We use cookies to enhance your experience on our website. By continuing to use our website, you are agreeing to our use of cookies. You can change your cookie settings at any time. Find out more


Music from the Earliest Notations to the Sixteenth Century


CHAPTER 4 Music of Feudalism and Fin’s Amors
Richard Taruskin


The earliest secular repertories of which we have direct knowledge consist of songs by knightly poets of courtly love and feudal service. Stylistically, they are remarkably like the sacred repertories with which we have been dealing so far. Not that this should surprise us: if these secular songs were thought worthy of commemoration and permanence (that is, worthy of writing down), they must have had some transcendent or elevating purpose like that of the sacred. And as surely as style follows function, a like purpose should entail a like manner.

The earliest such written-down knightly songs in a European vernacular (that is, a currently and locally spoken language) originated in Aquitaine, a duchy whose territory occupied parts of what is now southern and south-central France. It had been conquered by Charlemagne in the late eighth century and incorporated into the Carolingian Empire; but with the weakening of the Empire as a result of invasions by Normans on one side and Muslims on the other, royal influence over Aquitaine gave way over the course of the ninth and tenth centuries to several independent noble families who established local jurisdictions and maintained networks of patronage and protection among themselves. Eventually the counts of Poitou emerged as the most powerful among these clans and, from 973, asserted dominion over the whole territory and took the title of Duke. (Later, the marriage of the duchess Aliénor or Eleanor of Aquitaine to the French king in 1137 joined Aquitaine to France; and her second marriage, to the Norman duke who later became King Henry II of England, led to a long struggle over the territory that would not be ended until the fifteenth century.)

It was during the period of Aquitaine’s relative independence that its courtly poetic and musical traditions arose. William (Guillaume), seventh count of Poitiers and ninth duke of Aquitaine (1071–ca. 1127), was the first European vernacular poet whose work has come down to us. The tradition, socially speaking, thus began right at the top, with all that that implies as to “highness” of style, tone, and diction. The language William used was Provençal, alias Occitan or langue d’oc, from the local word for “yes.” (Old French, spoken to the north, was called langue d’orl for the same reason.) In Provençal, poetry was called trobar, meaning words “found,” and a poet was called a trobador, a “finder” of words. In English we use the Frenchified form, troubadour.

A troubadour’s subject matter was the life he led, viewed in terms of his social relations, which were ceremonial, idealized, and ritualized to the point of virtual sacralization. In keeping with the rarefied subject matter, the genres and styles of troubadour verse were also highly formalized and ceremonious, to the point of virtuosic complexity of design and occasional, sometimes deliberate, obscurity of meaning.

The genres reflected social relations directly. Feudalism, arising in unsettled conditions of weak central power and frequent ruinous invasion, was based on land grants and on contractual, consensual exchanges of service and protection on which everyone’s welfare depended. The bonds of honor thus pledged were taken very seriously indeed. The utopian ideal—never realized except (more or less theoretically) in the tiny, short-lived “Latin Kingdom of Jerusalem” established by the Crusaders in 1099—was a wholly hierarchical—and therefore, theoretically, a wholly harmonious—society. Under feudalism, all land was legally owned by an elected (rather than a hereditary) king, who deeded and parceled it out to the greatest nobles in the form of “fiefs” (from the Latin feodum, whence “feudal”), who deeded it in turn to lesser nobles, and so on down to the manorial barons and their serfs, who actually worked it.

The granting of a fief created the relationship of lord (or suzerain) and vassal. The bond thus created was solemnized downright liturgically, in a ritual of homage whereby the vassal, placing his hands in the lord’s, swore an oath of fealty that obliged him to perform certain specified acts and services, including military service. The suzerain, in turn, bound himself to protect the vassal from incursions. The feudal nobility was thus primarily a military caste system, a hierarchy of knights or warriors-in-service. These military bonds were at first envisioned as a system of mutual defense (although in reality disruptive conflict among lords was common), but in the period of the Crusades the knightly armies went on the offensive. William IX of Aquitaine, our first troubadour, led a Crusade himself in 1101; his unlucky army never reached the Holy Land.

