ORB Masthead with site navigation toolbar; see bottom 
of page for text version of toolbar

Encyclopedia | Library | Reference | Teaching | General | Links | About ORB | HOME


Songs in Fixed Forms

by Margaret P. Hasselman

(Last updated on 26 November 2001)


Fourteenth century France saw the development of several well-defined song structures. In contrast to the earlier troubadours and trouveres, the 14th-century songwriters established standardized patterns drawn from dance forms. These patterns then set up definite expectations in the listeners. The three forms which became standard, which are known today by the French term "formes fixes" (fixed forms), were the virelai, ballade and rondeau, although those terms were rarely used in that sense before the middle of the 14th century. (An older fixed form, the lai, was used in the Roman de Fauvel (c. 1316), and during the rest of the century primarily by Guillaume de Machaut.)

All three forms make use of certain basic structural principles: repetition and contrast of music; correspondence of music with poetic form (syllable count and rhyme); couplets, in which two similar phrases or sections end differently, with the second ending more final or "closed" than the first; and refrains, where repetition of both words and music create an emphatic reference point.



The three structures can be summarized using the conventional letters of the alphabet for repeated sections. Upper-case letters indicate that both text and music are identical. Lower-case letters indicate that a section of music is repeated with different words, which necessarily follow the same poetic form and rhyme-scheme.

1. Virelai

The virelai consists of a refrain; a contrasting verse section, beginning with a couplet (two halves with open and closed endings), and continuing with a section which uses the music and the poetic form of the refrain; and finally a reiteration of the refrain. There may be up to three verse sections; Machaut usually uses three, but many of the repertory manuscripts only include 1 or 2. The virelai begins and ends with its refrain. In short,

Virelai form: A bb' a A bb' a A bb' a A

2. Ballade

The ballade may be the oldest of the fixed forms, as something resembling it appears often in the troubadour repertoire. In the 14th century, and especially in the hands of Machaut, its tonal structure was clarified by the use of open and closed endings. The refrain comes at the end of each stanza, and may be set off by a dramatic rhythmic or tonal gesture. There are commonly three stanzas:

Ballade form: aa' b C aa' b C aa' b C

3. Rondeau

The rondeau is at once the smallest and the most intricate of the three forms, as all the complex formal procedures take place within a single stanza. While each section within the virelai and ballade consists of several lines of poetry and music, in the rondeau a section may contain only a single line (though more are possible, especially in the later part of the century.) Since the refrain contains two parts, and only the first part is repeated internally, the poetic and musical effect may be playful, ironic, or otherwise expressive.

Rondeau form (single stanza): A B a A a b A B

Like the virelai, the rondeau begins and ends with its refrain.

Historical context

The three forms were not clearly distinguished at the beginning of the fourteenth century. The manuscript Douce 308, copied early in the century, has a category called "ballettes", some of which look like ballades and some like virelais. (There is even one which can be regarded as either virelai or rondeau.) So, although the three forms are readily distinguishable in hindsight, we may well ask when the poet-composers themselves began to think in terms of separate genres. Manuscript fragments and insertions dating from the middle of the fourteenth century seem to select songs by genre. Thus, Ivrea contains only rondeax and virelais among its chansons, but does not label them; and one of the three distinct manuscript fragments in Cambrai B, another mid-century source, contains all three forms without any indication of genre labels. But in each case, scribal practice (such as the use of abbreviations for the refrains) indicates that the forms were well understood.

Philippe de Vitry (1291-1361) was credited with having contributed to the development of the fixed forms ("…[il] trouva la maniere des motes et des balades et des lais et des simples rondeaux" […[he] discovered the style of motets, ballades, lais and simple rondeaux]), and yet we have no songs that can clearly be attributed to him. Rather, the earliest examples of ballades, rondeaux and virelais are actually transmitted anonymously in Ivrea and Cambrai B. The first exemplar of Guillaume de Machaut's collected works ("Ms. C", Paris Bib. Nat. 1586) dates from around the same period (1350-56.)

Clearly, by the time of Machaut's first collected works the three genres were well established in his mind. He grouped and labeled his songs as ballade, rondeau or chanson baladee (his preferred term for the virelai.) A little later, the surviving index to the lost manuscript Tremoille, dated 1376, specifies ballades and rondeaux, but does not sort them by genre.

