguish French monophonic songs ca. 1500 from their courtly, polyphonic counterparts27 and serves to link them with the later tradition of French and French-Canadian folk song.

One can attain a more precise view of the subjects of songs by recourse to the sort of topology of themes or "motives" that is used in modern folklore research. For laisse-form songs in French tradition (both in Europe and North America), such a catalogue has been compiled by Conrad Laforte. If one examines Laforte's catalogue from the point of view of the present study, one can discern a very complex situation. Altogether a dozen items of French monophonic song ca. 1500 are listed.28 One can advance few generalizations about the twelve items; each seems to represent a special instance.

The song of "La Pernette" is found in the so-called Bayeux manuscript ca. 1500 where it tells much the same story and uses much the same rhyme syllables and metre as is found in its modern folkioric counterparts, but its rhyme scheme (as opposed to rhymes) and musical form differ substantially from the later versions. The song "Sur le point d'Avignon" as found in Petrucci's Canti C, 1503, has much the same tune as is found in folkioric versions; however, one cannot assess its text, which survives in the Renaissance source as a mere incipit. Two variants of the folk song "Les trois fleurs d'amour" are found in monophonic sources ca. 1500, but their tunes, and more importantly their forms and rhyme schemes differ substantially from each other and from later versions. The story of the folk song "Le message a l'ami" is told in a variant that is found in MS Paris, Bib. nat., f. fr. 12744 but that uses very different rhymes from the folkloric versions. Another song in MS 12744 has been grouped with the folkioric variants of not only "Mon père avait cinq cents moutons" but also "La bergère aux brebiettes." A similar situation holds for the song "Ce sont varletz de Vire" which is found in the Bayeux manuscript and has been grouped with the folkioric variants of both "L'embarquement de la fille du bourgeois" and "Le passage des bois." All in all, then, categorizing songs by motives does not lead to compelling results, at least as far as the relations between French monophonic song ca. 1500 and later French and French-Canadian folk song is concerned. This is not to deny that there might be direct or indirect genetic relations between the earlier repertoire and songs found later in French oral tradition. Rather one must acknowledge that, though the early and more recent songs might be historically related, their resemblance must have become considerably attenuated through the process of oral transmission. However, I do not believe that the student of survivals need despair in this situation. First, the evidence points to precise thematic connections between the Renaissance songs and later folk materials. Secondly, I feel that one can establish fairly strong connections not so much on the basis of individual items, which are subject to considerable change, but on the basis of persistent stylistic features of the repertoires between which otherwise attenuated relationships are evident.

Musical Features

Returning to features of style, one can note that there are a number of musical parallels between monophonic song ca. 1500 and later French and French-Canadian folk song. One of the most important of these parallels has to do with text placement (or underlay), more specifically the musical rhythm of the texts' syllables; this is a feature that links prosody or textual metre with musical metre. The systems of text rhythm that are discernible in French monophonic song ca. 1500 and in French-Canadian folk songs of the tempo giusto (as opposed to parlando rubato) type have been outlined by the present author in a pair of studies.29 Example 6 illustrates the groupings of syllables into "long-short" or "equal-equal" pairs in the middles or "cores" of lines that is characteristic of both systems (note that the first of each pair of syllables is, in musical terms, relatively more accented than the second). (See Ex. 6.)