Several genres of troubadour verse celebrated feudal ideals. A sirventes was a song from vassal to lord about knightly service or about some theme of political alliance; such a song could be either serious or satirical. An enueg (compare the French ennui) was a complaint about infractions of knightly decorum. A gap (compare jape or gibe) was a bluster-song, glorifying one’s own exploits or issuing a challenge. The most serious of such types was the planh (compare the French plainte), a eulogy on the death of a lord. There were also many songs about crusading zeal.

The true heart of the troubadour legacy, however, was the canso, which means a love poem—or better, perhaps, a poem about love. For the love celebrated by knightly singers was just as “high,” just as formalized and ritualized, as any other publicly enunciated theme. And the poems of the troubadours were always meant for public performance—hence the music!—not for private reading; theirs was still an eminently oral tradition. Modern scholars have christened the subject of the canso “courtly love” (amour courtoise); the troubadours themselves called it fin’ amors, “refined love,” defined by one modern authority as “a great imaginative and spiritual superstructure built on the foundation of sexual attraction.”

It is widely thought that Arabic sung poetry—known in southern Europe from the ninth century, and also emphasizing secret love and the spiritualization of the erotic (including the homoerotic)—had a formative influence on the concept of fin’ amors. Its main genre is the nawba (or nuba), a lengthy vocal performance accompanied by the oud (literally “wood”), a gourd-shaped plucked-string instrument from which the European lute had been adapted by the thirteenth century. It consisted of several stanzas (some in Arabic and some in Persian), connected with improvisatory instrumental interludes. No musical relationship between the nawba and the troubadour repertory has as yet been definitely established, but some modern performers of early music have experimented effectively with performance practices derived from those of present-day Arab musicians.

The love songs of the troubadours were like their knightly songs in that they emphasized service and the idolization of those above, as the lady was invariably held to be. The style was self-consciously lofty, as exemplified by the imagery of the most famous of all cansos, Can vei la lauzeta mover by Bernart de Ventadorn (d. ca. 1200), which begins with an unforgettable metaphor comparing the joy of love to the soaring and swooping of a lark in flight. This is contrasted with the lovesick poet’s unhappy state, condemned to adore a cold and unresponsive lady from afar.

Fin’ amors was furtive and hopeless as a matter of course, because the lady was always held to outrank her lover. She was married into the bargain, as a rule, though often left alone for long periods while her husband was out on campaigns or Crusades. At such times she was the effective ruler of his domain, as the Occitan word for lady—domna—already suggests (compare the Latin domina and the Italian donna). Her identity is always concealed behind a code name (senhal; in Bernart’s song it is Tristan), supposedly known only to the lady and her lover.


fig. 4-1 Medieval lutenist, from an illumination in a thirteenth-century manuscript now at the Biblioteca Nazionale in Florence, Italy. The instrument he is playing entered Europe as war booty from the Crusades.

But secrecy and illegitimacy should not be confused with licentiousness. The con-ventions of fin’ amors heightened the unavailability of the lady as actual lover and made her an object not of lust but of veneration. The canso was thus essentially a devotional song, a song of worship—another link with the sacred sphere, especially with the burgeoning liturgy of the Blessed Virgin. Veneration of the lady, like veneration of Mary, promoted not license or sensuality but rather the sublimation of amorous desire in charity, self-mortification, and acts of virtue. It was another bond of honor, hence a quintessentially feudal attitude.


In Bernart’s case it was easy enough to feel outranked by the lady: like many of the later troubadours, he was a commoner—according to various traditions the son of the baker or the furnace stoker in the castle of Ventadorn, near Poitiers—who rose to prominence, and received noble patronage, strictly on his merits as a poet. (Bernart’s patron was Eleanor of Aquitaine herself, with whom he traveled to France after her first marriage, and to England after her second, thus spreading his art abroad.) While the art of the troubadours was a quintessentially aristocratic art, an art of the castle, it was not an art practiced only by aristocrats. Rather, whoever the actual practitioner may have been, it was an art cultivated and patronized by aristocrats and expressive of their outlook.


fig. 4-2 Bernart de Ventadorn, from an illumination accompanying his vida, the biographical preface to his collected song texts without melodies, in a manuscript copied in Italy in the late thirteenth or early fourteenth century, by which time the Occitan culture of southern France had already been destroyed. (Only five of the thirty-seven manuscripts known to have contained troubadour songs had musical notations.) The manuscript found its way to the Bibliothèque Nationale in Paris as war booty, having been confiscated from the Vatican Library by Napoleon’s army. The portrait of Bernart, dating from at least a century after his death, is obviously fanciful.

