Med Song

Yale University Department of Music
Reconsidering "High Style" and "Low Style" in Medieval Song

Author(s): Elizabeth Aubrey
Source: Journal of Music Theory, Vol. 52, No. 1, Essays in Honor of Sarah Fuller (Spring
2008), pp. 75-122
Published by: Duke University Press on behalf of the Yale University Department of
Music
Stable URL: https://www.jstor.org/stable/40607030
Accessed: 04-04-2020 05:34 UTC
REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/40607030?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Duke University Press, Yale University Department of Music are collaborating with
JSTOR to digitize, preserve and extend access to Journal of Music Theory
This content downloaded from 149.171.67.148 on Sat, 04 Apr 2020 05:34:49 UTC
All use subject to https://about.jstor.org/terms
Reconsidering "High Style"
and "Low Style" in Medieval Song
Elizabeth Aubrey
Abstract Using the concept of "style" in analysis runs the risk of circularity, where features of individual
works are identified as belonging to a style whose definition itself is derived from those features. This pitfall
undermines studies of the songs of the twelfth- and thirteenth-century troubadours and trouvères that delin-
eate a "high style," including chansons, and a "low style," including dances and pastourelles. The dichotomy
originated in the nineteenth century with Gaston Paris's concept of amour courtois, from which Roger Dragon-
etti later derived the term grand chant courtois, now a common label for "high-style" songs. Other literary
scholars, notably Paul Zumthor and Pierre Bee, have discussed problems in classifying styles and genres.
References to genres in medieval texts are ambiguous, and manuscripts rarely group songs by genre. Theorists
such as Raimon Vidal, Jofre de Foixà, and Johannes de Grocheio do not present a clear-cut or consistent
stratification of genres. John Stevens, Christopher Page, and others have proposed features of "high style"
and "low style" that do not entirely agree. An examination of their examples and additional ones demonstrates
that a perception of "style" can be subjective and circular, and that the notion of "high style" and "low style"
is an oversimplification.
"style," writes Robert PASCALL in The New Grove Dictionary of Music and
Musicians (2001, 638), "is manner, mode of expression, type of presentation.
For the aesthetician style concerns surface or appearance, though in music
appearance and essence are ultimately inseparable. For the historian a style
is a distinguishing and ordering concept, both consistent of and denoting
generalities; he or she groups examples of music according to similarities
between them." Pascall goes on to say, "Style manifests itself in characteristic
usages of form, texture, harmony, melody, rhythm and ethos; and it is pre-
sented by creative personalities, conditioned by historical, social and geo-
graphical factors, performing resources and conventions." The author of the
entry "Style" in The Harvard Dictionary of Music points out that the rhetorical
tradition from which our concept of style emerged "distinguishes style from
It is with the greatest pleasure that I offer this study to honor my friend and colleague Sarah Fuller. From our first
encounter at Kalamazoo more than twenty years ago, in treks together along city streets and through the chambers of
off-the-beaten-path museums, during delightful meals whose spicy and delectable flavors proceeded more from the
conversation than the food, and certainly in my study of her many stunning contributions to scholarship, I have come
to a deep and abiding appreciation of Sarah's wit, warmth, and supreme erudition. The riches that she has imparted to
me, and to others, are boundless, and for them I am forever in her debt.
Journal of Music Theory 52:1, Spring 2008

DOI 10.1215/00222909-2009-011 © 2009 by Yale University 75
76 JOURNAL of MUSIC THEORY
content - the manner in which something is said as distin

being said," but that applying this distinction to music i
because music is essentially nonrepresentational. The pitch
that define the style of a composition also constitute its conte
music has only style" (Randel 2003, 846).
We use the concept of style as a fundamental tool of an
features in musical works that distinguish them from or conn
ers, and attempting to place them in some sort of context. Th
perils in this analytical approach, among the most insidiou
of circular argument. As The Harvard Dictionary of Music puts
The analysis of style ... is more likely to concentrate on the

normative categories against which to test the individual work
centrate on the uniqueness of the individual work. Here it runs
cularity. The norms of a particular style can only be discovered
study of individual works. But the criteria for significance
study of the individual work are likely to rest in some measure o
tion of the style to which the individual work is thought to
2003, 846)
As we shall see, this pitfall lies at the heart of many stu

song.
Also problematic is attempting to determine how consc
and performers were of their own stylistic heritage and whet
was or was not guided by deliberate incorporation of that her
individual style. Pascall (2001, 638) says that
by the act of creative will a composer asserts something; he m

of some kind. He inherits a usable past and acts by intuitive vis
of his vision builds on a stylistic heritage, has a style and impor
bequeaths an altered heritage. The stylistic heritage may be
procedures which condition the composer's intuitive choice and
general which limits the particular, the relevant available resou
tial context of creation.
From at least as early as the fifteenth century, writings of theorists and musi-
cians provide glimpses of how deliberately creators of music either were guided
by a received style or departed consciously from it. Such self-awareness is
difficult to document in any era, but even more so before the fifteenth cen-
tury, when few clues exist that reveal how composers and performers viewed
their own works, their art in general, their compositional processes, and their
performances.
The word style in the formulation "high style/low style" as it is applied
to medieval song suggests internal poetic and musical features that place a
work in one category or the other - language, rhetoric, theme, structure, and
melodic range, intervals, texture, and contour. Other expressions broaden
the dichotomy to include external factors: "courtly" and "popular" suggest a
Elizabeth Aubrey ~ "High Style" and "Low Style" in Medieval Song 77
social distinction of audience or venue; "aristocratic" or "learned" and "jon-

gleuresque" point to a difference in the social class of the performer or in the
educational level of performer or composer. We distinguish among types of
songs not only by their internal features but also by external conditions, such
as social setting, audience, function, or the training and erudition of
composer or performer. Consideration of factors such as these has become
intertwined with a delineation of genres, whereby different types of medieval
songs are assigned to either a "high-style" category or a "low-style" category.
This delineation, in turn, has produced broad theories about composition
and performance of different genres, especially the problematic questions of
rhythm and the use of instruments.
While scholars of medieval literature have vigorously debated questions
about how to arrive at a definition of style in medieval song and how to apply
it to composers and their works, including some healthy skepticism about the
"high style/low style" dichotomy (for a summary, see Switten 1995, 104-52),
most musicologists have accepted the framework of "high style" and "low
style" - especially for songs of the twelfth and thirteenth centuries in Occita-
nia and France - with little discussion of the hermeneutical or epistemologi-
cal issues that plague the dichotomy. But neglect of the two problems sug-
gested above - the circularity of identifying features common to a group of
works to delineate a collective style, which in turn serves as the norm by which
to define the style of individual works, and the hazards of discerning from
sparse, vague, or contradictory evidence how a composer or performer
thought about his or her own works - undermines the "high style/low style"
framework.
The earliest formulation of a distinction between a "high style" and a

"low style" in medieval song can be traced to the work of Gaston Paris, who
in the 1860s spoke of a "littérature des clercs" and a "littérature du peuple"
(in a lecture dated December 7, 1869, to the Collège de France, reprinted in
Paris 1885, 1:82; see Huit 1996, 207-8), the former of which he perceived as
artificial and conventional, and the latter as spontaneous and vital. Soon
scholars were differentiating between "learned" and "folk" songs and, more
particularly, between aristocratic love songs and popular dance songs (see
also Allen 1931, 67-79, which distinguishes between "popular song" and
"school poetry"). These early discussions address both internal features of
the songs and the sociological environment within which they were created
and performed.
In his influential 1960 study of rhetoric in "courtly" song, Roger Drag-
onetti tracked the three genera, or "styles," of discourse in ancient rhetoric,
the "grand" (ancient stylus gravis, medieval stylus grandiloquus) , "middle"
(stylus mediocris), and "simple" (stylus humilis), into later medieval writings
on rhetoric and poetry. Songs differ in their subject matter, the "dignity" of
the personages in the text, the words, figures, and rhetorical arguments,
the sincerity of the poet, and poetic structure. He saw features of grand
style in the love songs of the troubadours, trouvères, and late

ian poets - songs that he termed the grand chant courtois,
become a standard label for "high-style" songs. Any poem
elevated characteristics he classified as "lower" in styl
implied distinction of genre: songs about courtly love - th
troubadours and the chansons of the trouvères - are in t
while songs on other themes (which Dragonetti does not
"low" style.
Three years later, Paul Zumthor (1963, 123-78) postula
of poetic "register," whose complexity is too vast to be ex
which implies a stylistic fluidity from "high" to "low" (see als
161-85). Zumthor's wide-ranging discussions of poetics in p
the 1960s until his death in 1995 demonstrate how difficult it is to isolate
specific characteristics of subject matter, language and vocabulary, rhetor

gestures, poetic structure, and performative qualities to one type of song
asserted that medieval terminologies referencing style or genre are v
and inconsistent, and he pointed out that a theory of genres emerged
after the Middle Ages and remains elusive yet today. Delineating a ge
according to Zumthor, is complicated by the transmission of songs -
mouvance, itself a complex idea. While articulating his theory of reg
through which genres are expressed, he also introduced the idea of "regis
interference," for example, the presence of a refrain in a grand chant cour
which causes a "rupture" of its "high" register, a rupture of rhythm, me
syntax, language, and style (1972, 245-47).
Hans-Robert Jauss, in an important article published in the first
of Poétique (1970),1 offered a new approach. Famous for his articulat
reception theory, Jauss proposed that, though subject to the reader's
unique "horizon of expectations," and indeed, to variability within the me
eval tradition itself, the features of a medieval song nonetheless can b
ognized, and that among these features we can discern a "dominant
gives the song a generic identity. Jauss acknowledged that a modern u
standing of medieval genres must of necessity be imperfect, but he argue
the notion of genre as a useful construct.
Building on the largely structuralist approach of Dragonetti an
Zumthor, who focused on the songs and attempted to construe their
ing from within, Pierre Bee (1977-78) pursued a more sociological appr
postulating three sociopoetic registers, the aristocratisant, the jongleuresq
popularisant, and the folklorisant, of which the second is "pivotal," the jong
acting as a "kingpin" among the three (1:30). Even more than Zumth
discussion of poetic register and registrai interference, Bee's formulat
1 The article emerged from Jauss's work as coeditor

1968 of
and continued through 2005. The series remains
incomplete. See Gumbrecht 1996.
the Grundriss der romanischen Literaturen des Mittelalters,
whose many volumes and fascicles began appearing in
the registers as participles - "nobilizing," "popularizing," "folklorizing" -

evokes an active continuum along which a song might tend, rarely falling
exclusively into one register or another. The mouvance of a text is due princi-
pally to the mobility of the jongleur, who is a re-creator, a mediator between
the composer and the hearer (1:23-25). The jongleur sang all genres, all reg-
isters, and was the chief agent of registrai interference that one detects in so
many songs (1:27-28).
Like other littéraires, Bec acknowledged that medieval composers, theo-
rists, and performers did not themselves have a systematic or consistent view
of genres:
People in the Middle Ages seem not to have had the idea that the poetic texts
could be arranged into generic groups. Their thinking about poetry is always
fluctuating and their literary vocabulary, even their designations of the differ-
ent genres, must be handled with great care: in fact there is very little that can
be drawn from their terminology. (1977-78, 1:35)2
Bee's own system of registers and genres incorporates features of poetical and
musical form and vocabulary as well as external factors of class and function.
Still insisting on the demarcation of the grand chant courtois and its "satellites" -
the sirventes, planh, tenso, jeu-parti, and lai-descort - apart from other genres,
Bee yet acknowledged the difficulties of attempting to delineate among
them:
A sociological discrimination of different text types is thus extremely complex

because it must be placed not only on the level of the three dimensions ana-
lyzed earlier (creator, mediator, addressee), but also on the level of the multiple
combinations among them. In other words, the possibilities presented to the
creator (composer/jongleur/cleric/"people") can be found also at the level of
the mediator as well as that of the addressee (knightly /clerical/courtly/
"popular" public), thus multiplying and blurring the sociological connotations
of an individual text. (1:24)3
Conceding that no genre is "pure" in its register, Bee was forced to classify
certain genres as neither aristocratisant nor popularisant, but as "hybrid" (pas-
tourelle, reverdie, crusade song, motet, estampie, and rotrouenge) or "bourgeois"
(sotte chanson and fratrasie); Table 1 represents Bee's outline of genres (com-
ments in bold are my additions). Registrai interference can be seen especially
in the presence of "aristocratic" features in many "popular" genres, but also
2 "Les hommes du moyen âge ne paraissent pas avoir eu placer, non seulement au niveau des trois dimensions
l'idée que les textes poétiques pouvaient être rangés en précédemment analysées (créateur, médiateur, destina-
ensembles génériques. Leur réflexion sur la poésie est taire), mais aussi au niveau des multiples combinaisons
toujour fluctuante et leur vocabulaire littéraire, leurs désig-entre elles. En d'autres termes, les éventualités qui se
nations mêmes des différents genres, doivent être maniés présentent au plan du créateur (trouveur/jongleur/clerc/
avec beaucoup de prudence: il n'y a finalement que peu de 'peuple'), peuvent se retrouver au plan du médiateur comme
chose à tirer de leur terminologie." à celui du destinataire (public chevaleresque/clérical/
courtois/'populaire'), multipliant et brouillant ainsi les con-
3 "Une discrimination sociologique des différents types
notations sociologiques du texte singulier."
textuels est donc extrêmement complexe car elle doit se
Table 1. Pierre Bec, Classement typologique des genres (1977-78, 1:38-39)
I.- LE GRAND CHANT COURTOIS [registre aristocratisant

1. canso troubadouresque [O - f] (chanson de croisade1)
2. sirventes ( / serventois) [O - f] (chanson de croisade2)
3. planh [O- f]
4. tenson and jeu-parti [O - f]
IL- GENRES À PERTINENCE THÉMATIQUE [registre popula
A) Registre du "je" lyrique
1. chanson d'ami [F - o]
2. chanson de mal mariée [F - o]
3. aube [O- f] l > chansons de f
4. chanson de croisade3 [F]
B) Registre lyrico-narratif
1. chanson de toile [F]
2. pastourelle [ F- O]
3. reverdie [F]
C) Registre pieux
a) Genres à pertinence thématique:
1. Reverdie [F] (lyrico-narratif)
2. Pastourelle [f] (lyrico-narratif)
3. Aube [O] ("je" lyrique)
4. Chanson d'ami [F] ("je" lyrique)
5. Chanson de croisade3 [F] ("je" lyrique)
b) Genres à pertinence lyrico-formelle:
1. Ballette [F]
2. ROTROUENGE [F]
3. Rondet [F]
4. Motet [F]
5. Lai-descort [F]
6. Canso (chanson à la Vierge - chanson de croisade2
- chanson de croisade4) [F]
D) Registre de l'anti-lyrique
a) Formes lyriques parodiées (burlesque, bachique, obscène, etc.):
1. Ballette [F]
2. Motet [F]
3. ROTROUENGE [F]
4. Canso (sotte chanson) [F]
b) Formes spécifiques (registre du "non-sens"):
1. Resverie [F]
2. Fatrasie [F]
III.- GENRES À PERTINENCE LYRICO-FORMELLE [registre popularisant]
A) Registre lyrico-musical
1. rotrouenge [F - o]
2. lai-descort (troubadouresque) [F - O]
3. lai-farciture (arthurien) [F]
4. motet
B) Registre lyrico-chorégraphique
1. rondet de carole [F]
2. ballette [F] / balada [o]
3. vireli-virelai [F] / dansa [o]
4. estampie [F - o]
Table 1 (continued)
F - f = maximal to minimal frequency in French

