Of The Middle Ages: Oices Instruments
Of The Middle Ages: Oices Instruments
oices § Instruments —
n
aristopher Page
In the history of early music one of the most
pressing yet contentious questions concerns
instrumental practice in the performance of
medieval song, especially the lyrics of the
troubadours and trouveres.
Christopher Page has done more than
anyone in carrying out research in this field.
He has explored the vast amount ofliterary
and other evidence and is able to place our
understanding of these instruments and their
music on a new footing. After examining the
place of instruments in the genre-system of
medieval song, he discusses stringing and
tuning. He describes the instruments and
considers pitch, playing-techniques and
repertoire.
Dr Page’s book throws important new
light on his subject and will appeal to
listeners and performers, as well as to those
interested in the literature and history of the
Middle Ages.
Digitized by the Internet Archive
in 2022 with funding from
Kahle/Austin Foundation
https://archive.org/details/voicesinstrumentO000page
VOICES AND
INSTRUMENTS
Cir 102i
MIDDLE AGES
Frontispiece A fiddler, probably a minstrel, accompanying himself as he
sings. A twelfth-century sculpture from Cluny. Musée Ochier, Cluny.
VOICES AND
INSTRUMENTS
Or Tren
MIDDLE AGES
Instrumental practice
and songs in France 1100—1300
Christopher Page
Preface
Introduction
Notes
Bibliography
Index
For my parents and for Régine
We must all find our own doorway into the past. For some it is the study of
what can be seen in paintings or read in poems. For me the best entry has
always been through what is heard in music. The performance of old music
is perhaps our strangest way of confronting the past, for in order to do it we
must collaborate with the dead; when we hear a performance of a medieval
song we are hearing an echo of voices which, as if by some miracle, have not
been silenced despite the passage of seven or eight hundred years.
This book is an attempt to move a little closer to the source of that echo.
Of all medieval music, the monophonic songs of the troubadours and
trouvéres seem to have a special fascination for modern listeners—the
twelfth-century troubadour lyrics, for example, constitute the earliest
repertory of Western vernacular song in existence and lie at the centre of the
Occidental tradition of love-lyric. Yet for all their musical and poetic
interest, these songs sound very faintly now under the wind of seven or eight
centuries and little can be said with certainty about how they were
performed. Despite an enormous amount of scholarly effort, the question of
their rhythm will remain a matter of controversy until the great day when
musicologists will have the opportunity to consult the troubadours in
person.’ Nothing will ever be known for sure about the way in which
medieval singers paced and phrased them, about the vocal timbres which
they cultivated or their use of dynamic shading.
The past is nothing without its mystery, and these enigmas will probably
ensure that troubadour and trouvere songs will retain their fascination for
ever. Yet there is one area of performance practice where I believe the mist
needs to be cleared a little: the question of instrumental participation in the
delivery of these songs. In recent years performers and instrument-makers
have done a great deal to lift medieval instruments from the frozen silence of
church sculptures and from manuscript borders; many reproductions have
been made and some skilled performers have emerged. Now that the initial
burst of activity in this area has passed, along with the 1960s and 1970s
when it reached a peak, the time has come for a dispassionate assessment of
where musical instruments belong in the surviving repertories of music.
This book is an attempt to take a step in that direction. In writing it I
have been helped by many friends and colleagues, amongst whom it is a
pleasure to mention David Fallows and Stephen Haynes (both of whom
read parts of the book in draft and made many valuable suggestions), Ann
vii
Preface
Vill
Introduction
More often resting at some gentle place
Within the groves of Chivalry, I pipe
Among the Shepherds, with reposing Knights
Sit by a Fountain-side, and hear their tales.
Wordsworth’s image of the Middle Ages, when shepherds played their pipes
amidst ‘the groves of Chivalry’, evokes a world of pastoral innocence and
simple dignity where there is always music of instruments in the air. With
the rise of interest in the troubadours during the second half of the
eighteenth century, those were often the terms in which men of letters
imagined the medieval past. The poems and biographies of those Provengal
songwriters—virtually forgotten, at least in France, since the sixteenth
century—were a revelation to the literary public of Europe when Lacurne
de Sainte-Palaye published his Histoire Littéraire des Troubadours in 1774;' at a
time when any traveller in southern France could still hear the language of
the troubadours spoken even in high society,” this pioneering book showed
that Occitan had once been a great literary language cultivated by
flamboyant poets whose lyrics stood at the fountainhead of European love-
song.