Where, then, did the songs in the three fixed forms characteristic of the French 14th century come from? They had at least a double ancestry: the traditional chanson courtois of the trouvères, and the largely unwritten dance tradition. In addition, a third stream came from the motet and chace[1] of the late thirteenth century.

The trouvere tradition of fin'amor, or "courtly love," still supplied the content for the majority of poetry for the 14thc chanson. Thus, the lover is depicted as suffering and ill, perhaps crazy ("De ce que fol pense"); he venerates the lady in almost religious terms ("Rose sans per"); and he laments her cruel disregard, which is often due to the intervention of malignant bystanders ("Besier e acoler"). [2] Machaut's poetry exemplifies and fully develops all these themes.

Structurally, however, the older songs tend to follow their own individualistic patterns, rather than a recognized and predictable repeat-pattern. Antecedent and consequent phrases, with open and closed endings, appear in a variety of contexts. A common structural pattern which opens with an antecedent-consequent pair suggests the later ballade.

Though dances were popular all through the Middle Ages, they were usually improvised by instrumentalists, who were not trained in the literate courtly or churchly traditions, and may not have been able (or cared) to read music. A few instrumental dances do survive from the 13th century.[3] There were also songs for dancing, some of which had refrains and were called ballette or rondel. Unfortunately, most of the music for these dance-songs has been lost except for the refrains, which were often quoted in literary works or in motets.

Some experiments with dance-songs using polyphonic texture took place in the late 13th century: the songs of Adam de la Halle, for instance, which are in simple note-against-note style and are notated in score; and one three-part, homophonic rondeau [4] by Jehan de Lescurel.

The polyphonic texture of the mid-14th century songs, however, owed more to the complex motet and chace than to Adam and Jehan. Many songs have multiple texts, like the motet; the parts are notated separately, not in score; and non-texted parts are labeled Triplum, Tenor, or Contratenor , as in the motet. Imitative texture is often used, as in the chace. Also like the chace is the popular technique of mimesis, or imitating sounds such as bird-calls or musical instruments, which was often used in virelais ("Or sus vous dormes trop," from Ivrea; Pykini's "Or tost aeuz vous assembles," from Cambrai B). Machaut used similar techniques in his songs. His concept of mimesis, however, was extremely subtle: for example, in "Ma fin est mon commencement" (Rondeau 14) he develops a theological concept along with literal self-reference and a clever trick of notation.[5]

The song-writers of the twelfth and thirteenth centuries put a high value on innovation in poetic form, i.e. rhyme scheme and versification. It is interesting to speculate on why the fourteenth century saw a change, to put greater value on work within a set form. Did the political instability of fourteenth-century Europe lead courtly artists and audiences to prefer the fixed-form convention? Whatever the reason, a conventional form creates a different kind of tension within the work, as expectations are delayed or fulfilled.

In music, especially, tonal structures can be expanded in such a context. When the audience has become accustomed to a certain type of cadential progression, say for example a sixth on D followed by an octave on C, then a composer can set up a very effective "open ending" by pausing on D, the penultimate sonority, knowing that the audience will expect the final "closed" resolution on C. Machaut was particularly skilled at exploiting this kind of tension, often pausing on the penultimate sonority before proceeding to the refrain of a ballade, as in #31, De toutes flours. An analogous technique is used in the period structure of the Classical Period, where the dominant-to-tonic (V-I) cadence is so strongly expected that a half-cadence on V is immediately recognized for what it is.

Character and provenance; selected examples

While all three fixed forms were used for elaborate poetry on themes of courtly love, certain special styles were sometimes associated with each one. The virelai was more likely to resemble the motet and chace. It sometimes had two or more texts sung at the same time, like the motet. Like the chace, it sometimes used onomatopoetic bird calls, with a hocket-like texture, or with snatches of musical imitation. Even in the context of courtly love, it was likely to have a light-hearted character, using many short lines and frequent rhymes. Polytextual and imitative virelais are especially prevalent in the North French manuscript Cambrai 1328, section B2. The Italian compiler of Codex Reina was also fond of this genre. One of the best examples is Pykini's Or tost aeux (published as #86 in Apel, 1970.)