Ο - ο = maximal to minimal frequency in Occitan
Genre names in Small caps indicate genre defined mainly in another category but with
affinities in a different category (e.g., Marian song classed in the religious register but
having affinities with the courtly canso). Superscript numbers 1, 2, 3, and 4 refer to four
"classes" of chanson de croisade (Bec 1977-78, 1:153-56).
the other way around, such as in the already mentioned incorporation of the
refrain, a "popular" feature, in "aristocratic" genres (1:40-43). He persua-
sively argued that the songs of the troubadours mostly fall into the aristocrati-
sant register, while those of the trouvères increasingly tended to the populari-
sant register (1:37-38). He asserted that the chanson d'amour was imported
from the south into the north, and that its "courtly" register is shared by the
polemical or satirical sirventes, the dialogue types tenso, jeu-parti, and parti-
men, and the lax and descort (which he sees as northern and southern versions
of the same type).
* * *
The term grand chant court

the concept of amour courto
a study of Arthurian rom
expression "courtly love" it
texts (see Ferrante 1980); on
dour Peire d'Alvernhe (fl.
lines 57-61 (del Monte 195
Mas so non pot remaner, But this cannot continue,

cortez'amors de bon aire, courtly love of good nature,
don mi lais esser amaire, so I cease to be a lover,
tan m'agrada a tener I want so much to dwell
lai on vol sanhs esperitz. there where the Holy Spirit is.
Another occurrence is in the late-twelfth-century narrative Roman de Fla-

menca, lines 1200-1203 (Blodgett 1995, 64-65):
Ε que faria s'us truanz, And what would I do if one vile man,
que. s fenera d'amor cortes who feigns courtly love
e non sabra d'amor ques es, and does not know what love might be,
l'avia messa en follia? would put wild ideas in her head?
If these rare references to cortez amor provide little insight into what
"courtly love" might mean to these poets, scholars today find in the terms
fin amors (Occitan) and fine amour (French), literally "refined" or "pure" love,
much more fertile ground (see Lazar 1995 and Paterson 1999). The adjectives
corteis (French) and cortes (Occitan) and the nouns courteisie and cortezia are
quite common in medieval texts. "Courtliness" can refer to the features of a
luxurious life: fine clothing, food, and drink, skill in such activities as
hunting and hawking, grand feasts with singing and dan

forms of entertainment. "Courtliness" also suggests a vari
qualities: a noble ideal of moral excellence and decorous be
jovens, mezura, pretz, valor; a "courtly" person is honest, loya
generous, pure, even perhaps spiritual.
While love is not necessarily a central feature of "cou
understood today to be one particular expression of a "courtly
Paris wrote in 1883: "love is an art, a science, a virtue, which
those of chivalry or courtliness" (519) .4 Thus was born in the
tury the idea of a system of behavior associated with amour co
love" that seems to find support in the extraordinary list of d
the famous, but controversial, late-twelfth-century tract o
lanus (Trojel 1892; Parry 1941). 5 According to many scholars,
ships between a male subject and a female object, "courtlin
idealism of faithfulness, discretion, worship of the lady from
submission sometimes expressed with analogy to feudal va
sistic preoccupation of the poet with his plight. Some schol
tive connotations in a definition of "courtly love": adultery, s
cism, isolation and frustration in unfulfilled love, secrecy,
social standing but not necessarily moral in behavior.
Although today the idea of "courtly love" continues to
nents, both purposeful and tacit, there have been detractors f
almost from the moment of its first appearance. It has been d
ing from the titles of recent studies) an "invention" of moder
cism, a "myth," and an "impediment" to our appreciation o
The great diversity of its connotations in literature that cross
repertoires, and locales - earlier and later poetry, romance
and south - defies a single, all-encompassing definition. Qu
tinue to generate debate include the medieval origins of the id
love, the problem of how to understand songs that have a fem
a male object (see Burns and Krueger 1985; Pfeffer 2001),
poetic topos of "courtliness" is a reflection of real life and
have to do with the poetry.
As a tool of literary interpretation, the model of "courtline
love" is fraught with pitfalls. Circularity is endemic to studie
• framework of "courtly love," especially as it is applied to the
4 "L'amour est un art, une science, une vertu, qui a ses is less a text on the art of love than "a highly sophisticated
règles tout comme la chevalerie ou la courtoisie, règles contentious subversion of the accepted discourses of
qu'on possède et qu'on applique mieux à mesure qu'on a desire, social hierarchy, and ecclesiastical authority in
fait plus de progrès, et auxquelles on ne doit pas manquer twelfth-century Europe" (Andersen-Wyman 2007, pref-
sous peine d'être jugé indigne." ace, n.p.).
5 Some have argued that Capellanus's treatise is not a 6 Donaldson 1965, Robertson 1968, Huit 1996. See also
serious didactic work but rather an ironic parody. See, for Boase 1977 and Nichols 1998.
example, Monson 1988. A case has also been made that it
the extant poems are themselves the source of our definition, which we then
use to classify songs and genres. A definition must be broad enough to include
all of the widely varied, and not always compatible, topoi found in the poems,
but specific enough to constitute a category. Different critical approaches
may attempt to account for connections among "courtly love" as a theme in
the poetry, genres that incorporate that theme, a taxonomy of genres with
either "courtliness" or "courtly love" as its pivotal delimiter, and "courtliness,"
whether intrinsic to the songs or extrinsic as a factor of the sociological envi-
ronment within which they were composed and performed. Conflicting or
multiple generic labels of individual songs are common, even sometimes
unavoidable, in studies, editions, and indices.
The names that we give to genres are variously medieval and modern.
Medieval treatises and occasionally a lyric text give us many terms familiar
today, including vers, canso/ chanson, sirventes, alba, pastoreia/ pastourelle, planh,
tenso, descort, lai/ lay, dansa, estampida/ estampie, retroncha, rotunda, rondet, and
carole, along with a few that we do not use today, such as gayta, gelozesca, and
sompni. Some terms are found in manuscript rubrics, including jeu-parti and
ballette. Scholars have created other names and categories, such as chanson de
toile, romance, chanson de femme, chanson de mal-mariée, chanson de croisade,
chanson à refrain, chanson avec des refrains, Marian songs, and the like. A chan-
son d'amour is recognizable by its focus on love, but some types, such as pas-
tourelles and dances, also treat of love, in a specific way or with a specific form.
At what point does "registrai interference" - of theme, language, structure -
so obscure a line of demarcation that a generic label becomes meaningless?
What stylistic features are "courtly," "aristocratic," or "popular," and how do
we decide? If a song of excellent language, elevated tone, and sophisticated
versification is taken to be in a "high style," then must we not also include in
that category a chanson de femme, an estampie, or a chanson de croisade with
similar sophistication? Is a canso whose language and structure are quite sim-
ple thereby "popular" or "jongleuresque"? If we consider external factors,
such as function, venue, and audience, would we not consider, for instance, a
dance song that was performed at court to be "courtly"?7
Lyric, narrative, and romance texts, treatises, and manuscripts, which
one might hope would provide hints of how composers and performers
of medieval France and Occitania viewed style and genre, are glaringly vague,
contradictory, or silent on the subject of genre (see Aubrey 1989, 2000). Most
scholars of troubadour poetry agree that the early troubadours had no
clear sense of genre, and even later poets seem to have regarded generic des-
ignations as inexact (see Paden 2000; Pickens 2000). Scholars of trouvère
lyric generally presume that the idea of a system of genres began to take
shape in the later thirteenth century, but clear-cut evidence for this is scarce.
A few composers included reference to a specific genre in the verses of a
7 Page (1986, 38) makes a similar point.
poem, but in most of these rare instances it is not always

meant or whether the label qualifies as a genre as we
For example, Bernart de la Fon's Leu chansonet ad entendre (P
fied today as a canso, but its text refers to it with the diminuti
celebrates its "easy little tune" and declares both poem an
compose, to learn, and to understand (Appel 1915, 301; my
Leu chansonet' ad entendre An easy little song to understand
ab leu sonet volgra far, with an easy little tune, I want to make,
coindet' e leu per apendre graceful and easy to learn
e plan' e leu per chantar, and smooth and easy to sing,
quar leu m'aven la razo, for the subject came to me easily,
e leu latz los motz e.l so. and the words and melody lie easy.
We do not include chansoneta among our pantheon of genres, but might the
self-professed "easiness" of this song (whose melody does not survive) compel
us to place it in a "low-style" register rather than the "high style" that many
associate with the canso}
Manuscript rubrics that include generic labels are random, relatively

rare, and often simply bewildering. Two manuscripts of the songs of lat
thirteenth-century troubadour Guiraut Riquier have rubrics that suggest that
he may have been responsible for supervising the preservation of his songs in
writing (Paris, Bibliothèque nationale de France, f. fr. 856 and f. fr. 22543; see
Aubrey 1982, 86-87, 346-51; Monfrin 1955). Most of the rubrics give labe
for the songs; one song, which because of its theme of awaiting the dawn we
classify as an alba, Guiraut called a serena; his sirventes he named vers, religiou
songs canso, vers, or in one case alba; and he gave no names at all to his pa
toreias, tensos, and planks. How acute was his sense of style and genre? His la
date and affinity with courts in northern Iberia might lead us to expect that
he was at least aware of the taxonomy of genres found in some late Catala
treatises, but even if so, he did not adopt them in his manuscripts.
Fourteenth-century trouvère manuscript Oxford, Bodleian Library
Douce 308 (manuscript I), which contains no music, is famous for its grou
ing and labeling of songs by genre (grant chants, estampies, jeus partis, pastoure
les, balletes [sic], and sottes chansons contre amours). Yet the songs within thes
groups are not entirely uniform in style, form, or content (see Aubrey 1997,
40-52, for discussion of the estampies in manuscript I as well as those in othe
sources). The section of "balletes" is a case in point. The editors of the recent
edition find therein songs that have refrains and others that do not, songs in
the masculine voice and others in the feminine, and a variety of themes (see
Doss-Quinby and Rosenberg 2006, XXVII).
Table 2 provides a snapshot of the contents of the major trouvère man-
uscripts; because of conflicting or multiple generic designations by moder
scholars, as well as occasional duplication of songs within a manuscript, th
numbers are approximate. Most of the books are arranged not by genre b
by author, except for O and C (the latter of which has staves but no mus
entered), which organize the songs alphabetically by first letter of the

incipit; only two anthologies - I and a - present some genres as autonomous
groups. As the following list shows, other manuscripts include some dis-
crete groupings of certain kinds of songs - attributed and unattributed
songs (P, R, T, and X), troubadour and trouvère songs (M), jeux-partis (A,
Q, R, W, and Z, although in Q and R these are given without music), and
motets (M, W, a):
A - chansons', jeux-partis; chansons

I - grand chants; estampies; jeux-partis; pastourelles; ballettes; sottes chansons;
rondets and motets
M - trouvère chansons; troubadour cansos; motets; lais

Ν - attributed trouvère chansons and troubadour cansos; unattributed
trouvère chansons and troubadour cansos; lais and motets enté
Ρ - attributed chansons; unattributed chansons; attributed chansons
Q - chansons; jeux-partis (without music)
R - attributed chansons; jeux-partis (without music); unattributed chansons
Τ - attributed chansons; lais; motets; unattributed chansons
V - chansons; Marian songs
W - chansons; jeux-partis; rondeaux; motets
X - attributed chansons; unattributed chansons; Marian songs
Ζ - chansons; jeux-partis
a - chansons; pastourelles; motets; rondeaux; Marian songs; jeux-partis
The libellus ("little book," here meaning a manuscript devoted to the works of
a single composer) of Adam de la Halle (W) is exceptional in separating ron-
deaux from chansons, jeux-partis, and motets; in every other manuscript one
finds stray pastourelles, jeux-partis, and other so-called low-style genres scat-
tered among high-style chansons d'amour. It seems clear that most scribes did
not place a high priority on identifying a song by its genre, or a genre itself as
an exclusive group.
* * *
While littéraires continue t

and "courtly love," most m
terminology and the dichot
pins. Citing both Dragonett
study of the monophonie so
"high courtly songs"
all belong to the imagined, i

to his perennially inaccessib
length. . . . They are artefac
formal ingenuity whose pro
tained the aristocratic social
S ■* - ο 2 ο
<ί ιΜ (Π (Λ ιθ IM ί^ τ^ ^ Η γ-η 1> tJh (M .J2
■2 '
* » - - - ff
* s -ti
ϋζ CO ^Ο ©Ο ^Η τ-Η Γ-Η '■^ *^*
£3 C0J>(M (Μ Φ Ο Ι> ί^ γη ^ C5 Ο s? ι-ί