Indeed, the actual practice—the actual performance, that is, of the noble song-product—was usually left to what we now call minstrels: professionals of a lower caste, singer-entertainers called joglars in Provençal (jongleurs in French, both from the Latin joculatores, “jokers”); the derivation of our English word “juggler” from joglar should leave no doubt about its subartistic connotation. Most of the commoner-troubadours like Bernart started out as minstrels who learned the work of the noble poets by rote and who later developed creative facility in their own right. The relationship of troubadour to minstrel, and particularly the means of transmission from the one to the other, attest that the art of the troubadours remained an oral art. A noble poet would compose a song and teach it to a minstrel, thus sending it out into the oral tradition from which it might be transcribed, with luck, a hundred years later.

For written documentation of the troubadour art began only when the tradition was already moribund. The manuscripts containing troubadour songs, called chansonniers, are retrospective anthologies prepared in the middle of the thirteenth century. (Any song found in multiple copies in these late sources exists in multiple variants, thus precluding the restoration of a definitive “text,” assuming there ever was such a thing.) Besides the songs themselves, chansonniers contain fanciful portraits and biographies (vidas) of the poets. Their purpose was commemorative and decorative; they had nothing to do with practice or performance. They were “art objects,” rich “collectibles.”

That the composition of troubadour songs was just as much an oral practice as their transmission and performance is shown by a very revealing anecdote in the vida of Arnaut Daniel, one of the greatest knightly poets, known for his exceptional virtuosity in rhyme. It supposedly happened at the court of Richard I (Lion-Heart), Eleanor’s son. Another troubadour had boasted that he could compose a better poem than Arnaut and challenged him to a contest. The king confined the two poets to different rooms in his castle, stipulating that at the end of the day they were to appear before him and recite their new poems, whereupon Richard would determine the winner of the bet. Arnaut’s inspiration failed him; but from his room he could hear his rival singing as he composed his song, and learned it by heart. When the time of the trial came he asked to perform first and sang his rival’s song, leaving the latter to look like the copycat.

Like many of the anecdotes in the vidas, this one probably never happened. (There is no corroborating evidence that Arnaut, himself a nobleman, was ever in anyone’s employ, or that he knew Richard, or that he went to England.) But, as the Italian proverb has it, se non è vero, è ben trovato: “even if it isn’t true, it’s very apt (literally, well made-up)”—and note how the Italian for “making up” comes from the same root stock as trobar. What is so apt about it, and revealing, are the points the author of the vida took for granted: first, that a troubadour in the act of composition did not write but sang aloud; and second, that a troubadour could memorize a song at an aural glance. These are the assumptions of an oral culture.

It was that congruence of creating and performing as oral acts, and that ease of memorization, that made the minstrel an apt and necessary accessory to the troubadour. Yet because the creation of poetry, as opposed to its performance, was nevertheless viewed as a noble pastime rather than a profession, it could be practiced by lords—and by ladies, too. The vidas tell us of at least twenty lady troubadours (for which the Provençal word was trobairitz) who created courtly songs but never sang them, at least in public.


The only type of troubadour love song that emphasized the joy of consummation was the thrilling genre known as the alba, or “dawn-song.” The lovers, having passed a clandestine night together in oblivious bliss, are aroused—by the sun, by singing birds, by a watchman’s cry, or by a confidant—to the breaking day and to the peril of discovery. The most famous alba was Reis glorios by Guiraut de Bornelh, a contemporary of Bernart and, like him, a commoner whose skill found favor with the noble audiences (Ex. 4-1). Where Bernart’s stanza (or cobla, to use the Occitan word) had consisted, in Can vei la lauzeta, of eight different phrases, each corresponding to a line of poetry, Guiraut’s is regularized by an initial melodic repetition (or pes)—as in the Salve Regina melody discussed in the previous chapter—and by a concluding refrain as in, say, a Frankish Kyrie. The resemblance of Guiraut’s melody to that of the Kyrie verse Cunctipotens genitor (Ex. 2-14b) has been noted.