The ballade, on the other hand, tended to be more elegant and leisurely. It was the closest in form to the earlier chanson courtois. Though sometimes polytextual, ballades tended to be formal and elaborate, like Machaut's Ballade 31, De toutes flours. Towards the end of the century some composers gave the ballade some of the features of the grand dedicatory motet, as in Philipoctus de Caserta's Par les bons Gedeons (#82 in Apel, 1970.)

The poetic form of the rondeau was the most compact of the three. It was equally ingenious, however, as the refrain had to make sense in three different contexts., i.e. as a beginning, middle and end. Rondeau texts often have a sense of play and delight in puzzle-solving, which occasionally is expressed in the music by a puzzle notation. Machaut's Rondeau 14, Ma fin est mon commencement, is a famous example, with a double sense to the text: it is at once both a theological statement and a self-referential instruction for performance:

My end is my beginning,
And my beginning, my end
And tenor, truly;
My end is my beginning.
My third [part] sings only three times,
Reverses itself, and so end;
My end is my beginning,
And my beginning, my end.

The tenor is notated with the text, and when read backward and upside-down, it gives the cantus. At the same time, only half the contratenor part is given; the second half is derived by reading it backward. As a result, the complete rondeau has a second half which is the exact reverse of the first.

There is also the less polite Il vient bien sans apeler in the Codex Reina (Paris, n.a.fr. 6771). In this piece, like Machaut's, the second half is the exact retrograde of the first, but only half of each part is written. Unlike Machaut's text, this one has a sexual meaning.[6]

But Machaut and others also wrote straightforward and elegant rondeau texts, as in the beautifully expressive Machaut Rondeau 10, Rose, liz, printemps, verdure, which can be found in modern edition in several standard textbooks including the Norton Anthology of Western Music, vol. 1, ed. Claude Palisca (New York: W.W. Norton, 2001).

Examples of all three forms, as well as a few others, were included by Machaut in his Remede de Fortune. This long narrative poem explores the psychology of love with great sensitivity, and (not incidentally) links successful musical/ poetic composition with success in love. In the course of the poem Machaut presents an exemplar of : lai, complainte, chanson royale, baladelle, ballade, virelai and rondeau. The whole work, with performance, translation and somewhat abridged narrative, is available on CD-ROM, (Switten, 2001).


  • 1. Chace: a piece in imitative texture, where each voice follows or "chases" the one before it with identical music and words. Normally only voice is written out, and a rule or canon instructs the performer how to realize the piece. Some of these instructions are cryptic or humorous. The texts of the late 13th-early 14th century chaces often dealt with hunting, bird-calls, or serenades, inviting onomatopoetic effects. (return to text)
  • 2. These songs are transcribed and discussed in Hasselman, 1970.(return to text)
  • 3. Published in McGee, 1989.(return to text)
  • 4. As pointed out in Earp, 1991, Jehan's melodic and rhythmic style are too elaborate to be comfortable for actual dancing. But with respect to texture, his style hardly differs from Adam's. (return to text)
  • 5. For a fuller discussion, see Crocker, 1966, pp. 116-19 and 123-29, and Hasselman, 1970. A slightly different interpretation is found in Earp, 1991.(return to text)
  • 6. See Gunther, 1983 for a full discussion and edition. (return to text)

Selected Bibliography

Apel, Willi. French Secular Compositions of the Fourteenth Century. In Corpus Mensurabilis Musicae Vol. 53. American Institute of Musicology, 1970.

Crocker, Richard L. A History of Musical Style. New York: McGraw-Hill, 1966 and reprint: Dover, 1986.

Earp, Lawrence. Guillaume de Machaut: A Guide to Research. New York and London: Garland, 1995

_________. "Lyrics for Reading and Lyrics for Singing in Late Medieval France: The Development of the Dance Lyric from Adam de la Halle to Guillaume de Machaut." In Baltzer, Rebecca et al. (ed.) The Union of Words and Music in Medieval Poetry. Austin: University of Texas Press, 1991.

Gunther, Ursula. "Fourteenth-century music with texts revealing performance practice." In Boorman, Stanley, ed., Studies in the Performance of Late Medieval Music. Cambridge: Cambridge University Press, 1983.