0*ι cotomr-icoojcoi- ι cm cn cn οο η ca τ- ι ι- ι -^ s? „
ο ao ^ δ
^ m ^ m η η οο σο γ-η οο ι- ι ι- ι '53^^
ν- ^ eu
Q CM Í- »CM I Q '>
CL (N iC tû 00 ,-ηοΟ^η^ i>OCM CM Φ 00iO<7> "^S^
^COCMCOr-H ,_|,_|,_H,_| Î^r-HCMCMlOrHr-H r-HCMi> ggO
OcMr^ en oo(M<r>2 m ^ h σι b h co | § J
Sr-H T}H CMÍ^ M Φ nil
«^r-H-^ I> ΙΟ (û 00 i- I CM CM r-H COCM COr-Hi- I ^^^
• S ■«*· ν<50"
Oi>COiO en <£><£> 00 ,S -^ Ë
tO CM (N T-H- r-H r^Cji r-H cO^^CM r-H <D ^ ft- S
S ft- ^ s
«°h ζ£> en CD OCO O b CO CM W I> O O^S
Í Ü rfiOH0050íf)«b^ xf5 CN b CO CO cDCTiCO 43 § .
S*>i
1
cO Ô J> m
1
CO X>
δ
OI>cOr-H CM
S
00 η
r^ S mCMCMOOOOcDr-H^T^ Ο ^ φ. XOWCM^tû^b ^^g η
Ο Ο COCOCDtM^WCOCMCM CMCMr-HCMrHCMCOr-HCMCDr-H §s G
.S s 4 I 2
•S -S § & § -δ á
^ S ^ S s S ^ë -S ^ s^ ë S § s ||. g g g g g 3
(Λ
-s S §
I
C
(0 11 1 1 1 l|ö
Ε
£ i - s stg t! i. s ^ **î
>
1 il il Bin r nil ϊίίίϊϊ* ι ι
I
1-
a, <Jh1(hZ<<!phP3 clhPh<<;h-i<;<<<!<!< 'ö.S^
3 5^^ ^ « 33BBgp<^
c
g SSS §§ « S S S S =§ s M
c
φ COC5 COwiSoOo OOr-Í^H-Sor-HCOCOOOcOGO 2O-§
σ>
'S
c
i S^gsSSSîl S a f g ïï i -s -s g -s -s Isa
ο i-Jh O ^
+3
3
.Ü
u .S 'Ι-
1 la" -S 1 δ
ο
(Ν
_α>
.Ω
He offers Adam's Tant me piaist vivre en amoureus dangier (RS 1273), given here
as Example 1, as representative of these characteristics (all transcriptions are
my own). A graph of the poetic and musical structure of this song indicates a
poem of heterometric verses (ranging from five to ten syllables) and three
rhyme sounds, which are the same in all stanzas, a strophic construct known
as coblas unissonans. In the melody the first two verses are repeated, ABAB,
matching the alternation of rhymes, abab; after this the music proceeds with-
out repetition, and not matching the rhyme scheme (a prime mark in the
rhymes row indicates a paroxytonic rhyme; "+1" in the syllable count row indi-
cates the final unaccented syllable not included in the syllable count):
Melody ABABCD Ε FGH
Rhymes a b a b b c' c' a a c'
Syllable count 10 10 10 10 10 5 + 1 7 + 1 7 77 + 1
Stevens found in the poetic structures of such songs a "'harmoniousness' -

not mere euphony, but a positive balanced structural accord in sound," and
in the language a conventionality typical of the courtly rhetoric found in all
trouvère love songs (15). The melodies, he argued, are based on a convention-
ality similar to that found in the texts: correspondences between musical
phrase and verse length and between neume and syllable, overall structures
that follow the scheme AAB, and stepwise motion and formulaic melodic
figures. But this conventional elegance is articulated by "inexhaustible inven-
tiveness and sophistication" (25), subtle divergences from expectation that
might elevate it to a "high" style, such as the discord between musical phrase
and poetic rhyme, or unexpected tonal goals. Note, for example, that verse
6-9 of Tant me plaist end on four different pitches, none of them the C or F
established as tonal goals at the ends of the first four verses; only in the final
verse does the melody come to rest on F.
The single exception to this "high" style among Adam's thirty-six mono-
phonic chansons, Stevens claims, is Amours m'ont si doucement (RS 658; se
Example 2), a song that some literary scholars have labeled a chanson de femme,
partly because in one verse in one of the three extant readings the beloved is
referred to in the masculine (although another verse in the same reading uses
the feminine; see Marshall 1971, 126-27). The melody of this song, says Ste
vens, is "lighter" than the others, in "a more 'popular' courtly style" with
"simple structure, simple pattern and simple balance" (18).8 A single rhyme
for all verses (-ent), recurring in all stanzas, is the simplest of poetic structures,
and the overall effect of "balance" and "symmetry" that Stevens identifies is
apparent in the melody of the first five verses, which are largely repeated in
the last five verses:
8 Here and more expansively in a book published a in contrast to a mensurally regular performance of "lower
decade later (1986), Stevens argued for an isochronous style" songs, especially dances'.
rhythmic interpretation of "high-style" monophonie songs,
rhymes music
1 (CO - - 9 * m u m m * *V> : a A
* Tant me plaist vi - vreên a - mou- reus dan - gier
2 ^P ~ * * m m m m * m * m m *[W] I
qu'a pain - nés ai pen - see a guer- re - don,
° si ne chant pas pour mes maus a - le - gier,
4 fk' w m + a^l m _ m m + #1 ^7^- ι - b B

ν y ~ m _ * * * 'm ο" ' -
"ψ ν y m 'm ο" ' "- J "- J
quar je n'en plaing fors la ma - le foi - son.
5 £ m ^0 m ■ · β ^m *~, ^' m^m(.) = b c

° Tra- vail qui plaist ne doit on par rai - son
-$s- * · Ο _-^ ~ v
con- ter pour ha - schi - e:
7 fe- »^»* *~, » »~« "*^<;> c' E

* li mal d'à - mou - rou - se vi - e
8 <k> ' ' ' ' ' "^»' ; jP a F

ne me font fors cha - ti - Hier
9 (jk # , » * - - · <·,> a G
° de joie et de de - si - rier,
10 (fe * * « # · * *"· »^ II c< h

sans pen - ser nu - le fo - li - e.
Example 1. Adam de la Halle, Tant me plaist vivre en amou
Melody Α Β B' C D Α Β Β' C D'

Rhymes aaaaa aaaaa
Syllable count 77774 77776
Yet one might hear in this melody some of the

tion" of a "high-style" song. Observe, for exam
the contour of verse 1 a fourth higher, and ver
beginning a step higher than verse 1, convergin
melisma (D-C-B) on the fourth syllable, then
lowest pitch up to this point, F. Cadential motio
last two syllables of verse 3 are a step lower tha
is reflective of the cadence of verse 7; and w
ascends to lead into verse 6 on the same pitch (A
labeled "D" in verse 10 "closes" the melody with
even more final because it reinforces the cadential motion at the end of verse
9. These subtle variations, along with the floridness of the melody, might be
audible only in the "ear" of one particular hearer, but that one scholar can
notice their presence while another passes them over merely highlights the
subjectivity that is endemic to style analysis.
Christopher Page has argued persuasively for the framework of "High
Style" and "Low Style" (he capitalizes the terms) and for the centrality of
genre to understanding the composition and performance of medieval songs
in France and Occitania. In a series of articles and books (e.g., 1984-85, 1986,
1989) he has surveyed an important but hitherto neglected body of
evidence - nonlyric texts such as epics, romances, and chronicles, and cer-
tain treatises of poetry and music - and brought them all to bear on the ques-
tion of style and genre and especially the performance of medieval song.
Page readily acknowledges that these texts provide conflicting and ambigu-
ous impressions of the songs and their composers, performers, and hearers.
Although he pronounces all evidence as important, "since," as he puts it, "so
little of it survives" (1986, 40), Page discards "vague" evidence and focuses on
those "few sources which refer to different kinds of songs in contexts where
it is clear what kinds of songs are meant" (1986, xi). But one could argue that
vagueness itself is evidence of an ambivalence, perhaps even disregard, that
medieval authors had toward marked distinctions among different kinds of
songs.9
9 Page also interprets the absence of any allusion to of relying on a late source as a reliable witness (e.g., the
instruments in a text as evidence that instruments wereearly-fourteenth-century Leys d'Amors; see Page 1986,
250-52n6),
not present in the particular situation described in the text. but on occasion he ignores the dates and prov-
In such a case, though, as I have noted elsewhereenances(1989, of the manuscripts of the texts on which he relies,
as in his quotation of verses from a poem that is generally
131 n64), this is an argumentum ex silentio - no evidence
is no evidence one way or the other. He also sometimes dated to the thirteenth century, Gille de Chyn by Gautier de
fails to provide information about his sources thatTournai,
might but whose sole surviving text is in a sixteenth-
affect its interpretation. For instance, he cites the century
dangers manuscript (1986, 30).
Λ Ι
'J
1 λΓ~ m m
-W* m
° A - mours m'ont si dou - ce - ment
2 aI« * · *^''' ^ , B
* na vre que nul mal ne sent,
3 Û* ~ ' " ·- " *^7- , B'

•J I l
° si ser - vi - rai bon ne ment
4 A* m * =' ' * ± ./'*~m' _ ) c

* A - mours et ma ^/aouce a - mie,
° a cui me rent.
6 L » » # " * * » Ξ Α
* Et fas de men cors pre - sent;
= #=- B
_ ^
* ne ja mais por nul / tor - ment
■ç σ ;
σ e j ai - e η îere au - tre - ment
' ===f~
^^ ^
# * ν * » # » y c
8 ains wel u - ser mon jou - vent
* en a - mer loi - al - ment.
Example 2. Adam de la Halle, Amours m'ont si d
Although most scholars of medieval literature who have studied genre

have concluded that specific types were not clearly distinguished during the
Middle Ages, Page, like Stevens, presupposes that they were and that compos-
ers and performers were careful to respect the distinctions. This is a crucial
point, since this presupposition allows Page to establish a more or less con-
crete framework from which to extrapolate stylistic features of one genre or
another. As some late treatises do (see below), Page delineates genres by their
subject matter and categorizes the love songs of the troubadours and the
trouvères, along with the sirventes, tenso, partimen, and planh, as "High-Style"
songs, while dances, pastourelles, lais, descorts, jeux-partis, chansons de toile, and
narratives like cansos de gesta he views as being in a "Low Style." He distances
his interpretation of genres from a strict adherence to the aspects of "courtli-
ness" that concern venue, class of the audience, and participants:
A Lower Style song with a simple, refrain-based melody, sung to the fiddle for
dancing at court, would not be an "uncourtly" song; indeed, the fresh and pri-
maveral ethos of most aristocratic dancing lay very close to the essence of court-
culture in the twelfth and thirteenth centuries. . . . Courtliness is not the main
issue here, but art. While any simple dansa could be as courtly as a High Style
troubadour song, its appeal was not to good taste or judgement, but to the feet.
In the same way narrative songs, whether in the form of epic or lyric like the
chanson de toile, catered for a basic human desire - the desire for stories - in a
way that the High Style songs of the troubadours and trouvères resolutely
refused to do. (1986, 38)
Thus, Page removes consideration of the location or the status of the perform-
ers or the audience as determinative of a song's style, but at the same time
expands the definition of a song's theme to include performative elements -
on the part of the musicians, but also that of the listeners who reinforced the
generic identity of a song by how they responded to a specific type when it was
sung. A song that invited the hearers to tap their toes or leap from their seats
to dance, or to sit and listen to an absorbing tale of feats and heroism, was in
a "Lower" style than one that made them think deep thoughts about matters
of life, art, emotion, politics, religion, or philosophy.
Having defined the genres and the "styles" into which they fall, Page
(again like Stevens) draws from songs that fall into one or another category
a list of internal poetical and musical features that distinguish "High Style"
from "Low Style," a typology to which he subsequently appeals as support for
the classification of a specific song as falling into one group or the other. In
his typology, the stanzas of "High-Style" songs tend to be isometric (all verses
with the same number of syllables; note that Stevens's example of a "High-
Style" song, Example 1 above, is heterometric) and only rarely have a refrain;
the beloved is unnamed or provided with a pseudonym (Occitan senhal); the
melody is "rhapsodic"; internal relationships such as repetition are disguised
so that the listener only gradually grasps them; and focus is on the je - the
singer. "Low-Style" songs more often have heterometric stanzas, often with
markedly short verses, and refrains are common; the beloved

the melodies have "conspicuous short-range repetitions" an
with obvious internal relationships; and for dance songs, th
dance, the community, not the individual singer.
As an example of a "High-Style" melody, Page 1993 g
Marvelh's Las grans beutatz e.lsfis ensenhamens (PC 30,16; Exa
melody, according to Page (1986, 13-15), is "rhapsodic," w
"disguised" relationships such as the "closed" ending on D of v
by an "open" cadence on Ε in verse 4, and the resemblances am
and 6 (labeled A and A') and 2, 4, and 7 (labeled B, B', and
that the hearer does not perceive these relationships imme
the "coherence and focus in the rhapsodic flow of the m
emerge as the song proceeds.
These subtle interconnections contrast sharply with th
short-range patterns" of melodies in the "Low Style," as an ex
Page gives the anonymous C'est la gieus en mi les prez (not liste
300 in Linker 1979; rondeau 16 in van den Boogaard 1969,
scription of this song (see Example 4) is a modern re-cre
derived from one created by Friedrich Gennrich some deca
10). The full text of the single-stanza rondeau survives only as
without music, in one manuscript of the Roman de la Rose ou
Dole (Rome, Biblioteca Apostólica Vaticana, Reg. 1725, fol
source for music is at the end of the motetus of the motet Tout
Inseculum (Wolfenbüttel, Herzog-August-Bibliothek, Guelf. 10
222, and Montpellier, Bibliothèque Inter-Universitaire, Se
H196, fol. 259v-260), where only the rondeaus refrain verses (
in italics and labeled A and Β in Example 4) are quoted; Ge
conflate and adapt the Wolfenbüttel and Montpellier readi
(using the mensural notation as found in the Montpellier
this reconstruction to the other verses, following the princip
cal structure parallels the poem's metrical scheme,10 hence:
Melody Α Α Α Β Α Β
Rhymes a a a b A B
Syllable count 7 7 7 5 7 5
Example 4 gives the music of the two verses at the end of th

Wolfenbüttel and Montpellier sources (notice the Bb in t
Gennrich and Page ignore) followed by the conflated and
Gennrich and Page. Their interpretation is plausible at lea
of view of the classic rondeau structure of the fourteenth ce
that it was preserved as a full text only in a source without m
10 Tischler (1997, 14: rondeau 20) gives both ma

versions.
music
m) * Φ ·"" ·~ΊΡ
° Las grans beu-tatz e.ls fis en se - nha - mens
2 vy
ΠΤΝ « ^^« ~
· ^ « ^ «
^ m Λ _"""' Γ_
^ 1 _ ^, ^,
ι ^' ~~ B
·) -*■ » · ^ Λ ^ » V^ _ -#- *y _ ι "closed"
e.ls ve - rais pretz e las bo - nas lau - zors
jO , _ _
3 (fo m m φ , m m m ^ '"I*· _ "9 J~*K _ - A
^Y m m * ^ Φ 9 + "9 * Φ φ -
* e.ls au - très ditz e la fres - ca co lors
4 fts φ φ φ J~*^ f ,^ ' B'

ύ -Φ- 9 * W w J "open"
que vos a - vetz, bo - na do - na va - len,
5 WP *^^ ###^#^#* φ * φ μΦ á^=-. c

° me do - nan genh de chan - tar e sei - en - sa,
mas gran pa - or m Ό toi e gran te - men - sa
7 (fr> , , , , .""; _ -τ*. ~z- B"

qu'ieu non aus dir, do - na, qu'ieu chant de vos
É_
II ^
e re no sai si m'es o dans o pros.
Example 3. Arnaut de Marvel h, Las grans beutatz e.ls fis ense