Further details of the systems of musico-textual rhythm are provided in the earlier studies. Suffice it to remark at this point that there are three points of divergence between the earlier and later rhythmic practices: (1) French monophonic song ca. 1500 presents no consistently special musico-textual treatment of refrain lines as does more recent folk song; (2) in triple (musical) metres, one finds "short-long" pairs,30 whereas only the latter are found in modern giusto folk songs; and (3) there appears to be no record of a parlando nubato performance style during the earlier period. With regard to these divergences, one should note that: (1) prosodic complication in refrains is not a feature of the Renaissance songs whereas it is found fairly frequently in modern folk songs where it seems to give rise to special "rules" of text underlay for the refrains; (2) triple musical metre (including what has been notated as 3-4 or 3-8 time and the so-called "compound" metres of 6-8 and 9-8) is quite rare in French Renaissance monophonic song but quite common in later folk song, though it seems that in many cases Renaissance tunes were "coerced" into 2-4 time (i.e., tempus imperfectum diminutum) which was the favorite metre of the period;3' and (3) no particular system of melo-textual rhythm has as yet been discerned for modern parlando rubato performance in French and French-Canadian folk song, and such a style of performance seems to be difficult to convey in Renaissance notation, which, up to a certain-point, appears to force musical rhythms into a fairly rigid, albeit sometimes syncopated, framework. In short, then, some of these divergences between earlier and later practice might have been a result of stylistic change within the French monophonic song tradition, and some might be mere reflections of change in notational procedures. In this regard, one should note that those who notated the Renaissance songs often seem to have tried to "stretch" the conventions of their notational system a little by implicitly suppressing beats or introducing extra beats within a measure, yielding mixtures of duple and triple metre or what might in modern terms be described as "additive" rhythms.32 Such mixed or additive metres are fairly frequent in modern folk songs33 and are almost never found in Renaissance polyphony (notwithstanding some erroneous analyses of the latter).34 Moreover, additive or mixed metres are bound to arise if a system of underlay such as the ones referred to above is operant in a song.

Finally, with regard to musical style, one can note the use of modes in the various repertoires. Information on modal usage in modern folk song is scanty but what there is indicates the primacy of so-called lonian, Mixolydian, Dorian, and Aeolian as well as mixtures of these.35 The same preference is found in French monophonic song ca. 1500.36 In both the early and late repertoires, Lydian is almost never found and Phrygian is even less frequent. Use of Lydian and Phrygian seems to have been more typical of the more "elite" forms of music around 1500.37

The principal tonal divergence between the Renaissance and modern repertoires consists in the far greater proportion of Ionian (i.e., "major") melodies in the later repertoire and the appearance from time to time of modern minor. This would seem to be a reflection of stylistic change during the intervening musical era of so-called Common Practice during which major tonalities seem to have predominated not only in so-called "art" music but also (and perhaps especially) in more "popular" forms throughout the West.38

In sum, then, there would appear to be some fairly strong stylistic connections between French monophonic song ca. 1500 and later French and French-Canadian folk song. There are stylistic parallels between the Renaissance and modern repertoires of French monophony with regard to prosody, subject matter, and music. Interestingly, many of these common features are found in so-called "popular" song ca. 1500 but not, or much less frequently, in more elite forms of the period. In some cases, these distinguishing characteristics were recognized as such and even as worthy of censure by elite commentators ca. 1500. And such stylistic disjunctions around 1500 seem to have coincided with socio-cultural differences and indicate a large-scale rift in what might be termed French "song-culture." All the same one cannot ignore the fact that there are also some systematic differences between French monophonic song ca. 1500 and later French and French-Canadian folk song. Some of these appear to represent stylistic change. In future historical studies, one would hope to determine, for example, when major tonality became dominant, when complicated versification for refrains came into vogue, and when "short-long" syllable pairs dropped out of use in triple-time tempo giusto melodies. Other discrepancies between early and late monophony might be more apparent than real. Performance practices that might have existed in the Renaissance might have gone unrecorded because of the notational conventions of the time. Renaissance temporal notation discourages the rendering of elaborate parlando nubato rhythms; early pitch notation provides no means of indicating the micro-tonal inflections that are sometimes found in later folk music and leaves certain intervals open to interpretation because of the unwritten conventions of musicaficta.

It seems desirable to keep all of these observations in mind when assessing individual instances of apparent survival. To illustrate this point, I will consider, by way of summary, the early and modern versions of "M'en revenant" mentioned at the outset of this study (Examples 1 and 2).

M'En Revenant de la Joli' Rochelle

As mentioned above, more than ninety variants of "M'en revenant" have been recorded in modern times. Space does not permit a detailed consideration of all of these, though such a study would be of interest because of the piece's historical importance.39 Fortunately, the modern variants present a sufficiently unified picture to allow generalization, and, for the purposes of the present discussion, the version presented in Example 1 can be considered representative of the modern forms that the song has taken.