Hasselman, Margaret Paine. "The French Chanson in the Fourteenth Century." PhD. Dissertation, University of California, Berkeley, 1970. 2 vols. Ann Arbor: University Microfilms, order #719830.

Hoppin, Richard, ed. Anthology of Medieval Music. New York: Norton, 1978.

McGee, Timothy J. Medieval Instrumental Dances. Bloomington and Indianapolis: Indiana University Press, 1989.

Palisca, Claude. Norton Anthology of Western Music, vol. 1. New York: Norton, 2001.

Reaney, Gilbert. "Concerning the Origins of the Rondeau, Virelai and Ballade Forms." Musica Disciplina 6: 155-66.

________. "The Development of the Rondeau, Virelai and Ballde Forms from Adam de la Halle to Guillaume de Machaut." In Festgabe Karl Gustav Felerer zum sechzigsten Geburtstag am 7. Juli 1962, ed. Heinrich Hueschen, 421-27. Regensburg: Bosse, 1962.

Switten, Margaret L. Teaching Medieval Lyric with Modern Technology: New Windows on the Medieval World, a project supported by The National Endowment for the Humanities and Mount Holyoke College. South Hadley, Mass: The Medieval Lyric, Margaret Switten, Director, 2001.

Stevens, John. Words and Music in the Middle Ages: Song, Narrative, Dance and Drama, 1050-1350. Cambridge: Cambridge University Press, 1986.

Wilkins, Nigel. The Lyric Art of Medieval France. Fulbourn: New Press, 1988, rev. ed. 1989.


Selected Discography

The French Fixed Forms (repertory listed here in rough chronological order):

The Mirror of Narcissus: Songs by Guillaume de Machaut (1300-1377). Gothic Voices, Christopher Page, Director. Hyperion CDA 66087, 1987. [A seminal recording of Machaut's music.]

For other Machaut recordings, see listings at: http://www.medieval.org/emfaq/composers/machaut.html.

The Spirits of England and France: Music for Court and Church from the Later Middle Ages. Gothic Voices, Christopher Page, Director. Hyperion CDA 66739.

Ars Subtilior. New London Consort, Philip Pickett, Director. Linn Records CKD 039.

The Garden of Zephirus: Courtly songs of the early fifteenth century. Gothic Voices, Christopher Page, director. Hyperion CDA66144, 1985.

The Castle of Fair Welcome: Courtly songs of the later fifteenth century. Gothic Voices, Christopher Page, director. Hyperion CDA 66194, 1986. [Includes Binchois' wrenching setting of Christine de Pizan's "Deuil angoisseux"]

Sweet Love, Sweet Hope. The Hilliard Ensemble. Isis records, CD030, 1996. [Songs from Oxford 213, by Dufay, Johannes Rezon, Jo. Hasprois, Bartolomeus Brollo, Paullet, Guillaume Malbecque, Prepositus Brixiensis and Anon.]

Guillaume Dufay, Complete Secular Music. The Medieval Ensemble of London,
directed by Peter Davies and Timothy Davies. L'Oiseau-Lyre (Decca), 5 CD set: 452 557-2, 1981/1997.

Works from the Italian Trecento:

A Song for Francesca: Music in Italy, 1330-1430. Gothic Voices, Christopher Page, Director. Hyperion CDA 66286, 1988.

Decameron: Ballate monodiques de l'Ars Nova Florentine. Esther Lamandier, voice and portative organ, harp, vielle and lute. Astree E 7706, 1986. [A French performer who taught at the Scola Cantorum in Basel performs music from the Rossi codex and the Squarcialupi manuscript.]

Ecco la Primavera. Early Music Consort of London, dir. David Munrow. Decca 436 219-2,1970/1993. [Particularly nice recordings of Landini's "Questa fanciull'amor" and the caccias "Con dolce brama" by Magister Piero and "Chon brachi assai" by Giovanni da Firenza.]

Return to:

|top | definitions | historical context | character and provenance |
|notes | bibliography |

Encyclopedia | Library | Reference | Teaching | General | Links | About ORB | HOME

Copyright (C) 2001, Margaret P. Hasselman. This file may be copied on the condition that the entire contents, including the header and this copyright notice, remain intact. (The contents of ORB are copyright © 1995-2001 Laura V. Blanchard and Carolyn Schriber except as otherwise indicated herein.)