22543 fol. 52
Wolfenbüttel 1099, fol. 222
8 A - mor ai a ma vo - len - te te - les con ge vueill.
. Montpellier H 196, fol. 259v-260 ^
φ Ι' ι
J'ai a-mours a ma vo - len - te te - les corn je wiel.
Γ ' Ι
A Gennrich, Page
1 <fe Ρ Γ f# ^ Γ Γ Γ-Γ
* C'est la gieus, en me les prez,
2 L· r r r r y Lr r m « a
* j'ai a - mors a ma vo - len - te,
3 l[ r r cr r Lr ρ ^= »
" da - mes i ont baus le - vez.
4 | r r Lr r Lr
Ga - ri m'ont mi oel.
• l·1 r r r r i' üt · -
° J'ai a - mors a ma vo - len - te
- l'- r [ r r cj^î b '

te - les com ge voei.
Example 4. Anonymous, C'est la gieus en mi les prez (Linker 19
music is found only as a two-line refrain within the context of a motet, sug-
gests a life as an elusive dancing song, not an expressive, deliberately com-
posed love song.11 This melody as transcribed certainly fits Page's description
as simple and "tuneful," but it should be noted that strict rondeau form was
not stable in the thirteenth century, and to create such an extrapolation
depends upon, among other things, a preconception of the style and form of
a song with a refrain.12
Yet many songs in the troubadour and trouvère repertoires defy Ste-
vens's and Page's classifications. Example 5 gives the first stanza of a self-
styled dance by mid-thirteenth-century troubadour Guiraut d'Espanha (PC
244,1a), which calls itself a dansa in the final stanza.13 In Page's view, this
dance should be a "Low-Style" song, yet contrary to his characterization of
dances, it is isometric; each of the three stanzas has its own rhymes (coblas
singulars), except for the last four verses, whose rhymes and music match
those of the refrain, as this graph of stanza 1 shows:
Melody Α Β C D Ε C F Ε' C F' A B C D

Rhymes ABBAc c de c dABBA
Syllable count 77777777777777
The song also has three four-line tornadas

cally found in many cansos. The transcriptio
of the manuscript.
One might compare this song with a chans
century trouvère Gace Brulé (RS 1757), given
chansons, like this one, have refrains (Pag
withstanding), although here the refrain rhy
the stanza. The poem uses the sophisticated c
metric (again contrary to Page's characterizat
the contrast between the oxytonic six-syllab
a' rhyme, giving the sonic effect of seven syl
than the graph suggests:
11 Page draws striking conclusions about

12 See, for example, Doss-Quinby the2006,
and Rosenberg perfor-
LXXXIII; for discussion
mance of lyric songs, especially their of arhythm
similar example, seeand
Aubrey the use
1997, 10-14.
of instruments. "Low-Style" Example 5 has
songs, he structural
posits,features of the later
might have
been sung to instrumental virelai (AbbaA), although, again, not as consistent
accompaniment, and asthe they dance
wouldrhythms,
types, at least, had regular be in the "fixed" form while
of the fourteenth century.
"High-Style"
songs were sung without instrumental accompaniment
13 Hoby 1915, 38-41. See Aubrey 1996, 124-26. This
and in a nonmeasurable rhythm. Near the end of the thir
song is extant in only one source (M, fol. 186-86v); it was
teenth century, Page argues, a new scholastic environ-
added on a blank page sometime after the original layer
ment took root in Paris, and a new ethos of "High Style"
had been copied, but probably by the end of the thirteenth
and "Low Style" developed, in which playing musical
century.
instruments became part of a literate cleric's accomplish-
ments, and performers began to sing "High-Style" songs 14 Text from Rosenberg, Danon, and van der Werf 1985,
to instrumental accompaniment. 66-70. Gace's chanson survives in the same manuscript to
which Guiraut's dance was added (M fol. 28v-29). Both
were written with mensural notation, although Gace's
melody is part of the original layer.
A B
1 iten vol - gra s 'es -ser po - gro, 2 ca - mors si gar - des d'ay - tan
'c' D
TO F$L
Γ * # J ' ■ J J'#F
/^1 * ^ J ' Jf^
d + d 7 ' *
3 ^m^ non fe - ses fin ay - man 4 chau - sir en luec que. I pla - ges.
φ J «Π J 1' ^p ft Γ 7 ρ p J ji /J
5 Ε per que? car per pla - ser 6 qu'ieu ere - sia de vos a - ver,
λ<^
~ψ I
7 don - na, vos mi fes chau -sir 8 a - mor, don a - via es - per,
'Ç' <F>
ί Ρ ft J Ji J
9 que mi de - ges - ses va - 1er 10 del joy don ieu tant sos - pir;
A B
(cyJ LJ* J^^ ^J ? J ^jJ ^ ^^J 7=

liar m'a - ves a tal punchmes 12 que tot iorn vauc de - si - ran
{c' D
I r ^
13 la mort don ay do - lor gran, 14 car non faitz so c'a - mors fes.
Example 5. Guiraut d'Espanha, Ben volgra s'esser poges (PC 244,1a), M fol. 186-86
Melody A B A BCDE FGH

Rhymes a' b a' b a' b a' b C C
Syllable count 6 + 1 6 6 + 1 6 6 + 1 6 6 + 1 6 8 8
The melodies of these two songs are similar in their mostly syllab
ture, range of about an octave, and conjunct melodic motion. Gace
(Example 6) follows the typical ABABx form, and the rhyme sch
musical phrases agree in the first four verses, features that Stevens a
with a "high" style, but no obvious "divergences from expectation" st
ear. The first four phrases begin similarly, and the cadences are
Verse 5 begins like verses 2 and 4 and leads predictably upward t
on C in verse 6, followed by a gradual descent to the octave below at
of verse 8. The ends of verses 5, 6, 7, and 9 move smoothly into the f
verse on the same note. Verse 10 reflects the contour of verse 1,
notably the nudge up to Bb, seeming to round the melody to a cl
might hear simplicity and balance in this chanson d'amour such a
associates with a "low-style" song.
Guiraut's dance (Example 5) on first examination appears to be
through-composed, except for the presumed repetition of the four refrain
verses between stanzas. In Gace's song, since the music of the refrain verses
(9-10) does not match that of any other verses, the repetition of the refrain
is heard only as a reiteration of the text; the refrain of Guiraut's song (verses
1-4) uses the music of verses 11-14 and thus draws attention to it as well as
to the words. But "obvious internal relationships" such as Page expects to
find in a "Low-Style" song are difficult to detect in this dance; rather, as the
song proceeds, one can hear more subtle interconnections, for example,
verses 6 and 9 (labeled C') as a variation of verse 3 (labeled C), verse 8
(labeled E') as a more noticeable reflection of verse 5 (labeled E), and verse
10 (labeled F') as a "closed" version of verse 7 (labeled F). These reworkings,
which are deployed asymmetrically as the form progresses, might qualify
as the sort of "inexhaustible inventiveness" that Stevens associates with
"high-style" melodies, but as analyses such as these demonstrate, a percep-

tion of style is subjective.
* * *
As noted above, literary a

from medieval sources, part
of the later twelfth throug
a summary of the authors,
Latin works are mainly fro
from the south. The readers of the Latin treatises were scholars and clerics,
students of literary composition who were familiar with the ancient arts of
grammar and rhetoric - Horace, Cicero, Quintilian, Donatus, Priscian, the
_ rhymes music
1 fej ]) J. J1 kfl JJ J. a' A

Quant li tanz ra - ver- doi - e
2 |j J1 j. j j j,
con - tre le tanz Pas- cour,
3 (L Jl J. J^ k J"J J. Ξ - A
•I r
vis m'est que chan - ter doi - e
a - près ire et do - lour,
5 £j J1 j~ j j Ji i1
* donc tant a - voir so - loi - e
6 jpj * ρ» ρ -J> J. b D
pour fur- nir fine a - mour.
7 (fcj Jl J. J. J^ J"]
* Mes se ma dame a - voi - e,
8 |j ji j- iTj i'j. b f
as - sez a - vroie ho - nour,
9 |j Jl J· J Jl J
° g'î/^ mes cuers ne veut nule a - voir
- L· >p ^ j J7j. ^ c «
ybrs /i, cite n'tfn áaz- ^72^ cão- /ofr.
Example 6. Gace Brûlé, Quant ΙΊ tanz raverdoie (RS 1757), M fol. 28v-29
Table 3. Theorists
Date Author Title Provenance
Before 1140 Hugues de Saint Victor Didascalicon: de studio legendi Paris

c. 1159 John of Salisbury Metalogicus France
c. 1170-75 Matthieu de Vendôme Ars versificatoria Tours, Orléans,
Paris
c. 1181-86 Andreas Capellanus De amore or De arte honeste amandi Paris?

c. 1190-1213 Raimon Vidal Las Razos de trobar Catalonia
c. 1208-15 Geoffroi de Vinsauf Poetria nova France and England

(Northampton)
After 1213 Geoffroi de Vinsauf Documentum de modo et France and England
arte dictandi et versificandi (Northampton)
c. 1208-13? Gervasius de Melkley Ars versificaria Rouen? England?
(de Salta lácteo)
c. 1225-45 Uc Faidit Donatz proensals Italy
1229-35 Jean de Garlande Parisiana poetria Toulouse and Paris
c. 1260 Brunetto Latini Li livre dou trésor Florence
c. 1260 Brunetto Latini Rettorica Florence
Before 1250 Eberhardus Alamannus Laborintus Paris/Bremen/Orléans
or 1280
1288 Matfre Ermengaud Breviari dAmor Occitania

c. 1289-91 Jofre de Foixà Règles [de trobar] Sicily (Catalan court)
c. 1290 Jofre de Foixà? Doctrina de compondre dictats Catalonia (Sicily?)
After 1310 Guilhem Molinier et al. Leys dAmors Languedoc
c. 1302 Dante De vulgari eloquentia Florence
c. 1300 Johannes de Grocheio De musica Paris
Rhetorica ad Herennium - and they were exhorted to imitat

Virgil to the writings of contemporaries, including the the
Their discussions are couched in the traditions of the scholastic milieu of the
cathedral schools (Kelly 1991, 41-44, 47-57). 15

In the classical treatises on rhetoric, the issue of style concerned mainly
the use of language, the figures and colors, which were grand, medium, or
simple; the distinctions were carried over into the medieval arts of poetry and
prose and were to be placed in service of the theme, or "material." According
to grammarian Jean de Garlande, "the level of style itself amplifies the mate-
rial, when high sentiments are chosen for the high style, middling ones for
the middle style, low ones for the low style" (Lawler 1974, 78-79). 16 In their
treatment of genres, these Latin theorists concerned themselves with the wid-
est spectrum of Latin literature: tragedy (couched in the language of ornatus
16 Lines 416-18: "ipsius stili ampliat materiam quando ad

15 Kelly asserts that the treatises were used "for class-
grauem stilum graues eliguntur sentencie, ad mediocrem,
room instruction and illustration" (1991, 60; see 89-96 for
full discussion of the authors and their readers). He divides médiocres, ad humilem, humiles." This is not the Johannes
the treatises into groups that focus more on the rhetorical de Garlandia familiar to musicologists as a later thirteenth-
tradition, the grammatical tradition, and a more elemen- century theorist of Notre Dame polyphony and notation.
tary level of teaching; see 1991, 63, 107-9. See Reimer 1972, 1:12-17, and Pinegar 1991, 96-102.
difficultas and concerned with the noble class), comedy (in the o
and for middle and low classes), and so forth, and to a limite
song. In the latter category, they focused on metrical poetry in
tradition; the later treatises attended to the medieval tradition o
and accentual poetry. The styles of language could be mixe
genres.17
In medieval teaching, the stylus materialis, or material style, was broad-
ened to include, as Douglas Kelly puts it, "the quality of persons and things
depicted" (1991, 71), or various facets of the subject matter of the poem - the
personages (nobility, middle class, low class) by whom, about whom, or for
whom a poem was composed (see Kelly 1991, 71-78; Aubrey 1996, 68-70).
These, according to Jean de Garlande, fall into three classes:
Three kinds of characters ought to be considered here, according to the three

types of men, which are courtiers, city dwellers, and peasants. Courtiers are
those who dwell in or frequent courts, such as the Holy Father, cardinals, leg-
ates, archbishops, bishops, and their subordinates, such as archdeacons, deans,
officials, masters, scholars; also emperors, kings, marquises, and dukes. City
dwellers are count, provost, and the whole range of people who live in the city.
Peasants are those who live in the country, such as hunters, farmers, vine dress-
ers, fowlers.18 (Lawler 1974, 10-11)
Jean describes the subject matter as either "honorable" or "disreputable," the

language of the former being straightforward, of the latter circumlocutious:
There is honorable subject matter and disreputable subject matter. In honor-

able subject matter use plain sentences and words that put the case in the open.
Disguising disreputable subject matter calls for subtlety. . . . That means touch-
ing on the issue with various circumlocutions, which will keep the disreputable
subject from showing through.19 (20-21)
Treating material style in a similar way, literary theorist Geoffroi de Vinsauf

enjoins the composer to fashion his language to accord with the subject mat-
ter, but also to take into account the hearers:
17 See Jean de Garlande, for instance (Lawler 1974, reges, marchiones, et duces. Ciuiles persone sunt consul,
86-87, lines 53-54): "potest et humilis matéria exaltari, ut
prepositus, et cetere persone in ciuitate habitantes.
in graui matéria coli muliebres uocantur 'inbelles haste'"
Rurales sunt rura colentes, sicut uenatores, agricole, uini-
(low matter can be exalted, as when in a treatment of a tores, aucupes."
high subject women's distaffs are called "the spears of
19 Lines 317-23: "sicut dicit Tullius, 'Est genus cause
peace"). See Kelly 1991, 79-80.
honestum et turpe'; sic est matéria honesta et matéria tur-
18 Lines 124-32: "Tria genera personarum hie debent pls. In matéria honesta utendum est sentenciis planis et
considerari secundum tria genera hominum, que sunt curi-
uerbis materiam declarantibus. In turpi matéria, si velimus
ales, ciuiles, rurales. Curiales sunt qui curiam tenent aclatere, vtendum est insinuacione . . . qua propositum qui-
celebrant, ut Dominus Papa, cardinales, legati, archiepis- busdam tangimus circumlocutionibus quibus manifestari
copi, episcopi, et eorum suffraganei, sicut archidiaconi,non poterit turpitudo."
decani, officiates, magistri, scolares. Item, imperatores,
Regard not your own capacities, therefore, but rather his with whom you are
speaking. Give to your words weight suited to his shoulders, and adapt your
speech to the subject. ... In a common matter, let the style be common; in
specialized matters let the style be proper to each. Let the distinctive quality of
each subject be respected.20 (Gallo 1971, 72)
In further comments about oral delivery, Geoffroi mentions such

performative - but probably not musical - techniques as a "caustic voice," "an
inflamed countenance," and "turbulent gestures" to reflect a poem's meaning
(Gallo 1971, 124).21
Geoffroi 's Poetria nova was by far the most widely disseminated of the
Latin treatises - today extant in at least 185 manuscripts from all over Europe,
as well as more than eighty glosses and commentaries. The other treatises
seem to have been less widely read.22 The short tract of Eberhardus Alaman-
nus survives in forty-three manuscripts, most of them from Germany, while
that of Matthieu de Vendôme is found in nineteen extant sources. Gervasius
de Melkley's treatise survives in only three manuscripts, and Jean de Gar-