Both the modern and Renaissance variants are strophic. Both have a laisse construction consisting of isosyllabic lines of 4 + 6f syllables and both make use of imbrication. Both feature assonances in "è-(e)" where the "è" vowel is followed by rand land even

In the modern variants one finds a considerable number of refrains in use. Thus, it is not surprising that none of these corresponds to the refrain of the Renaissance variant, since a great variety of refrains might have been introduced and coexisted over time. In both early and modern variants, nevertheless, one finds parallel construction in the refrains:

Faisons bonne chère (feminine)
Faisons-la, faison(masculine)

and

C'est l'aviron qui nous men', qui nous mont' (feminine)
C'est l'aviron qui nous monte en haut (masculine)

In the refrain of the Renaissance variant, one also finds the 5f+ 5 syllable pattern characteristic of later French folk song. In the modern variants, the refrain first appears at the end of the first verse; in the Renaissance variant it appears at the outset. This seems to represent a stylistic change, for the laisse-form songs of the Renaissance uniformly present their refrains at the beginning in contrast with laisses found in later folk song.41

Since the texts of Examples 1 and 2 are taken from written (rather than acoustically recorded) variants, it is difficult to assess the appearances of elisions and irregular syllable counts. However, in the French-Canadian variant, e's are clearly elided in an "unofficial" way in the words "joli'" and "demoisell's" of the first verse and "mont' "of the refrain, and the appearance of e in "jolies" in verse 2 might represent an extra syllable. In the Renaissance variant, "monte' " in verses 3 and 4 and "ell' "in verses 5 and 6 seem to represent an "unofficial" elision; if the e's were not elided, they would provide instances of extra syllables.42

Both variants could be classified thematically as "erotic" or "chivalrous adventures." Thematically, they are very similar up to verse 4, that is, throughout their first halves. The male singer says that he was returning from La Rochelle, met three girls, chose the fairest, and raised her onto his saddle. At this point, the narratives diverge. Up to this point, however, the words chosen to convey the opening motif just described are very similar. One should note as well that in both repertoires often only part of one variant's theme is found in the text of another variant.43

The modern variants employ a number of different tunes; it is, thus, not surprising that none of these corresponds very closely to that of the Renaissance variant. One should note also that in the rare instances where there is more than one musical source for a monophonic French song ca. 1500, one finds that there is only one tune, albeit in variant forms, associated with a given text. However, this might be merely a reflection of the relatively similar provenance of French monophonic song sources during this period. The modern tunes all appear to be in giusto rhythm (this is sometimes difficult to determine in the case of written variants for which no corresponding recording survives). The modern tunes are Ionian (or major), Dorian, and modern minor in tonality; the Renaissance variant is clearly Ionian (or major). Finally, both the modern and Renaissance variants manifest those aspects that are shared by the two text underlay systems mentioned above, save that the "irregular" prosody sometimes found in the modern refrains is reflected in correspondingly "irregular" rhythmic patterns typical of the more recent songs. In sum, the stylistic evidence appears to support the hypothesis that there is a genetic connection between the Renaissance and modern variants despite the differences — possibly reflective of stylistic change — to be expected between items separated by four centuries of oral tradition.

Conclusion

In the future, one might study other such instances of survival. Prime candidates for this sort of study include early and modern songs that are similar thematically. In this regard, the ongoing work of Laforte's Catalogue is of great utility.44 Also worthy of investigation are stylistic survivals. Here one would hope to isolate songs which, though their literary themes seem not to have survived, nevertheless contain a high concentration of stylistic characteristics typical of modern folk songs. It is hoped that the present study has provided some initial leads in this respect.45 Furthermore, one would be interested in whether tunes or "tune-types" have survived. In this regard, there would appear to be much work to be done, for a detailed classification of melodies has yet to be arrived at for tunes in either early sources of French monophony or more recent collections.46 In all such studies, one would hope that as much detail as possible concerning the stylistic and cultural contexts of the items compared could be brought to bear on the questions that arise in each case. One hopes again that the present study might serve as a guide in this regard.