ex. 4-1 Guiraut de Bornelh, Reis glorios (first verse)

The higher the style of a troubadour melody, the more likely were its chant affinities. Comparison of Bernart’s or Guiraut’s melodies with those of the late Frankish chant discloses a great similarity of style. Like Salve Regina or Kyrie IX, they are exemplary “first mode” melodies according to the rationalized concept of mode studied in chapter 3. (The composers probably picked up the style by ear on the basis of the chants they heard sung.) There are even instances that show the influence of late chant poetry on troubadour diction and sentiment. A familiar Provençal dictum—fin’ amors, fons de bontat, “courteous love is the source of all goodness”—echoes in close cognates the tenth-century Kyrie verse Fons bonitatis. (Its third-mode melody, shorn of the verse but still sporting the incipit as a title, can be found in modern chant books as “Kyrie II.”) The hypothesis that the music of troubadour song—the performance medium of a liturgy of aristocratic mores—aped the actual liturgical music of its time has been gaining strength as more is learned about actual twelfth-century chant. Quite near William the Ninth’s seat at Poitiers was Limoges, another Aquitanian town and the site of the Benedictine abbey of St. Martial, the greatest center for the production of Latin versus on which the troubadours modeled their vers (to use the Provençal word for poetry-with-music).


fig. 4-3 The Languedoc region in the twelfth century. Note the close proximity of Limoges, the monastic center, where Latin liturgical poetry to music was composed in profusion; and Poitiers, the center of troubadour activity.

Rhythmically and formally, Latin versus (or conductus as it was called in northern France) was just as various and almost as virtuosic as troubadour songs. Actual Provençal words occasionally appear in versus from St. Martial. Most striking of all, there are subgenres of versus that, like the most elevated troubadour genres, straddle the nebulous line between the sacred and the secular. Some of them have actual troubadour parallels, for example theplanctus (=planh). St. Martial manuscripts contain a celebrated planctus for Charlemagne and an even more illustrious song called planctus cigni (“The lament of the swan”), a moving metaphor of exile, in which a swan, caught in a storm over the sea, laments the loss of its verdant homeland. The famous theologian Pierre Abelard (1079–1142), an exact contemporary of William of Aquitaine, wrote six planctus on biblical themes, two of which exist in staff notation. A large collection of versus in an early thirteenth-century French manuscript best known for polyphonic music includes planctus commemorating a whole honor roll of recently deceased aristocrats and churchmen, and what may be described as Latin sirventes (one of them concerning Pope Innocent III’s excommunication of Otto IV, the Holy Roman Emperor, in 1210 and the ensuing war of succession). Many planctus settings (including the planctus cigni) are in a form resembling the older (pre-“Victorine”) liturgical sequence, with its melodically paired verses of differing lengths. This form, too, had its Provençal counterpart, called the descort. The name, translated literally, means “discordant,” the idea being that its component stanzas are in varying (disagreeing) rhyme and meter schemes, requiring new melodies for each. Such a structure particularly suited narrative poems.

Other troubadour genres or individual melodies affected a mock-popular style that may have drawn stylistically not on chant but on otherwise unrecorded folk idioms. A chantar m’er de so gu’en no volria (“I must sing of that which I would rather not”), the one poem with surviving tune attributed to the late twelfth-century trobairitz Beatriz, Countess of Dia, departs markedly from the chant idiom. Its overall structure is that of the regularized or “rounded” canso with its repeated couplet and final refrain (AB AB CD B), but the tune alternates cadences on E and D in a fashion never encountered in first-mode chants but common in the dances and dance songs of a slightly later period. (Such endings would be designated ouvert and clos—“open” and “shut”—in thirteenth century dance manuscripts; they correspond to, and prefigure, what we would now call half and full cadences.) One genre that always affected a mock-popular tone was the pastorela, in which a knight seduces (or tries to seduce) a shepherdess. The best-known survivor of this genre is L’autrier jost’ una sebissa (“The other day by a hedge row”) by Marcabru (or Marcabrun), one of the early troubadours, who served in his youth at the court of William of Aquitaine and who memorialized his patron in a Crusader song (Pax! In nomine Domini, “Peace! In the name of the Lord”) that actually mixes Latin verses with vernacular ones. Texts that do this are called macaronic (“jumbled” like macaroni).