lande's in only six. It is impossible to determine whether the trouvères knew
of these treatises or studied them, but the works certainly cannot be inter-
preted as being descriptive of the trouvères' art, as the teachings are devoted
to the composition of Latin literature. And as Quadlbauer (1962, 71), Kelly
(1991, 85-88), and others have pointed out, the theorists drew their exempla
from "small forms" - sentences, couplets, single stanzas - not from the long
strophic structures created by trouvères or even by the authors of Latin versus
and conductus such as are associated with the Aquitanian and Parisian schools
of the twelfth and thirteenth centuries. Further, these writings teach careful,
deliberative, intentional authorial creation, whether devoted to the composi-
tion of works falling in a low, middle, or high style. The dances, rondeaux, and
caroles that musicologists discuss as being representative of a "low style" were
invented in a social context removed from the sophisticated authorial process
that these treatises describe, and thus can hardly be seen to be subsumed into
the systems that they describe.
Vernacular treatises, all from the south (mostly Catalonia and Italy, not
Languedoc; see Table 3), started appearing in the late twelfth century. In
20 Lines 1089-91, 1096-98: "Proprias igitur ne respice flammae materque furoris, ab ipso /folle trahens ortum,
vires, /immo suas, cum quo loqueris. Da pondera verbis / cor et interiora venenat; /pungit folle, cremat flamma, tur-
aequa suis humeris, et pro re verba loquaris. ... In re com-batque furore, /exit in hac ipsa forma vox fellea, vultus /
muni communis, in appropriatis /sit sermo proprius. Sic accensus, gestus turbatus" (English translation in Nims
rerum cuique geratur /mos suus" (English translation in1967,90).
Nims 1967, 55).
22 See Kelly 1991, 96-101, and 110-19, for discussion of
21 Lines 2041-50: "Domes ita vocem, /ut non discordet the varied dissemination of these treatises. A thorough
study of the manuscript tradition of any of these texts has
a re, nee limite tendat /vox alio, quam res intendat; eant
yet to be conducted.
simula ambae; /vox quaedam sit imago rei; res sicut habet
se, /sic vocem recitator habe. Videamus in uno. / Ira, genus
contrast to the clerical audiences of the northern Latin treatises

ences were lay amateurs, bourgeois and aristocratic, aspiring
lived outside the land of the troubadours. The later of these
taxonomy of genres of medieval songs, although squaring th
of specific genres with surviving songs is often problematic (see
70-73; 1997, 34-52; 2000). For these theorists, subject matter
polemic, pastoral narrative - is the salient distinguishing featur
but other relevant factors have to do with structure and format,
the presence of a refrain or arrangement as a dialogue; in on
treatises is music a salient feature in the description of genres.
Raimon Vidal, the Catalan author of the earliest vernacular treatise,
Las Razos de trobar, was addressing an aristocratic audience of Catalan and
Occitan natives (see Marshall 1972, lxvi-lxxi, on manuscripts, dating, prov-
enance, authorship, and audience). The text is extant in three manuscripts
from Italy and Catalonia from the late thirteenth century to the late four-
teenth. Raimon devotes more attention to grammatical features of Occitan
(and to a limited extent French) than he does to concepts (e.g., razo or theme)
or mechanics (e.g., versification) of composing poetry. He does not present a
taxonomy of lyric genres, nor does he mention music. The principal generic
distinction that he makes is between "cantar" and "romans" - that is, lyric and
narrative. In the context of a discussion of the relative merits of the French
and Occitan languages, the latter of which he declares to be excellent, c

rect, and authoritative (Marshall 1972, 4 and 6),23 Raimon proposes th
French is better suited to romans and pastourelles, while Occitan is better fo
vers, cansos, and sirventes (6).24
At the beginning of his treatise Raimon colorfully describes the ubi
uity of song singing, regardless of class, social or economic status, or occupa
tion. He credits the troubadours with capturing every conceivable topic that
might be put into song, claims that their art is "the driving force behind al
valor," and declares that everyone - aristocrats, clergy, pagans, bourgeo
and working classes - appreciates the songs of the masters:
All good Christians, Jews and Saracens, emperor, prince, king, duke, coun
viscount, commander, vassal, cleric, citizen, and peasant, small and great, dail
give their minds to composing and singing, by either inventing or listenin
23 Lines 59-60, 74-76: "Totz horn qe vol trobar ni enten-

24 Lines 72-74: "La parladura francesca val mais et [es]
dre deu primierament saber qe neguna parladura non
pluses
avinenz a far romanz et pasturellas, mas cella de Lem-
naturals ni drecha del nostre lingage. ... Et per totas
osin
lasval mais per far vers et causons et serventes" (The
Frenchde
terras de nostre lengage son de maior autoritat li cantar language is more worthy and suitable for compos-
la lenga lemosina qe de neguna autra parladura" (Anyone
ing romances and pastourelles, but that of Limousin is bet-
who wants to compose or listen should know first ter
of for
all making vers and cansos and sirventes; my transla-
that no language is as excellent or correct as ours. .tion). See Aubrey 1996, 85.
. . And
in all lands, songs in our language of Limousin are of
greater authority than those in any other language; my
translation).
speaking or hearing; no place is so isolated or solitary that, as long as there are

a few or many, you will not hear singing either by a single person or by many
together; even the shepherds in the mountains know of no greater amusement
than song. All good and evil things in the world are commemorated by the
troubadours. Indeed there is no [good] or evil saying put into rhyme by a trou-
badour that is not remembered every day, because inventing and singing are
the driving force behind all valor.25 (Marshall 1972, 2)
Undoubtedly hyperbolic, this passage nonetheless conveys the message that

for Raimon the composition and performance of the troubadours' songs
knew no social distinction.
To aspire to the greatness of the troubadours, poets must approach the

act of composing the texts of both romance and lyric with great care:
the theme must be judiciously and consistently developed and deployed, and
the language should be appropriate to the theme; notably, Raimon warns
against using "low discourse" or "words" (paraulas) or mixing languages or
dialects (parladura): "One also should take care, if he wants to compose a
song or a romance, that he arranges its theme and discourse continuously
and properly and pleasingly, and that his song or romance not be of low dis-
course or of two [different] languages or of a theme badly continued or badly
pursued" (22). 26 While Raimon recognizes no social distinction of high or
low in the class of performer or composer, he does insist that composers who
wish to emulate the masters must craft the songs, specifically the subject mat-
ter and the words, with a degree of sophistication.27
A Catalan monk and diplomat in Italy from 1289 until at least 1295,
Jofre de Foixà produced the Regies [de trobar] on the commission of the future
Jacme II of Aragon during the latter 's reign as king of Sicily from 1285 to 1291
25 Lines 20^31 : "Totas genz cristianas, iusieuas et sara- 27 While he shows every evidence of being intimately
zinas, emperador, princeps, rei, duc, conte, vesconte, con- familiar with the works of many troubadours, both living
tor, valvasor, clergue, borgues, vilans, paucs et granz, and dead, among them Bernart de Ventadom, Guiraut de
meton totz iorns lor entendiment en trobar et en chanter, o Bornelh, Guillem de St. -Didier, Arnaut de Marvelh, Bertran
q'en volon trobar o q'en volon entendre o q'en volon dire o de Born, Folquet de Marselha, Peire Vidal, Peirol, and Rai-
q'en volon auzir; qe greu seres en loc negun tan privat ni mon de Miraval (see Marshall 1972, xxii-xxiii), Raimon's
tant sol, pos gens i a paucas o moutas, qe ades non auias writings betray only a little firsthand familiarity with French
cantar un o autre o tot ensems, qe neis li pastor de la mon- literary works or with specific French authors (see Mar-
tagna Io maior sollatz qe ill aiant an de chantar. Et tuit li mal shall 1972, Ixviii n3), although by the time of this treatise
e.l ben del mont son mes en remembransa per trobadors. such well-known and prolific trouvères as Blondel de
Et ia non trobares mot [ben] ni mal dig, po[s] trobaires l'a Nesle, Conon de Béthune, the Chastelain de Couci, and
mes en rima, qe tot iorns [non sia] en remembranza, qar Gace Brûlé were already producing love chansons and cru-
trobars et chantars son movemenz de totas galhardias" sade songs. Perhaps his allusion to romans and pastourel-
(my translation). See Aubrey 1996, 84-85. les in French suggests that such works as the Chanson de
Roland, the Roman d'Alexandre, Thomas's Roman de
26 Lines 451-54: "Per aqi mezeis deu gardar, si vol far un
Horn, the romans of Chrétien de Troyes, or the numerous
cantar o un romans, qe diga rasons et paraulas continuadas
other narrative works that had come into being before the
et proprias et avinenz, et qe sos cantars o sos romans non
end of the twelfth century, some of whose subject matter
sion de paraulas biaisas ni de doas parladuras ni de razons
certainly can be described as pastoral, were more readily
mal continuadas ni mal seguidas" (my translation).
to hand than the lyric songs of the north.
(Marshall 1972, lxxii-lxxiv). Written probably for the Catalan lay

me's court, the treatise is based in part on Raimon Vidal's R
eighty years earlier. The text survives in two Catalan manusc
middle to the end of the fourteenth century. Coming as it di
of the classic period of the troubadours, Jofre's work is ret
descriptive regarding the poets whose works he cites, but it mu
as prescriptive in that he was addressing amateur poets (in
himself) who wished to learn Occitan grammar and how to us
correctly to create songs. Jofre includes more details than Raim
sification of Occitan poetry, and he eschews any discussion that
readers to have knowledge of Latin grammar or vocabulary. Like
does not mention music.
Jofre echoes Raimon's injunction that consistency and probity of theme

are paramount, to which he adds comments about the themes of sever
genres - war, or praise and blame in the sirventes, love in the canso or dansa
and he urges that the themes should not be mixed: "For if you begin making
a sirventes about deeds of war or about reproof or about praise, it is not appr
priate to mix in a theme of love; or if you make a canso or a dansa about love
it is not appropriate to mix in deeds of arms or men's slander, if it does not
resemble or communicate the theme" (Marshall 1972, 57). 28 It is notable that
Jofre mentions both canso, which Page, Stevens, and others would consider t
be the epitome of "high-style" songs, and dansa, viewed by many as a "lo
style" genre, in the same sentence, not only making clear that a dance can be
about love, but also implying that composing a dance was as much a sel
conscious and deliberate creative process as composing a canso.
For musicologists, the most intriguing of the vernacular treatises on poetr
is the Doctnna de compondre de dictais, because it is the earliest document to gi
succinct but detailed descriptions of numerous genres, including commen
oh their melodies. The text survives uniquely in a late-fourteenth-centu
Catalan manuscript that also contains Raimon Vidal's and Jofre de Foix
treatises, a poetic version of the early fourteenth-century Leys d' Amors, an
several other lesser treatises on grammar and poetry (see Marshall 1972, xi-xi
John Marshall (1972, lxxvi-lxxviii) has argued compellingly that although the
Doctnna is unattributed, its style, contents, and linguistic peculiarities point
Jofre as its author. A late-thirteenth-century date that such an identification
implies makes this treatise earlier than Johannes de Grocheio's, whom musi-
cologists generally claim to be the first to address the music of secular song.
28 Lines 33-37: "Car si tu comences a far un sirventesch
de fayt de guerro o de reprendimen o de lausors, no. s

conve que.y mescles raho d'amor; o si faç canço o dança
d'amor, no. s tayn que.y mescles fayt d'armes ne maldit de
gens, si donchs per semblances no o podiets aportar a
raho" (my translation).
Elizabeth Aubrey ^ "High Style" and "Low Style" in Medieval Song 105
Elsewhere I have discussed details of the Doctnnds descriptions of genres