Finally, it seems worthwhile to consider what further value might arise from such studies of survival. According to a current school of ethnomusicological thought, such studies, based as they are on comparison of items from diverse contexts, are not meaningful.47 However, I feel such studies are of considerable value. They can serve to point up continuities and changes within a tradition. And perhaps more importantly, framing such comparisons draws one's attention to aspects of the different repertoires that might otherwise have been overlooked: in other words, in the process of studying survivals one finds that the items that are compared with each other turn out to cast light on each other.

York University
North York, Ontario

Notes

*This is a revised version of a talk originally given at the Graduate Colloquium, Music Department, York University in 1984. In the meantime, many scholars have offered advice and comments on the paper's written form. These include especially Rika Maniates, Tim McGee, and John McLelland of the University of Toronto, Leeman Perkins of Columbia University, Adrienne Fried Block of CUNY, Staten Island, Andrew Porter of UCLA, and Donald Deschênes of Université Laval. Any flaws that remain are my own fault.

1. See Conrad Laforte, Le Catalogue de la Chanson Folklorique Fran çais, vol. 1 (Chansons en Laisse) in Les Archives de Folklore, 18(1977), no. K-4, 352-56.

2. The source is MS Paris, Bibliothèque nationale, fonds français 12744. On the dating of this manuscript, see Jay Rahn, Melodic and Textual Types in French Monophonic Song, ca. 1500 (Ph.D. dissertation, Columbia University, 1978), rpt. Ann Arbor, University Microfilms, 2 vols., 64.

3. For the purposes of this study, "ca. 1500" refers to the years 1490-1520, that is, the period immediately surrounding the time when MS 12744 appears to have been compiled.

4. The secular songs with tunes appear in a) MS 12744 (see note 2, above; the texts and melodies are edited in Gaston Paris and Auguste Gevaert, eds., Chansons du XV3 Siècle, Paris: Firmin-Didot et cie, 1875, rpt. 1935; the melodies are re-edited in Rahn, 414-566) and b) MS Paris, Bib. nat., f. fr. 9346 (the texts and melodies are edited in Theodore Gérold, ed., Le Manuscrit de Bayeux, Strasbourg, Commission des Publications de la Faculté des Lettres, 1921; the melodies are re-edited in Rahn, 567-696). Secular songs lacking tunes appear in seven sources of the time; these are edited in Brian Jeffery, ed., Chanson Verse of the Early Renaissance, vol. 1, London: Tecla, 1971, 39-183.

5. About a dozen sources of religious French monophony survive from around 1500. These are listed in Rahn, 369, and they are described and edited in a) ibid., 44-60 and 701-1045; b) Henry Chardon, ed., Noels de 1512 de François Briand, Paris: H. Champion, 1904; and c) LaBorderie, ed., Oeuvres Fran çaises d'Olivier Maillard, Paris, 877. On the tradition of monophonic religious song, see Adrienne Fried Block, Pierre Sergent's "Les Grans Nolz", ca 1537, and the Early French Parody Noel: History and Analysis (Ph.D. dissertation, City Univ. of New York, 1979) rpt. Ann Arbor, Univ. Microfilms, 79-13 115, 2 vols. Most frequently the phrase "Surle chant de" appears in the abbreviated form "Sur."

6. One should distinguish at this point three types of French song ca. 1500: songs, mostly rondeaux, that exist almost entirely in the courtly, polyphonic tradition; other songs, mostly in non-rondeau forms, that exist almost entirely in the "popular" (see below), monophonic tradition; and courtly polyphonic arrangements of the latter. The contrasts drawn in this section are between the first two categories, and thus do not involve, save indirectly, the third, mixed genre. That the monophonic songs existed independently of polyphonic arrangements of them that were made for elite consumption is attested to by Gustave Reese and Theodore Karp, "Monophony in a Group of Renaissance Chansonfliers," Journal ofthe American Musicological Society, 5 (1952): 4-15, where it is con-elusively shown that the monophonic tunes were not extracted from polyphonic originals.