Another mock-popular genre was the balada, or dance-song, of which a rare surviving example is the anonymous A l’entrada del tens clar (Ex. 4-2). This melody, of an altogether different character from the preceding ones, seems to be a sophisticated imitation folk song with its call-and-response verses (ending with “Eya,” Occitan for “Hey!”), its half and full cadences, and its lengthy refrain.


ex. 4-2 A l’entrada del tens clar (balada)


If, as the text suggests, this song is meant to accompany a carole, an actual public dance, then its rhythm has got to be metrical. No such information is conveyed by the actual notation, which like that of the other troubadour melodies is indistinguishable from the “quadratic notation” (notation with square -note-heads) used in contemporary chant manuscripts. (“Contemporary” here means thirteenth-century, the period of the retrospective chansonniers.) Meter has to be supplied conjecturally, on the basis of the words. In the case of a dance, no one is likely to object to such a conjecture; but the question of the proper rhythmic style for troubadour songs—indeed, for verse-music in general before the thirteenth century—is one of the most hotly contested issues in musical scholarship.

The poetry is of course metered; indeed, metrical design (along with rhyme scheme) was a field in which the troubadours vied to excel one another in virtuosity. The question is whether the metric patterning was reflected in the music in patterns of note lengths or by means of stress patterns. Or, as some maintain, were the troubadour songs performed in a supple rhythm modeled, perhaps, on that of the sacred chant? Such an attribute of “high” style would free the troubadour art from any taint of the “popular”—if indeed that was thought desirable.

This is certainly not the place to adjudicate such a question; but it should be emphasized that we are just as much in the dark about the intended rhythmic performance of versus (or even Frankish hymns and sequences) as we are about vers. And it should be noted that the scholarly consensus has lately been swinging, in the case of all of these repertories, away from the a position (now regarded as anachronistic) that favored applying quantitative meters wherever possible. This theory, which dominated editions and performances of troubadour melodies and late chants alike in the earlier part of the twentieth century, was based on a misreading of treatises on the rhythmic performance of late medieval polyphonic music (to be described in chapter 6).

The approach most favored now is the so-called “isosyllabic” approach, whereby all syllables, whether sung to a single note or to a group or two or three notes, are given roughly equal length. Another possibility, which also has its adherents, is the “equalist” approach that makes precisely the opposite assumption. It gives all notes the same length, regardless of how many of them are given to a syllable. This fairly radical “solution” to the problem of rhythmic interpretation is the one recommended by the editors of modern chant books (officially adopted by the Catholic church in 1904 and in general Catholic use until 1963), and is therefore the one most widely practiced today wherever the Gregorian chant is still sung as service music.


There is one more important genre of troubadour poetry with music: the tenso (or joc-partit), an often jesting debate-song that involves two or more interlocutors, and that was sometimes, but not necessarily, actually a joint composition by two or more poets. The subject matter could be some fine point of love or feudal service (like, “if you love a lady, would it be better to be married to her or to have her love you back?”), or it could be—and most often was—about poetry itself. Here the troubadour addressed his craft directly and with marvelous self-consciousness. The tenso was thus a sort of school for poets and can be extremely instructive for us.

One of the favorite themes for debate was the eternal conflict between trobar clus and trobar clar, between “closed” or difficult poetry for connoisseurs and “clear” poetry designed for immediate pleasure and easy communication. The virtues claimed for the first were its technical prowess, its density of meaning, and the exclusive nature of its appeal, which lent it an ability to create an elite occasion and foster solidarity among a coterie of insiders. It promoted social division and hierarchy, and was therefore an art quintessentially expressive of aristocratic values. The virtues claimed for the second were its greater technical prowess (or so it was argued, since as the Roman poet Horace famously remarked, the greatest art is the art that conceals art) and its power to create a sense of community and shared values. Within the narrow social context of troubadour culture this is hardly to be looked upon as a “democratic” ideal. It might better be regarded as a feudal piety.