and how they relate to extant examples (1996, 86-131; 2000, 279-88), but here
I wish to consider whether anything in the text suggests that the theorist held a
notion of "high" and "low" styles in different genres.
The author distinguishes among the genres by their themes, structures,
functions, and performance. The composer should take great care, he says,
in choosing language, tone, versification, and melody that properly embody
the theme. Most of the author's comments about music have to do with
whether or not the genre deserves a newly composed melody; he says that the
canso, vers, retroncha, alba, estampida, dansa, and discort [sic] should, while the
sirventes, lay, pastora, and planh can use borrowed tunes.
Love is the subject matter of certain songs that are supposed to have
new melodies, including the canso, retroncha, alba, dansa, and discort, as well as
two that can have borrowed melodies, the pastora and plank, some of these
types treat love in quite distinctive ways. The canso merits primacy of place,
speaking "pleasingly about love," and it should always have a new melody, one
as beautiful as the composer can make it (Marshall 1972, 95). 29 Likewise, the
dansa treats "of love well and pleasantly," a subject matter that should be
coherent and carefully developed from beginning to end, and the poem
always should be given a new melody (96). 30 The author mentions the pres-
ence of a refrain in a dansa and says that "one performs it with instruments"
(98). 31 Although the author makes the obvious remark that a dansa accompa-
nies dancing, the song he describes does not seem to be a simple one that
might have been improvised or patched together from formulaic material,
but instead one crafted with the same care given to creating a canso (as well
as a planh, lay, or retroncha, whose descriptions in the Doctrina include similar
language about unity of theme). Guiraut d'Espanha's Ben volgra s'esser poges
(Example 5) might be just such a dance.32
29 Lines 7-9: "E primerament deus saber que canço deu 31 Lines 115-17: "Dansa es dita per ço com naturalment
parlar d'amor plazenment, e potz metre en ton parlar exim- la ditz horn dança[n] o bayllan, cor deu [haver] so plazent;
pli d'altra rayso, e ses mal dir e ses lauzor de re sino e la ditz horn ab esturmens, e piau a cascus que Ia diga e
d'amor" (And first you should know that a canso must la escout" (A dansa is so called, naturally, because one
speak pleasingly of love, and you can put in your poem dances or leaps to it, so that it must [have] a pleasant mel-
examples of other themes, but without speaking evil, and ody; and one performs it with instruments, and it delights
praising nothing but love). Line 15: "E dona li so noveyl co everyone who performs and hears it; my translation).
pus bell poras" (And give it as beautiful a new melody as
32 Page remarks that troubadours deliberately "ignored"
you can; my translations).
the dansa (1986, 25), since there are none extant in the
30 Lines 51-56: "Si vols far dança, deus parlar d'amor be sources that can be dated before the end of the thirteenth
e plasentment. . . ; totes vegades so novell. ... Ε aquella century. Yet their absence from the manuscripts is surely
raho de que la començaras deu[s] continuar e be servar ai a result of what he terms their "ephemeral" nature (24)
començament, ai mig, e a la fi" (If you want to compose a whose existence was primarily oral and not dependent on
dansa, you must speak of love well and pleasantly ... ; writing. This does not mean that troubadours did not com-
every time [it has] a new melody. . . . And whatever theme pose them, it only means that they were rarely copied into
with which you begin, you must continue and make serve the manuscripts that survive.
well for the beginning, the middle, and the end; my
translation).
But the author does make clear that a dansas melody has
style - not a "low" one necessarily, but one that distinguishes it
of a very different genre, the lament or plank. About the latter
that "you can make it to any tune you wish, except that of a
implying that the melody to express a pleasant dansa would b
with the mournful mood of a plank. A pastora, whose stereotyp
author describes without suggesting that it is in any way "light
than a canso, can have a new or a borrowed melody or a "foreign
no longer current" (96), 34 a comment that suggests the melody
might evoke an "otherness" of regional or temporal strangeness.
The author describes a discortns a song that treats love in a n
as "when a man is separated from it [love] or when a man wh
pleasure from his lady and is tormented" (97). 35 The theori
distinct generic label to a song with the theme of "unrequite
today we would consider to be simply a song about "courtly love
to say, to reinforce this "discordant" theme, that the melody sh
opposite of all other songs" (97), rising and falling in ways th
dies do not.36 If a composer takes care to create "as beautiful a m
can" for a canso, it would require as much skill to create a melod
in an "opposite" manner, whether that means with a reversed co
terms "rising" and "falling" imply - or in other ways that the th
explain.
The alba, or dawn song, is about love, but according to the Doctrinds
author its text also involves praise and blame, the distinctive topos of epideic-
tic songs, including the vers, lay, estampida, and sirventes. One should speak of
love "pleasantly" in the alba, and praise or censure the lady and the dawn
depending on whether the lover achieves or fails in his pursuit. Like a canso,
the alba merits a newly composed melody (96). 37
33 Line 58: "pot[z] lo fer en qual so te vulles, salvant de 36 Lines 83-84: "E que en Io cantar, lia hon lo so deuria
dança." muntar, que.l baxes; e fe lo contrari de tot l'altre cantar"
(Wherever the tune ought to rise, make it low; make it do
34 Lines 45-50: "Si vols far pastora, deus parlar d'amor
the opposite of all other songs; my translation). Page
en aytal semblan com eu te ensenyaray, ço es a saber: si.t
(1 986, 23) uses the idea that the descort is the opposite of
acostes a pastora e la vols saludar o enquerer o manar o
a canso in support of his classifying it as a "Low-Style"
corteiar, o de qual razo demanar o dar o parlar li vulles. . . .
song. There is no hint in this treatise of a formal affinity
Ε potz li fer ... so novell o so estrayn ia passât" (If you
between the descort and the couplet-structured lai;
want to compose a pastora, you must speak of love in the
indeed, the author describes its structure as having stan-
way I will teach you, thus: if you meet a shepherdess and
zas (coblas) like those of other genres, and a refrain (Mar-
wish to greet her or woo or pursue or court her, or to ask
shall 1972, 97, lines 84-85): "E deu haver très cobles e
or give or speak to her about something, as you wish. . . .
una o dues tornades e responedor" (And it must have
And you can give it ... a new melody or a foreign melody
three stanzas and one or two tornadas and a refrain; my
that is no longer current; my translation). Compare Rai-
translation). See Aubrey 1994.
mon's comment that French is a better language for a pas-
tourelle, p. 102. 37 Lines 62-66: "Si vols far alba, parla d'amor plazent-
ment; e atressi [deus] lauzar la dona on vas o de que la
35 My translation; lines 81-83: "Si vols far discort, deus
faras. E bendi Talba si acabes Ιο plazier per lo qual anaves
parlar d'amor com a horn qui n'es désemparât e com a horn
a ta dona; e si no.l acabes, fes l'alba blasman Ia dona e
qui no pot haver plaser de sa dona e viu turmentatz."
The sirventes is not a love song, but instead is about "feats of arms" or
"praise of a lord or of calumny or of some new deeds" and is full of "proverbs"
and "examples" and other rhetorical devices. Its tune is nearly always bor-
rowed from another song, especially that of a canso (95-96). 38 Thus, while the
composer pays special attention to creating a compelling text for a sirventes,
setting it to a borrowed tune implies that the melody is not central to the rhe-
torical purpose of the new poem. A few lines further on the author seems to
reinforce this, making an attempt at etymology with the claim that "a sirventes
is so called because it serves and is subordinated to the song from which the
melody and rhymes are taken" (97). 39 Unlike the case of a dansa and a planh,
where the themes of the two genres are so conflicting that using the same
melody is inconceivable, here the theorist sees no discord at all in the melodic
styles of a song about love and one about war or good or bad deeds.40
The lay, whose text the author of the Doctrina says is about God or
worldly matters (not love between man and woman), can have either a bor-
rowed melody, notably a sacred one, or a new one (95), 41 while the estampida
he viewed as a genre worthy of a newly composed melody (97).42 A bit further
I'alba. . . . e deus hi fer so novell" (If you want to compose

39 My translation; lines 104-5: "Serventetz es dit per ço
an alba, speak pleasantly of love; and so praise the lady
serventetz
to per ço com se serveix e es sotsmes a aquell
whom or about whom you compose it. And praise the cantar de qui pren lo so e les rimes."
dawn if you have won the pleasure for which you went to
40 For examples of how troubadours adapted the struc-
your lady; and if you do not win it, make the alba censure
ture of the melody of a preexisting song to a sirventes, see
the lady and the dawn. . . . And it must have a new melody;
Aubrey 1996, 112-21. Recent work by Christopher Calla-
my translation).
han (in a preliminary report, "Contrafacture and Interge-
38 Lines 28-39: "Si vols far sirventez, deus parlar de fayt neric Play in Trouvère Lyric," delivered at the Forty-third
d'armes, e senyalladament o de lausor de senyor o de International Congress on Medieval Studies, Western
maldit o de qualsque feyts qui novellament se tracten. Ε Michigan University, May 8-11, 2008) highlighting the fact
començaras ton cantar segons que usaran aquells dels that a single melody not uncommonly was used for two or
quais ton serventez començaras; e per proverbis e per more songs of contrasting genres suggests the difficulty
exemples poretz hi portar les naturaleses que fan, o ço de we have today in identifying musical features that medi-
que fan a rependre o a lausar aquells deis quais ton ser- eval musicians might have associated with one genre to
ventez començaras. . . . E pot[z] lo far en qualque so te the exclusion of another.
vulles; e specialment se fa en so novell, e maiorment en ço

41 Lines 22-25: "Si vols fer lays, deus parlar de Deu e de
de canço. E deus Ιο far d'aytantes cobles com sera Ιο can-
segle, o de eximpli o de proverbis, de lausors ses feyment
tar de que pendras lo so; e potz seguir las rimas con-
d'amor, qui sia axi plazent a Deu co al segle; e deus saber
trasemblantz dei cantar de que pendras lo so, o atressi Io
que. s deu far e dir ab contriccio tota via, e ab so novell e
potz far en altres rimes" (If you want to compose a sir-
plazen, o de esgleya o d'autra manera" (If you wish to
ventes, you must speak of feats of arms, and in particular
make a lay, you must speak of God or of the world, or of
either of praise of a lord or of calumny or of some new
examples or proverbs, of praise without pretense of love,
deeds. And begin your song following the customs of
which thus would be pleasing to God as well as the world;
those with whom you begin your sirventes; and by prov-
and you should know that it must always be done with
erbs and examples you can bring in the allegiances they
contrition, and with a new and pleasing melody, or one of
swore, or reprove or praise the deeds of those with whom
the church or another type; my translation).
you begin your sirventes. . . . And you can compose it to
whatever tune you wish, rarely a new melody, and chiefly42 Lines 72-74: "Si vols far estampida, potz parlar de
that of a canso. And you must give it as many stanzas as qualque fayt vulles, blasman o lauzan o merceyan, qui.t
there are in the song from which you took the melody; and vulles; e deu haver ... so novell" (If you want to compose
you can follow the rhymes corresponding to the song froman estampida, you can speak of whatever you wish, blam-
which you took the melody, or you can make other rhymes; ing or praising or supplicating; and it must have . . . a new
my translation). melody; my translation). This contradicts the early
on the theorist comments on its performance with the brief rem

name "estampida" comes from the fact that such a song is "taken
he does not mention the possibility of dancing to it (98) .43
The author of the Doctrina does not present a hierarchica
for the genres, nor does he anywhere imply that some genre
quality, profounder rhetorical power, or more sophisticated
others, or have composer, performer, or audience of only
social class. As for the music specifically, the genres that requir
ody include several that some today consider to be in a "low
estampida, descort, retroncha - and the theorist's discussion imp
would demand as much attention and skill as composing a new
All of these southern vernacular treatises are essentially pre
the sense that they clearly were intended to teach their readers
pose poetry, although their readers, amateurs of aristocrati
class, were not the troubadour masters whose famed songs were
posterity in manuscripts. The documents also are descripti
their authors' observations on current practices, although
tions are less definitive than we might wish. Their didactic
largely based on the rhetorical tradition of enarratio poetarum -
the great poets; all of the authors quote liberally from songs
dours. But given their audience, we must be cautious in inferrin
texts any insight into how the troubadours themselves viewed s
as genre or style.
As is well known, the sole music theorist of this period to a
nacular genres is the Parisian Johannes de Grocheio (c. 1300)
other Latin theorists was addressing a scholastic audience (R
translation of the relevant passages in Page 1993). Grocheio's f
stands apart in many ways. It is a musical tract, not an art of po
like these other Latin documents, yet it incorporates an impo
unique - discussion of vernacular song, which follows neither
the vernacular tradition, nor indeed that of other treatises on m
Grocheio's categorization of vernacular types is notorious
sial. Without belaboring the details of his fulsome discussio
those concerned with form, rhythm, and instruments, which ar
my chief concern, or the intractable problem of matching Groc
tions with surviving songs (on this, see McGee 1989; Aubrey
43 Lines
fourteenth-century report about the well-known 124-25: "Stampida es dita per ço stampida
"Kalenda
maia," whose tune Raímbaut de Vaqueiraspren
was purported
vigoria en contan o en xantan pus que null autr
tar" performed
to have appropriated from an instrumental work (An estampida is so called because it is taken vig
by a couple of traveling jongleurs, although
ously
as this
in [re?]counting
report or in singing, more than any
postdates the earliest extant version of Raimbaut's
song; mysong,
translation).
it
is difficult to confirm its veracity. I have suggested else-
where that the self-designation of this song as an estamp-
ida is a tenuous poetic conceit; see Aubrey 1997, 40-52.
on his taxonomy of vocal vernacular song.44 In this discussion, Grocheio com-

ments on who composed and performed different types of songs, subject mat-
ter, audience and venue, and, of course, poetic and musical structure, all of
which is summarized in Table 4.
Grocheio divides vocal song into two types, cantus and cantilena, whose
main distinguishing feature is a structural one: each subcategory of cantilena
has a refrain at its beginning and end. Despite the commonly held view today
that the presence of a refrain implies a connection with dance, Grocheio
refers to dance in his discussion of only one subcategory of cantilena, the
ductia, but not that of the rotunda or the stantipes, the latter of which he says
lacks the regular pulse of the ductia, which implies that it might not be suit-
able for dancing. Grocheio cites several examples of cantilenae of different
types, but only one survives: a rotunda, Toute sole passerai le vert boscage (Linker
1979, 265-1673; van den Boogaard 1969, 85, rondeau 175). As with C'est la
gieus (Example 4), the full text of this song is extant only as a one-stanza ron-
deau without music, in a fourteenth-century manuscript (Paris, Bibliothèque
nationale de France, f. fr. 12786, fol. 78v); an apparent version of its refrain is
found with music in the motetus of the motet Amours qui vient par message ne
pourroit/Toute soûle passerai li bois rame/NoTUM,45 from which various editors
have reconstructed a melody for the entire stanza (Gennrich 1921, 1:79,
Tischler 1997, 14: rondeau 39). As for the other vernacular type, cantus, songs
without refrains, Grocheio includes both lyric chansons and nonlyric chansons
de geste, a poetic form with a laisse (rather than stanzaic) structure whose sub-
ject is heroic deeds.
Regarding subject matter, Grocheio's categories do not follow the
generic distinctions that we find in poetic treatises. Of the lyric types, he
describes the content of only the cantus coronatus, and he does not specifically
mention love, certainly not what we would interpret as "courtly love," but
rather topics such as friendship and kindness. As far as audience is concerned,
44 One particular of Grocheio's cursory comments has

only plainchant and polyphony, whereas he intends to
excited much debate: his mention of a division advocated include vernacular song in his taxonomy, requiring a tripar-
by "others" (besides himself) who classify music as either
tite division. The bipartite division that depends on a dis-
measurable - meaning polyphony - or immeasurable - tinction between music with measurable rhyth'ms and
plainchant - to which he objects that all music is measur- music without thus is inadequate for his purposes, but
able in the sense that "every process of music - and of anyrather than dismiss it entirely, he defines the term "mea-
art - must be calculated according to the rules of that art"surable" to mean not rhythm but instead broadly the "rules
("quaelibet operatio musicae et cuiuslibet artis débet illius of the art," so his inclusion in his discussion of music that
artis regulis mensurari," Page 1993, 20). Grocheio goes on is neither plainchant nor polyphony is justified. On various
to say that he can accept the division if by "immeasurable"views of this passage, see Stevens 1986, van der Werf
one really means "not so precisely measured," a phrase 1988, Aubrey 1997, and Mullally 1998.
that has been taken to mean that monophonie music is not
45 Complete in Bamberg, Staatsbibliothek, Lit. 115, fol.
governed by precise rhythms. Reading the passage strictly,
58-58v, and as a fragment in Rome, Biblioteca Apostólica
one must acknowledge that Grocheio is speaking only of
Vaticana, Reg. 1543, vol. lav; the text in the motet is
monophonie church music, not vernacular monophony,
slightly different than in the full rondeau.
which he has not yet mentioned. What he seems to be
doing is giving a nod to previous discussions that concern
1
îi
d S
'S
fi
"3d % j«
■S ·§■ -S s g ^2
2 fi -s s S -g, ΐ
111 I i I i 1 1 1
$
,D "^
1 ■
S, S
o -§ -g
s
■il
to 35
1 s
Ü
s
5
(A
Φ
ο.
£
>
"(5
υ
ο
> S "^5 ■ rt bo ri be
•i II "^5 ■ ΐ bo § -s § be
■S*
ã
I lof 38 Jü
s
•5
§ ü-S
ü
g SI be
ΐ
-§ bo
^ "g
rt
12 1 || II ï
o
1 1 ε. Ι ι s « I
■S
1
ι
s 'S
fifi Λ
ίΛίΛ α
C«C«
^ ^ fi
1
31 t è s ·§ I ·§ I s „. ä g
Ic
o
o <u<^fi ooogoo Ofi
l! h ill
"õ>
s:
υ
S
CD
ω
■σ
i
c
ta
1
I | -a Hi I
H j S u £ S <
S ^ . *
s ξ . fc <υ
„ε g
S 2 » υ 1
ι ·§ £ 1 g
I -g -S f -5 I ■=·
■s § S ■§ is 2 H
■s l| 2 s » 1 i S
II S I ï £ 31
!s 1 .,
"t: ι §
Is If il i il
11 1 1 il ., 1 s SB
Π tîïlliîèï
flllilifniil
! S ï ί I S 1 1 I
η il ^ ni
'S s ■«
Β υ "g
S s t
<υ ο x
> fi w
Grocheio does not distinguish consistently between cantus