7. On aristocratic use of the monophonic repertoire, both sacred and secular, see Rahn, 46-53 and 61-69. On the use of secular monophonic songs in farces, sotties, and moralities of the time, see Howard Mayer Brown, Music in the French Secular Theater 1400-1550, Cambridge: Harvard Univ. Press, 1963. On the use of monophony in devotions, see Rahn, 53-60 and especially Block, passim. On the batelleurs, see Rahn, 34-41. These street singers appear to have been the descendants of the medieval jongleurs (see Edmond Faral, Les Jongleurs en France au Moyen Age, Paris: Champion, 1910, rpt. 1964) and the predecessors of the later chanteurs publics (see Patrice Coirault, Formation de Nos Chansons Folkioriques, Paris: Scarabée, 1953, 3 vols, esp. vol. 1).

8. See Jeffery, 14-19 for a general description of the chapbooks. The few manuscripts containing monophonic songs are exceptional for the time, albeit important for a study of the genre.

9. An annotated list of variants of the monophonic songs appears in Rahn, 368-412. Some specific examples of divergence among monophonic variants are discussed in ibid., 67-69.

10. Unlike the case with French polyphonic song of the time, the composers of the music of French monophonic songs Ca. 1500 are entirely unknown. The authors of a few of the texts are unknown. Perhaps "author" is too strong a word in this case, for in almost every instance where an author is known, the process of composition consists merely of contrafactum, that is, writing a parody of a pre-existent song, and/or the author is not a professional poet (cf. note 5, above, for the texts by the schoolmaster François Briand, the theorist and publisher Guillaume Guerson, and the preachers Frère Jehan Tisserant and Olivier Maillard). Symptomatically, one of the few original texts of which the author can be traced appears in a monophonic source without an attribution (cf. Jeffery, 74-76).

11. Since Howard Mayer Brown's "The Chanson Rustique: Popular Elements in the 15th-and 16th-Century Chanson," Journal of the American Musicological Society 12 (1959): 16-26, the scholarly consensus has been that the monophonic repertoire consists of "popular" songs.

12. Cf. the accounts of assonance by prosodic theorists of the period in Ernest Langlois, ed., Recueil d'Arts de Seconde Rhetorique, Paris, Imprimerie Nationale, 1902, passim, e.g., 249-51 and 314-16, and Pierre Fabri's Le Grant et Vrai Art de Pleine Rhetorique (ed. Alexandre Heron), Rouen, 1889-90, vol. 2, 27. For examples of courtly verse of the period, see Marcel Françon, Poèmes de Transition (XVe-XVIe Siècles): Rondeaux du MS 402 de Lille, Cambridge: Harvard Univ. Press, 1938. General observations on the works of courtly versifiers of the time appear in Henry Guy, Histoire de la Poésie Francaise au XVIe Siècle, Paris: Honoré Champion, rpt. 1968, vol. 1, "L'école des Rhétoriqueurs." These can serve as illustrative material for the remarks on courtly prosody below. With regard to assonance, one can consult as well Jeannine Alton and Brian Jeffery, Bele Buche e Bele Parleure: A Guide to the Pronunciation of Medieval and Renaissance French for Singers and Others, London: Tecla, © 1976, for a clear introduction to French pronunciation of the time. Examples of assonance, half-rhyme, and non-rhyme can be found in MS 12744, nos. 1, 2, 4, 5, 7, 11, 14, 15, 16, 20, etc. and in modern field collections of French and French-Canadian folk song. Such partial or non-rhymes are often "cleaned up" in modern editions, especially in the so-called "textes critiques" which almost invariably provide a "normalized" version.