These arguments were given an early, classic exposition in a tenso by Guiraut de Bornelh, a recent convert to trobar clar, in mock debate with a fellow troubadour, Raimbaut d’Aurenga (called Linhaure), who remained loyal to trobar clus. The melody, unfortunately, has not survived. In somewhat abridged translation, the dispute runs as follows:

  1. 1. I should like to know, G. de Bornelh, why you keep blaming the obscure style. Tell me if you prize so highly that which is common to all? For then would all be equal.

  2. 2. Sir Linhaure, I do not take it to heart if each man composes as he pleases; but judge that song is more loved and prized which is made easy and simple, and do not be vexed at my opinion.

  3. 3. Guiraut, I do not like my songs to be so confused, that the base and good, the small and great be appraised alike; my poetry will never be praised by fools, for they have no understanding nor care for what is more precious and valuable.

  4. 4. Linhaure, if I work late and turn my rest into weariness to make my songs simple, does it seem that I am afraid of work? Why compose if you do not want all to understand? Song brings no other advantage.

  5. 5. Guiraut, provided that I produce what is best at all times, I care not if it be not so widespread; commonplaces are no good for the appreciative—that is why gold is more valued than salt, and with song it is just the same.3

For the troubadours, these opposing sentiments were not so much passionately held convictions as postures; many poets cultivated both styles depending on the occasion and saw no compelling reason to choose between them. And yet the debate continues. It will run through this book like a red thread, steadily gathering force and urgency as the audience for art changes (and inexorably widens) over time. For one of the enduring characteristics of “high art,” and a perennial source of contention, is the fact that it is produced by and for political and social elites. That, after all, is what makes it “high.” But then there can be many reasons for hiding meaning, and not all of them are proud.

Nor is original purpose an inherent limitation on meaning or value. Art devised to serve the interests or the needs of a feudal aristocracy must be serving other interests now, if it is serving any interests at all. And yet trobar clus and trobar clar, by other names and in other forms, are with us still. Each still has its ardent defenders and its adamant detractors. Their subtexts and agendas are many. There is no more consequential theme in the history of art.

The art of the troubadours lasted about two hundred years. It declined together with the Provençal culture that sustained it. Many of the later troubadours fled southward, into present-day Spain and Italy, at the time of the so-called Albigensian Crusade (from 1208). This was a drawn-out, devastating war of aggression waged by the northern French against the courts of Languedoc under pretext of a religious campaign. (The ostensible targets were the so-called Cathari or Albigenses, adherents of an old philosophical tradition called Manicheism that had been declared a heresy by the Catholic Church.) Guiraut Riquier (ca. 1230–ca. 1300), the last of the troubadours, found employment at the court of Alfonso X of Castile, which became a major center of vernacular courtly and devotional song in the later thirteenth century, but no longer in Provençal. Nor was Guiraut the only troubadour who stimulated the spread of vernacular poetry into other languages. As we will see, the nobleman Arnaut Daniel (by common consent the preeminent master of trobar clus) had a formative, if posthumous, influence on the art of Dante and Petrarch, as well as their fourteenth-century Italian musical contemporaries. But long before that, indeed by the end of the thirteenth century, the art of the troubadour was at an end.


(3) H. J. Chaytor, The Troubadours (London, 1912), pp. 38–9.

Citation (MLA):
Richard Taruskin. "Chapter 4 Music of Feudalism and Fin’s Amors." The Oxford History of Western Music. Oxford University Press. New York, USA. n.d. Web. 18 Oct. 2021. <https://www.oxfordwesternmusic.com/view/Volume1/actrade-9780195384819-div1-004002.xml>.
Citation (APA):
Taruskin, R. (n.d.). Chapter 4 Music of Feudalism and Fin’s Amors. In Oxford University Press, Music from the Earliest Notations to the Sixteenth Century. New York, USA. Retrieved 18 Oct. 2021, from https://www.oxfordwesternmusic.com/view/Volume1/actrade-9780195384819-div1-004002.xml
Citation (Chicago):
Richard Taruskin. "Chapter 4 Music of Feudalism and Fin’s Amors." In Music from the Earliest Notations to the Sixteenth Century, Oxford University Press. (New York, USA, n.d.). Retrieved 18 Oct. 2021, from https://www.oxfordwesternmusic.com/view/Volume1/actrade-9780195384819-div1-004002.xml