Nobles, masters, and students - upper and educated classes - -
ence for the cantus coronatus, but the wealthy at festivals and e
enjoyed the cantilenae stantipes and ductia. Intriguingly, accordi
the main venue for chansons de geste, which he placed in the cate
was the city, and the chief listeners the working classes.46
All told, it is difficult to trace in Grocheio's remarks a distinction
between a "high style" and a "low style" as these are currently defined. It is
tempting to view his cantus and his cantilena as a medieval articulation of the
notion of "high style" and "low style." But this is not as tidy as it might seem.
By including both chansons and chansons de geste as types of cantus, he defies
many modern classifications of medieval song into lyric and nonlyric types,
as well as the separation that Page makes between "High-Style" lyric love
songs and "Low-Style" narrative works.47 He only obliquely implies that the
refrain types - the cantilenae - are of less quality than the cantus types, but
he suggests that the stantipes and the ductia can be rather complex in poetic
and musical structure, as well as powerful in their impact. The stantipes is
"difficult" in its diverse rhymes and musical sections so as to distract "young
men and girls" - it is unclear whether Grocheio means as listeners or
performers - from their "depraved thoughts," and the ductia has the power
to divert these same youth from the pursuit of "erotic love."
The only suggestion Grocheio makes that any type is of lesser quality
than the others is in reference to one of the nonrefrain types, the cantus ver-
sualis, which lacks the "bonitas" of the other cantus types, leading "some" to
call it a cantilena. One of his two examples of a cantus versualis is a chanson
d'amour of Thibaut de Navarre, Chanter πι estuet quarne m'en puis tenir (RS 1476;
Example 7). He cites another song by Thibaut, Ausi com Vunicorne sui (RS
2075, Example 8), as an example of a cantus coronatus, "crowned" because
unlike the cantus versualis it has "bonitas" of words and music. Several scholars
have compared these two songs in search of these distinctions, with varying
results. Mullally (1998, 7) claims that despite a striking metaphor in stanza
four, "The Phoenix seeks the pyre and the vines/in which he burns himsel
and gives up his life" ("Li Fenix qiert la busche et le sarment/en quoi il s'ar
et gete fors de vie"; Brahney 1989, 84-85), "the language [of Chanter m'estuet]
is direct rather than the figurative use observable in the cantus coronatus.
Both songs use coblas doblas, wherein the rhymes change every two stanzas
and both have isometric poetic structures that Page would categorize as
46 The most recent attempt to explain Grocheio's discus-

47 Page 1986, 38, 199, and 250n13, equates Grocheio's
sion of secular song is the penetrating and nuanced cantus
study coronatus with the trouvère grand chant courtois, in
by Mullally (1998). a "High-Style register," but he dismisses Grocheio's inclu-
sion of chanson de geste in the same category, implying
that the chanson de geste is in a "Low Style"; see also
Page 1984-85.
"High Style," although the paroxytonic b' rhymes in Chanter m'estuet give the
impression of different syllable counts:
Chanter m'estuet quar ne m'en puis tenir (cantus versualis)
Melody A A' A" A" A"" Β A"" C

Rhymes ab' ab' c c b' c
Syllable count 10 10+1 10 10 + 1 10 10 10 + 1 10
Ausi com Vunicorne sut {cantus coronatus)
Melody ABCDEFGHI
Rhymes abbaccbdd
Syllable count 888888888
Karp (2001, 391) finds the melody of Chanter m'est

cal," with a smaller range than that of Ausi com V
syllabic and have a range of about an octave; both m
occasional leaps of a third both ascending and desc
ence between the two is the repetitive structure of
com Vunicorne is through-composed. But the music
continually evolving - not strict, but varied in crea
1 is never repeated exactly, but progresses thro
decreasing ornamentation (compare the cadence of
1, the reduction of a two-note neume in the midd
in the middle of verse 4), and shifting contours (v
to a D, a step higher than the C of verses 1 and 3;
return in the phrase labeled A until near the end
the Β phrase, takes the first four notes of the A
ing the same progression of whole- and half-ste
drop of a minor third, all of which is echoed in v
below. Only two pitches serve as cadence points at t
A. The "open" cadence on A occurs first in verse 5
up to the highest note in the melody, F, before d
the G of the first four verses; the other occurren
in the variation in verse 7 of the music of verse 5
The final verse introduces new music, but its final
2 and 4. To this beholder, the ebb and flow of t
Grocheio says lacks "bonitas" is full of the "inventi
expectation" that Stevens claims for the "high sty
On the other hand, the through-composed melo
sui does not follow the structural conventions that Stevens associates with the
"high style" (see discussion of Example 1, pp. 87-89) or the coherence articu-
lated by interrelationships among verses that Page cites as a characteristic of
"High-Style" melodies (see discussion of Example 3, pp. 92). There is not the
same kind of variation that one finds in Chanter m'estuet, nor a clear-cut tonal
rhymes music
° Chan - ter m'e - stuet, que ne m'en puis te - nir,
2 <£ # m *
_ÏT 1
' '?/
I
Β * · # *^# ^= b' A'
8 et si n'ai je fors qu'a - nui et pe - san - ce;
3 jk # , # ' * , , , >^ί »~#!Ξ a A"

° mes tout a - des se fait bon res - jo - ir
4 <£ # » # » Φ Β * # # #^# >^

° qu'en fe - re duel nus dou mont ne s'a - van - ce.
° Je ne chant pas com horn qui soit a - mes,
6 ZU * m * ) m m m J7' c b
* mes corn des - troiz, pen - sis, et es - ga - rez;
7 ^(« « * - m) φ · ''V ' ' * (*^^ b' A'""

* que je n'ai mais de bien nule es - pe - ran - ce,
8 <fr>#*#«#. » » · gi^Si c c
8 ains sui toz jors a pa - ro - les me - nez.
Example 7. Thibaut de Navarre, Chanter m'e stuet quar ne m'en puis teni
65-65v
Elizabeth Aubrey *~ "High Style" and "Low Style" in Medieval Song 115
. rhymes music
1 #t) m * m #~# *~# # rj a A

* Ein- si com u - ni - cor - ne sui
2 Μφ) Φ ' Φ Μ Φ Φ~· # ί b B

* qui s'es- ba - hist en res - gar - dant
3 ψ) ' " ' l ' · ^P b c

quant la pu - ce - le va mi - rant.
4 Κ ' · ' . Γ* ΓΗ 3 D
Tant est li - e de son en - nui,
* pas - me - e chiet en son gi - ron;
6 mfò φ m φ m #~# φ <Π^ΊΞ c F

lors Γ ο - cist en en tra - i - son.
7 fom Φ - * - Φ m #|ΒΞ b G
° Et moi ont mort d'au - tel sem - blant
s μφ) mm ^ j-η ^ d h
a - mors et ma da - me por voir:
9 dL) # * · * * »^ * il d
* mon euer ont, n'en puis point ra - voir.
Example 8. Thibaut de Navarre, Ausi corn l'unicorne sui (RS 2075), Mt fol. 75v-76
center, as nine phrases begin on six different pitches (E, F, G

cadence on another six (low C, D, E, F, G, middle C). Groch
perceived a qualitative difference between the two songs by T
still placed both in the category of cantus; Mullally concludes
Vunicorne simply "is a better . . . example of what is generally k
chanson courtoise' (1998, 7-8). Although Grocheio is not more
this, it seems that he regarded a cantus coronatus as different f
versualis not because it occupies a different stylistic space defin
contrast to "low," but simply because it deploys words and musi
another song in the same generic category.
It is in fact extraordinarily difficult for us today to perceiv
cal and poetic features struck Grocheio as "better" than oth
"better" if its words were literarily profound or linguistically a
they were more direct? Unexpected or conventional? If its struc
metrical or symmetrical? Complex or simple? Varied or regular,
through-composed? If the melody was wide of range or melis
pact or syllabic? Tonally coherent or inventive? "Tuneful" o
Analysis of these two songs, deemed by Grocheio as differe
does not readily yield an answer, nor does it shed light on the la
of whether Grocheio's notion of bonitas can be applied to our co
and "high" style.
Grocheio's knowledge of the repertoire of trouvère song, jud
songs that he cites, was limited. For his cantus he chooses love s
known trouvères (although he does not mention their names
baut de Navarre and one by the Chastelain de Coucy, as well
mous love song, while all of his cantilenae are anonymous, a
extant. In contrast, he alludes to more than sixty plainchants by
(see the catalogue in Rohloff 1967, 200-1). For a theorist wh
much verbiage categorizing and describing the repertoire of ver
the sparseness of his citations is striking, but so also is the fact t
half of the songs he gives as exempla do not survive. These
songs current in Paris or in Grocheio's native Normandy (ind
cally says that the cantilena rotunda is common in Normandy) d
but they are not the songs that scribes valued enough to record
chansonniers, most of which date from before Grocheio's time. Of his five
examples that do survive, one surviving manuscript includes four of them
(ms. X), scattered widely among its leaves, and five other sources include
three of them (mss. K, O, R, U, and V), but again, widely dispersed in each
volume.48 It would be difficult to argue that Grocheio had access to any of
these manuscripts, most of which were produced in the Artois-Picardy region
before Grocheio's time (see Aubrey 2001).
48 Ausi com I'unicorne is in manuscripts A, B, C, F, K, Mt,0, P, T, U, V, and X; Chanter m'estuet is in K, Mt, 0, R, S, T,

0, R, T, U, V, X, Z, and a; Quant li roussignol is in C, K, M,V, X, and Z; and Au repairier que je fis is in Ν, Ρ, R, and X.
Grocheio's system is an elaborate attempt to provide an authoritative

justification for describing a repertoire - vernacular song - that by his day
had reached a respectable level of sophistication, not to mention a ubiquity
at all levels of society. Certainly the Aristotelian framework that underlies his
discussion and the concepts from the art of rhetoric on which he draws pre-
sent the reader with a compelling case for viewing vernacular song as on a par
with plainchant and polyphony (see Aubrey 1997). Whether or not poets and
musicians viewed vernacular songs in this scholastic way remains an open
question.
* * *
Although providing plent

posers, performers, audien
among the songs of the tr
manuscripts and theoretical
the songs themselves paint
coherent. The sundry dates
purposes for which they we
works and the songs, instru
tion, and perhaps editing
onciling their wildly diverg
The manuscripts and th
scribes and theorists, at lea
tinctions among different g
not perceive some genres as
The mostly random placeme
and the mixture of stylistic
theme, structure - in theoretical discussions reinforce the evidence of the
songs themselves, that generic distinctions are difficult to pinpoint, and that
even when they are clear, composers of most of the songs that survive devoted
as much care to creating them no matter what their theme or structure. It is
hard to escape the conclusion that notions of "high" style and "low" style,
specifically in reference to genres, were unknown to scribes, theorists, and
composers.
All of the evidence we have is what was written down - texts that were
deemed worthy of preservation by edμcated individuals. We have every rea-

son to believe that numerous songs were never written, songs that existed in
an idiom for which writing was considered unnecessary. This surely includes
dances, simple refrains, working songs, lullabies, and other musical expres-
sions that spiced the daily lives not just of peasants, but of members of every
stratum in society. Such songs can be considered "popular" in that they were
"of the people," but of all the people, not only those in the lower classes.
Given the certainty that such never-recorded songs existed, nonetheless
without a written record we have no evidence by which to compar

their stylistic features might have been with those of songs th
served, no way to define differences between them, no way to artic
ing characteristics of the style of songs written down and songs n
down.
Whether we are concerned with the songs of which we have written

records, or with a larger sociological milieu of written and unwritten music,
both the notion of genre and the hermeneutical construct of "high style" and
"low style" are not as helpful as they seem. Although identifying attributes
that might be specific to a genre can serve a useful purpose to help sort out
the complex picture of medieval song, especially regarding the text, whose
"dominant" theme usually is obvious, using either a system of genres or a
"high-low" dichotomy as a framework to analyze extant melodies is demon-
strably unenlightening, if for no other reason than that there is no agreement
on what the salient or defining characteristics of a style are; we are mostly in
the dark about how medieval musicians perceived the elements of melody or,
indeed, if they even recognized such musical features as are listed at the
beginning of this article. The evidence we have remains rich and full of
potential, and a more nuanced approach to it, unencumbered by a narrow
view of style and genre, still holds the promise of a greater understanding of
the songs of the troubadours and the trouvères.
Abbreviations
A Arras, Médiathèque municipale, ms. 657

BnF Paris, Bibliothèque nationale de France
C Bern, Burgerbibliothek, ms. 389
I Oxford, Bodleian Library, Douce 308
Κ Paris, Bibliothèque de l'Arsenal, ms. 5198
L Paris, Bibliothèque nationale de France, f. fr. 765
M Paris, Bibliothèque nationale de France, f. fr. 844
Mt Paris, Bibliothèque nationale de France, f. fr. 844,
fols. 13 and 59-79
Ν Paris, Bibliothèque nationale de France, f. fr. 845

Ο Paris, Bibliothèque nationale de France, f. fr. 846
Ρ Paris, Bibliothèque nationale de France, f. fr. 847
PC Pillet, Alfred, and Henry Carstens. 1933. Bibliographie der
Troubadours, Schriften der Königsberger Gelehrten Gesellschaft.
Halle: Niemeyer.
Q Paris, Bibliothèque nationale de France, f. fr. 1109
R Paris, Bibliothèque nationale de France, f. fr. 1591
RS Spanke, Hans. 1980. G.-Raynauds Bibliographie des altfranzösischen
Liedes, erster Teil, ergänzt mit einer Diskographie und einem Register
der Lieder nach Anfangsbuchstaben hergestellt. Edited by
A. Bahat. Leiden: Brill.
Τ Paris, Bibliothèque nationale de France, f. fr. 12615