13. Prosodic theorists of the time directed their remarks to courtly versifiers and were very fussy about distinguishing masculine from feminine lines and pronouncing or eliding the unaccented e in the proper manner. On this point, see the treatises cited in note 12, above, passim. In written versions not only of French monophonic song ca. 1500 but also of French and French Canadian folk song, it is often not clear where "non-standard" elisions and pronunciations were made. Confusion arises in many cases as to whether a given line might be pronounced in (a) the "normal" way and have "too many" or "too few" syllables, or (b) a "non-standard" way and have the "right" number of syllables. Such situations are further obscured by modern editions and textes critiques.

14. See notes 12 and 13, above. "Discrepant" syllable counts are found in MS 12744, nos. 1, 2, 3, 5,6,7,8, 9, 10, 11, 12, 13, etc. In some of these instances, a "correct" count might result from a "non-standard" pronunciation, or elision, of the unaccented e

15. The 5 + 5 division is found, for example, in MS 12744, nos. 5, 21, 22, 104, 129, and 142 (in all the refrains and some of the verses) and in nos. A-I 1, C-22, C-25, C-27, D-4, E-6, etc. of Laforte. Though syllable counts might vary in these locations, 5 + 5 represents the most frequent count at a given location in the prosodic form in each case.

16. See al/the songs in Laforte, and MS 12744 nos. 21, 39, 71, 78, 81, 88, 97, 104, 117, 126, 142 and 143. No. 53, which never settles into a single rhyme, and no. 103 which has some verses with three (rather than two) lines, might be considered laisses as well. In MS 9346, one might also cite no. 82. No. 86 might also be a laisse but only one verse is preserved. No. 46 might be a laisse but the tierce which introduces its refrain, varies from verse to verse suggesting a virelai form. Nos. 17 and 45 have a /aisse-like structure but have true rhymes at their caesurae. No. 16 has a laisse-like rhyme-scheme but there are six (!) rather than one or two isosyllabic rhyming lines in its verses. And no. 15 seems somewhat suspect from the perspective of traditional laisse structure because all its rhymes involve puns on the syllable "point." See also Jeffery Chanson Verse, 90(a), no. 29 (= 90(b), no. 30); 12, nos. 2 and 6; MS 2368, nos. 37, 47 and 49, and 50 (the latter three are contrafacts of"M'en revenant"); and MS 3653, no.48.

17. See many of the variants listed in Laforte, and nos. 21, 53 (see note 16, above), 78, 81, 88, 104 and 142 in MS 12744. As well no. 97 is probably imbricative. In the case of written, as opposed to acoustically recorded, versions, it is often not clear whether imbrication would be present in performance; there is an obvious economy in notation if one does not write down the repetitions, and the scribe of MS 12744 seems to have arrived at a number of solutions to this problem. In MS 9346, no. 82 features imbrication. Jeffery Chanson Verse, 90(a), no. 29 and 12, no. 2 are imbricative as is MS 2368, no. 37.

18. Langlois, 249-51; cf. also 2 14-16.

19. One might also add an eighth, namely, inexact repetition, as represented by the last line of verse 2 and the final of verse 3 in Example 5, and a ninth, inconsistent structure as represented by the absence of imbrication in the first two verses of Example 4. However, in the case of written texts, it is difficult to determine whether such features (which are found often in MS 12744, for example) are truly representative of a loose, orally-based prosody, or slips of the pen on the part of those who recorded them.

20. Langlois, 249-51.

21. Op. cit., vol. 2, p. 27. Some idea of the implications of "going to the mustard" can be obtained from the following passages from the fifteenth century:

a) "Little children sang in the evening while going to the wine or to the mustard all together:

Your ... has a cough, gossip,
Your ... has a cough, a cough.

Journal d'un Bourgeois de Paris, ed. Tuetey, 1414, 49, "Chantoient les petiz enfans, au soir, en allant au yin ou a Ia moustarde, tous communement:

Vostre con a la toux, commere,
Vostre con ala toux, la toux,"

cited in Louis Thuasne, ed., François Villon: Oeuvres, Paris: August Picard, 1923, vol. 3, 499-501. (Note the parallelism of the first and second halves and the feminine/masculine alternation in what appears to be a refrain);

b) "Item on Jacquet Cardon — nothing
Since I have nothing decent for him.
(Mind you I won't abandon him),
Except perhaps this little song,
If it could have the tune of 'Marionette'
Composed for Marion la Peautarde
Or that of 'Open your door, Guillemette,'
It might then serve for getting mustard."
François Villon, La Testament, ed. and trans., Anthony Bonner, New York, David<
McKay, 1960, 11, 1776-83:

"Item, riens àiacquet Cardon,
Carje n'ay riens pour luy d'honneste.
(Non pas que le gette habandon),
Si non ceste bergeronnette
S'elle eust le chant 'Marionette'
Fait pour Marion la Peautarde
Ou d' 'Ouvrez vostre huys, Guillemette,'
Elle allast bien a la moustarde"

(note that the bergerette is a specifically monophonic form ca. 1500: see on this point, Jay

Rahn,' " 'Fixed' and 'Free' Forms in French Monophonic Song, ca. 1480-1520," in Mary

Beth Winn, ed. , Musique Naturelle et Musique Artificielle: In Memoriam Gustave Reese (Le Moyen Français, vol. 5), Montreal: Ceres, 1980, 130-58, esp. 147-149.

c) "Children who go to the mustard
Sing of you (a prostitute) at the crossroads."

M. Schwob, Le Parnasse, no. XXV, p. 81, of a prostitute:

"Enfans qui vont ala moustarde
Chantent de vous aux carrefours."

22. See Langlois, 321: "Another scheme for double rondeaux which are called simple virelais because laymen put them into their rural songs such as 'Gente de Corps' "(Autre taille de rondeaux doubles qui se nomment simple virlais pour cc que les gens lais les mettent en leurs chansons rurales comme "Gente de Corps"). De Croy's reference might be to the song "Gente de Corps" which appears in MS 9346. This passage only makes complete sense if chanson rurale denotes a strophic song.

23. The virelai and rondeau as they appear in courtly song are essentially only one "strophe" long (i.e., non-strophic): the "proper" form of the ballade, a rarity at this time in courtly song, consists of three strophes plus an envoy and hence is not strictly speaking strophic either. Cf. Rahn," 'Fixed' and 'Free' Forms," passim.

24. Gérold, 128.

25. See the headings in Dumersan and Noel Ségur, comps., Chansons Nationales et Populaires de France, Paris: Garnier, 1866, 2 vols, and Paul Arma, comp., The Faber Book of French Folk Songs, London: Faber, © 1972.

26. See the headings in Marguerite and Raoul d'Harcourt, Chansons Folkloriques Fran çaises au Canada, Quebec: Presses Universitaires Lava!, 1956.

27. Cf. Brown, "The Chanson Rustique," 19.

28. See Laforte, B-3, D-2, E-2, F-4, G-2, 1-4, J-2, J-8, K-6, K-7, L-13, N-10. Though Laforte says that a variant of the song he labels L-2 is found in S'ensuy vent dix-sept chansons (15 15-1530), which is edited in Jeffery, Chanson Verse, I have not located it here.

29. See Jay Rahn, "Text Underlay in Gagnon's Collections of French Canadian Folk Songs," Canadian Folk Music Journal, 4(1976): 3-14, and "Text Underlay in French Monophonic Song, ca. 1500," Current Musicology, 24 (1977): 63-79. The system described in the former appears to apply quite well to tempo giusto melodies in Continental French folk song as well as those found in Canada.

30. Clear examples of "short-long" pairs can be observed in MS 12744, nos. 30, 78, 89, and 136, and in MS 9346, nos. 90 and 93.

31. See d'Harcourt, 49, on the frequency of triple metres in a sample of French-Canadian folk song, and Rahn, Melodic and Textual Types, 123-28 on triple passages notated in a duple manner in French monophonic song ca. 1500.

32. See Rahn, Melodic and Textual Types, 115-22 on "extra" and "missing" beats.

33. See d'Harcourt, 49 for a summary of mixed and additive metres in a sample of French Canadian folk song.

34. See Edward Lowinsky, "Early Scores in Manuscript," Journal of the American Musicological Society, 13 (1960): 126-71, esp. 156-69 on the metrical framework of Renaissance polyphony.