U Paris, Bibliothèque nationale de France, f. fr. 20050
V Paris, Bibliothèque nationale de France, f. fr. 24406
W Paris, Bibliothèque nationale de France, f. fr. 25566
Wa Paris, Bibliothèque nationale de France, f. fr. 25566, fols. 2-9
X Paris, Bibliothèque nationale de France, nouv. acq. fr. 1050
Ζ Siena, Biblioteca Comunale degli Intronati, H.X.36
a Rome, Biblioteca Apostólica Vaticana, Regina 1490
Works Cited
Allen, Philip S. 1931. Medieval Latin Lyrics. Chicago: University of Chicago Press.
Andersen-Wyman, Kathleen. 2007. Andreas Capellanus on Love? Desire, Seduction, and Subversion
in a Twelfth-Century Latin Text. New York: Palgrave Macmillan.
Appel, Carl, ed. 1915. Bernart von Ventadorn, seine Lieder, mit Einleitung und Glossar. Halle: Nie-
meyer.
Aubrey, Elizabeth. 1982. "A Study of the Origins, History, and Notation of the Troubadour
Chansonnier Paris, Bibliothèque nationale, f. fr. 22543." Ph.D. diss., University of Mary-
land.
tural Milieu of the Troubadours and Trouvères, ed. Nancy Van De

tute of Mediaeval Music.
Indiana University Press.
Music History 16: 1-53.
Trouvères." In Medieval Lyric: Genres in Historical Context, e

Urbana: University of Illinois Press.
of Music and Musicians, 2nd ed., ed. Stanley Sadie and Jo

London: Macmillan.
Bee, Pierre. 1977-78. La lyrique française au moyen âge (XlIe-XIIIe siècle): Contribution à une typolo-
gie des genres poétiques médiévaux. 2 vols. Centre d'Études Supérieures de Civilisation
Médiévale de l'Université de Poitiers, 6-7. Paris: Picard.
Blodgett, E. D., ed. and trans. 1995. The Romance of Flamenca. New York: Garland.
Boase, Roger. 1977. The Origin and Meaning of Courtly Love: A Critical Study of European Scholar-
ship. Manchester: Manchester University Press.
Brahney, Kathleen, ed. and trans. 1989. The Lyrics of Thibaut de Champagne. Garland Library of
Medieval Literature, A, 41. New York: Garland.
Burns, E.Jane, and Roberta L. Krueger, eds. 1985. Romance Notes, Vol. 25/3: Courtly Ideology and
Woman's Place in Medieval French Literature [special issue] .
Del Monte, Alberto, ed. 1955. Peire d'Alvernha: Liriche; testo, traduzione, e note. Turin: Loescher-
Chiantore.
Donaldson, E. Talbot. 1965. "The Myth of Courtly Love." Ventures: Magazine of the Yale Graduate
School 5/2: 16-23.
Doss-Quinby, Eglal and Samuel N. Rosenberg, eds., with music editions and commentary by
Elizabeth Aubrey. 2006. The Old French Ballette: Oxford, Bodleian Library, MS Douce 308.
Geneva: Droz.
Dragonetti, Roger. 1960. La technique poétique des trouvères dans la chanson court
l'étude de la rhétorique médiévale. Werken uitgegeven door de Faculteit va
wijsbegeerte 127° Aflevering. Bruges: Rijksuniversiteit Gent. Reprint. Ge
1979.
Ferrante, Joan. 1980. "Cortes 'Amor in Medieval Texts." Speculum 55: 686-95.
Frappier, Jean. 1973. Amour courtois et table ronde. Publications romanes
Geneva: Droz.
Gallo, Ernest, éd. and trans. 1971. The Poetria nova and Its Sources in Early Rhetorical Doctrine.
Hague: Mouton.
Gennrich, Friedrich. 1921, 1927. Rondeaux, Virelais und Balladen aus dem Ende des XII., dem
und dem ersten Drittel des XIV. Jahrhunderts, mit den überlieferten Melodien. 2 vols. Gesellsch
für romanische Literatur, 43, 47. Göttingen: Gedruckt für die Gesellschaft für ro
nische Literatur.
Gumbrecht, Hans Ulrich. 1996. "A Sad and Weary History: The Grundriß der romanischen Litera-
turen des Mittelalters." In Medievalism and the Modernist Temper, ed. R. Howard Bloch and
Stephen G. Nichols, 439-71. Baltimore: Johns Hopkins University Press.
Hoby, Otto. 1915. Die Lieder des Trobadors Guiraut d 'Espanha. Freiburg: St. Paulusdruckerei.
Huit, David F. 1996. "Gaston Paris and the Invention of Courtly Love." In Medievalism and the
Modernist Temper, ed. R. Howard Block and Stephen G. Nichols, 192-224. Baltimore:
Johns Hopkins University Press.
Jauss, Hans-Robert. 1970. "Littérature médiévale et théorie des genres." Poétique 1: 79-101.
Jauss, Hans-Robert and Erich Köhler, eds. 1968-2005. Grundriss der romanischen Literaturen des
Mittelalters. Heidelberg: Winter.
Karp, Theodore. 2001. "Thibaut IV. In The New Grove Dictionary of Music and Musicians,
2nd ed., ed. Stanley Sadie and John Tyrrell, 25: 391-92. London: Macmillan.
Kelly, Douglas. 1991. The Arts of Poetry and Prose. Typologie des Sources du Moyen Age Occiden-
tal, 59. Turnhout: Brepols.
Lawler, Traugott, ed. and trans. 1974. The Parisiana poetria of John of Garland. New Haven: Yale
University Press.
Lazar, Moshé. 1995. "Fin'amor." In A Handbook of the Troubadours, ed. F. R. P. Akehurst and
Judith M. Davis, 61-100. Berkeley: University of California Press.
Linker, Robert White. 1979. A Bibliography of Old French Lyrics. Romance Monographs, 31. Uni-
versity, MS: Romance Monographs.
Marshall, John H., ed. 1971. The Chansons of Adam de la Halle. Manchester: Manchester
University Press.
University Press.
McGee, Timothy J. 1989. "Medieval Dances: Matching the
tions." Journal ofMusicology 7: 498-517.
Monfrin, Jacques. 1955. "Notes sur le chansonnier provença
M. Clovis Brunei par ses amis, 2: 282-312. Paris: Société
Monson, Don A. 1988. "Andreas Capellanus and the Problem
Mullally, Robert. 1998. "Johannes de Grocheio's 'Musi
79: 1-26.
Nichols, Stephen G. 1998. "Et si on repensait le 'grand chant courtois'?" In Paul Zumthor ou
l'invention permanente: Critique, histoire, poésie, ed. Jacqueline Cerquiglini-Toulet and
Christopher Lucken, 33-50. Geneva: Droz.
Nims, Margaret F., ed. and trans. 1967. Poetria nova of Geoffrey ofVinsauf Toronto: Pontifical
Institute of Mediaeval Studies.
Paden, William D. 2000. "The System of Genres in Troubadour Lyric." In Medieval Lyric: Gen
in Historical Context, ed. William D. Paden, 21-67. Urbana: University of Illinois Press.
Page, Christopher. 1984-85. "Music and Chivalric Fiction in France, 1150-1300." Proceedings of
the Royal Musical Association 1 1 1 : 1-27.
1100-1300. Berkeley: University of California Press.
University of California Press.
tion." Plainsong and Medieval Music 2: 1 7-41 .

Paris, Gaston. 1881. "Études sur les romans de la table ronde:
10: 465-96.
12: 459-534.
Parry, John J., trans. 1941. The Art of Courtly Love [Andreas Capellanus,
New York: Columbia University Press.
Pascall, Robert. 2001. "Style." In The New Grove Dictionary of Music
ed. Stanley Sadie and John Tyrrell, 24: 638-42. London: Macmillan
Paterson, Linda M. 1999. "Fin amor and the Development of the Courtly
dours: An Introduction, ed. Simon Gaunt and Sarah Kay, 28-46. Cambridge:
Cambridge University Press.
Pfeffer, Wendy. 2001. "Fin'amors and the Women Trouvères." In Songs of the Women Trouvères,
ed. and introd. Eglal Doss-Quinby, Joan Tasker Grimbert, Wendy Pfeffer, and
Elizabeth Aubrey, 35-44. New Haven: Yale University Press.
Pickens, Rupert T. 2000. "The Old Occitan Arts of Poetry and the Early Troubadour Lyric." In
Medieval Lyric: Genres in Historical Context, ed. William D. Paden, 209-41. Urbana: Uni-
versity of Illinois Press.
Pillet, Alfred and Henry Carstens. 1933. Bibliographie der Troubadours, Schriften der Königsberger
Gelehrten Gesellschaft. Halle: Niemeyer.
Pinegar, Sandra. 1991. "Textual and Conceptual Relationships among Theoretical Writings on
Measurable Music of the Thirteenth and Early Fourteenth Centuries." Ph.D. diss.,
Columbia University.
Quadlbauer, Franz. 1962. Die antike Theorie der genera dicendi in lateinischen Mittelalter. Vienna: H.
Böhlhaus Nfg., Kommissonsverlag der Österreichischen Akademie der Wissen-
schaften.
Randel, Don Michael, ed. 2003. The Harvard Dictionary of Music, 4th ed. Cambridge, MA:
Belknap Press of Harvard University Press.
Reimer, Erich, ed. 1972. Johannes de Garlandia: De mensurabili musica. Kritische Edition mit Kom-
mentar und Interpretation der Notationslehre. Beihefte zum Archiv für Musikwissenschaft,
10-11. 2 vols. Wiesbaden: Steiner.
Robertson, D. W. 1968. "The Concept of Courtly Love as an Impediment to the Understanding

of Medieval Texts." In The Meaning of Courtly Love, ed. F. X. Newman, 1-18. Albany: State
University of New York Press.
Rohloff, Ernst, ed. 1967. Die Quellenhandschriften zum Musiktraktat des Johannes de Grocheio, im
Faksimile herausgegeben nebst Übertragung des Textes und Übersetzung ins Deutsche, dazu Bericht,
Literaturschau, Tabellen und Indices. Leipzig: Deutscher Verlag für Musik.
Rosenberg, Samuel N., Samuel Danon, and Hendrik van der Werf, eds. 1985. The Lyrics and Melodies
of Gace Brûlé. Garland Library of Medieval Literature, A/ 39. New York: Garland.
Spanke, Hans. 1980. G. Raynauds Bibliographie des altfranzösischen Liedes, erster Teil, ergänzt mit
einer Diskographie und einem Register der Lieder nach Anfangsbuchstaben hergestellt. Ed.
A. Bahat. Leiden: Brill.
Stevens, John. 1974-75. "Lagrande chanson courtoise: The Chansons of Adam

ceedings of the Royal Musical Association 101: 1-30.
Cambridge Studies in Music. Cambridge: Cambridge Un

Switten, Margaret L. 1995. Music and Poetry in the Middle Ages
Occitan Song, 1100-1400. Garland Medieval Bibliographi
Tischler, Hans, ed. 1997. Trouvère Lyncs with Melodies: Com
Corpus Mensurabilis Musicae, 107. Neuhausen: America
Hänssler-Verlag.
Trojel, E., ed. 1892. Andreae Capellani regii Francorum
Gandiana.
van den Boogaard, Nico. 1969. Rondeaux et refrains du Xlle siècle au début du XTVe. Biblioth
française et romane, D/3. Paris: Klincksieck.
van der Werf, Hendrik. 1988. "The 'Not-So-Precisely Measured' Music of the Middle A
Performance Practice Review 1 : 42-60.
Zumthor, Paul. 1963. Langue et techniques poétiques à l'époque romane. Paris: Klincksieck.
Elizabeth Aubrey is professor of music at the University of Io

Women Trouvères (2001 ) and author of The Music of the Troubadours (1996) as well as numerous arti-
cles on the music of medieval France; she held a fellowship from the National Endowment for the
Humanities in 2006-7. (*%*

Med Song

Uploaded by

Copyright:

Available Formats

Med Song

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Med Song

Uploaded by

Copyright:

Available Formats

Yale University Department of Music

Reconsidering "High Style" and "Low Style" in Medieval Song

Journal of Music Theory 52:1, Spring 2008

content - the manner in which something is said as distin

The analysis of style ... is more likely to concentrate on the

As we shall see, this pitfall lies at the heart of many stu

by the act of creative will a composer asserts something; he m

social distinction of audience or venue; "aristocratic" or "learned" and "jon-

The earliest formulation of a distinction between a "high style" and a

style in the love songs of the troubadours, trouvères, and late

specific characteristics of subject matter, language and vocabulary, rhetor

1 The article emerged from Jauss's work as coeditor

the registers as participles - "nobilizing," "popularizing," "folklorizing" -

A sociological discrimination of different text types is thus extremely complex

Table 1. Pierre Bec, Classement typologique des genres (1977-78, 1:38-39)

I.- LE GRAND CHANT COURTOIS [registre aristocratisant

F - f = maximal to minimal frequency in French

The term grand chant court

Mas so non pot remaner, But this cannot continue,

Another occurrence is in the late-twelfth-century narrative Roman de Fla-

hunting and hawking, grand feasts with singing and dan

7 Page (1986, 38) makes a similar point.

poem, but in most of these rare instances it is not always

Manuscript rubrics that include generic labels are random, relatively

entered), which organize the songs alphabetically by first letter of the

A - chansons', jeux-partis; chansons

M - trouvère chansons; troubadour cansos; motets; lais

While littéraires continue t

all belong to the imagined, i

£3 C0J>(M (Μ Φ Ο Ι> ί^ γη ^ C5 Ο s? ι-ί

Stevens found in the poetic structures of such songs a "'harmoniousness' -

° si ne chant pas pour mes maus a - le - gier,

4 fk' w m + a^l m _ m m + #1 ^7^- ι - b B

5 £ m ^0 m ■ · β ^m *~, ^' m^m(.) = b c

7 fe- »^»* *~, » »~« "*^<;> c' E

8 <k> ' ' ' ' ' "^»' ; jP a F

10 (fe * * « # · * *"· »^ II c< h

Example 1. Adam de la Halle, Tant me plaist vivre en amou

Melody Α Β B' C D Α Β Β' C D'

Yet one might hear in this melody some of the

° A - mours m'ont si dou - ce - ment

3 Û* ~ ' " ·- " *^7- , B'

4 A* m * =' ' * ± ./'*~m' _ ) c

* en a - mer loi - al - ment.

Example 2. Adam de la Halle, Amours m'ont si d

Although most scholars of medieval literature who have studied genre

markedly short verses, and refrains are common; the beloved

Example 4 gives the music of the two verses at the end of th

10 Tischler (1997, 14: rondeau 20) gives both ma

4 fts φ φ φ J~*^ f ,^ ' B'

5 WP *^^ ###^#^#* φ * φ μΦ á^=-. c

mas gran pa - or m Ό toi e gran te - men - sa

7 (fr> , , , , .""; _ -τ*. ~z- B"

e re no sai si m'es o dans o pros.

Example 3. Arnaut de Marvel h, Las grans beutatz e.ls fis ense

Wolfenbüttel 1099, fol. 222

8 A - mor ai a ma vo - len - te te - les con ge vueill.

. Montpellier H 196, fol. 259v-260 ^

- l'- r [ r r cj^^i b '

7 fe- »^»* ~, » »~« "^<;> c' E

5 WP ^^ ###^#^# φ * φ μΦ á^=-. c