35. See d'Harcourt, 42, George Proctor, "Musical Styles of Gaspé Songs," National Museum of Canada Bulletin, 190 (1960); Part II, 209-12, esp. 210, and Lorraine Thibault, "The Complainte in French Canadian Folk Music," unpub. Master's Thesis, Univ. of Washington, Seattle, 1968, 22, for tabulations of modes found in samples of French-Canadian folk song. Though there are sampling difficulties in these studies, their unanimity cannot be overlooked.

36. See Rahn, Melodic and Textual Types, 159-65 on the modes used in French monophonic song ca. 1500. Note that the rules of musica ficta generally render the songs notated in Lydian as Ionian in tonality.

37. On the frequency of various modes in polyphony ca. 1500 see Rahn, Melodic and Textual Types, 186-89.

38. For speculations on the intrusion of major-minor tonality into folk music see Walter Wiora, European Folk Song: Common Forms in Characteristic Modifications, Cologne: Arno Volk Velag, © 1966 (Anthology of Music, no. 4), 6, 7, and 8.

39. Models for such a study include Marius Barbeau, "Trois Beaux Canards (92 Versions Canadiennes)," Les Archives de Folklore, 2 (1947): 97-138, and Marguerite Béclardd'Harcourt, "Analyse des Versions Musicales des 'Trois Beaux Canards'," Les Archives de Folklore, 4(1949): 129-36.

40. Note that "boire" in verses 7 and 8 of Example 1 is pronounced in such a way as to rhyme with "verre" in French Canada, as it was in France ca. 1500. Cf. Jeffery and Alton, Bele Buche, 34.

41. Thus the laisses that are mentioned in note 16 above and that have a refrain (i.e., MS 12744, nos. 21, 71, 78, 81, 97, 103, 104, and 142) present the refrain consistently at the beginning.

42. Another feature that might be considered is the use of nonsense syllables as is found in MS 12744, no. 104. This feature is like the eighth and ninth described in note 19, above, for though they are characteristic of French monophonic song ca. 1500 and later folk song, they are not germane to the examples of "M'en revenant" that are considered here. Nonsense syllables would be condemned as "sans raison" by elite prosodists ca. 1500.

43. Cf., for example, the annotations in Rahn, Melodic and Textual Types, 368-412, passim.

44. See the songs cited in note 28 above as well as numbers C-3, C-il, and C-Si in Conrad Laforte, Le Catalogue de la Chanson Folklorique Fran çaise, II: Chansons Strophiques in Les Archives de Folklore, 20 (1981); "Nous Sommes de l'Ordre de Saint Babouin," "Sus les Pons de Lyesse", and "Je My Levay Par Ung Matin" discussed in Georges Delarue's review of Jeffery's Chanson Verse in Le Monde Alpin et Rhodanien, 2, (1973): . 1-65; and the noel "Chanton, Je Vous Emprie" which is discussed in Jan Reiner Hendrik de Smidt, Les Noëls et la Tradition Populaire, Amsterdam: H.J. Paris, 1932, 120-24. (Note, however, that no version of "Chanton Je Vous Emprie" appears in the Livret de Noëls as de Smidt asserts on p. 120.)

45. See the songs cited above, especially in notes 15 and 16.

46. Models for such a study might be the work of BartOk on East European folk song.

47. Alan P. Merriam, "On Objections to Comparison in Ethnomusicology," in Robert Falck and Tim Rice, eds., Cross-cultural Perspectives on Music, Toronto: Univ. of Toronto Press, 1982, 174-89, surveys arguments both for and against comparison.

Résumé: Jay Rahn présente une étude détaillée de la chansonfrancaise M'en revenant de la joli' Rochelle, chanson qui date du debut du seixième siècle et dont maintes versions ont été notées en France et au Canada. Ii discute les differences entre des chansons monophoniques et polyphoniques, et illustre comment les caractéristiques des folkloriques sont retenues a travers le temps.