This interest in the troubadours was accompanied by a developing
appreciation of the role which minstrels and professional entertainers had
played in medieval life. Between 1759 and 1798 a hoard of references to
histriones and joculatores was made available to readers in Mansi’s volumes of
ecclesiastical Concilia, and these documents left no doubt that the medieval
monk in his cloister, like the chivalrous lord in his keep, had drawn upon a
host of itinerant singers and instrumentalists for his entertainment.° At the
same time the revival of interest in medieval literature—especially in
England, where Chaucer had been continuously read since the fourteenth
century—prompted antiquarians such as Thomas Percy, Joseph Ritson and
Walter Scott to study the part played by minstrels in composing or
performing the romances and ballads which survived in manuscripts and
in oral tradition.* Their disagreements were sometimes public and acri-
ix
Introduction
monious, but they were agreed that ‘the groves of Chivalry’ had been
thronged with minstrels, especially singers and instrumentalists.
It was perhaps inevitable that these interests in troubadour lyric and
minstrelsy should coalesce, bringing eighteenth-century antiquaries to the
view that troubadour songs had been originally performed by minstrels as
combined vocal and instrumental music. Fresh discoveries only seemed to
confirm it. By 1774 substantial extracts from the Vidas (‘Lives’) of the
troubadours were available in paraphrase in Lacurne de Sainte-Palaye’s
volumes mentioned above, and one of these relates how Guiraut de Bornelh
went from court to court leading two singers (cantadors) with him, while
another records that the troubadour Perdigo ‘knew very well how to play
the fiddle’.° Such evidence seemed definitive, and by the time of Charles
Burney’s A General History of Music (1776-89) it was in order for the songs of
the troubadours to be presented as the first concerted art-music for voices
and instruments in Western history:°
It was about this time [cl119] that Provengal Poetry arrived at its
greatest point of perfection; and that it began to be sung to the sound
of instruments: for at this period Violars, or performers on the Vielle
and Viol; Juglars, or Flute-players; Musars, or players on other
instruments; and Comics, or Comedians, abounded all over Europe.
This view has recently been questioned by Hendrik van der Werf in his book
on the songs of the troubadours and trouvéres. In a few paragraphs he
clarifies the accompaniment question in a most useful way. The central
issue, he argues,® .
I take my point of departure from these lines. The task is not to establish
‘whether medieval singers ever sang to instrumental accompaniment’, for
there is a good deal of literary evidence that they sometimes did (see
Appendix 3, passim). My purpose is to distinguish the kinds of songs which
were considered more appropriate for instrumental accompaniment from
those considered less suitable for performance in that way.
When the issue is expressed in these terms, much of the literary evidence
for accompaniment is disqualified for being too vague; it does not reveal
what kind of song it is that lies behind such terms as chanson, son, lai, vers, and
so on.” I have therefore based the argument of this book upon the few
sources which refer to different kinds of songs in contexts where it is clear
what kinds of songs are meant, and where instrumental accompaniment is
mentioned in explicit terms. The pattern that emerges from these sources is
probably only a fragment of the whole truth, and nothing more can be based
upon it than the most tentative conclusions. In a situation such as this the
danger of exhausting the reader with qualifications and multiple hypotheses
is clearly very great; | have attempted to be brisk and plain as an invitation
to the reader to consider the issues afresh.
The book is in two parts. The first is an attempt to determine which kinds
of songs may have been associated with accompaniment, and which not.
The second gathers evidence relating to the kinds of accompanimént which
stringed instruments may have supplied in performances of monophonic
songs. This restriction of the field of view to stringed instruments seems
justified by the lack of any definite indications that wind instruments
participated in the performance of courtly monody during the Middle Ages.
In the Appendices I have gathered source material relating to various
matters which, though important, do not belong within the substance of my
argument. The first deals with the terminology of medieval instruments,
since no attempt to use literary evidence as I wish to use it can bejustified
without some clarification of the nomenclature employed by medieval
writers. The second offers a selective typology of musical references in Old
French literature such as is necessary, I believe, to interpret what comes
next in Appendix 3, a collection of literary references to musical
performance. Finally, Appendix 4 confronts an issue which bears directly
upon the sound of courtly monody: the string-materials used on medieval
instruments. Here I have attempted to establish a tentative chronology for
gut, metal, silk and horsehair strings as a guide to instrument-makers and
performers who wish to do in a much more effective way what this book
attempts to do: to capture sounds after a lapse of many centuries.
xl
~*~
+O 1 gtr sia
as “" j
ne Shoe
e ‘¥
me P
id ‘4 a a
fs 1thy
¢ é “Aes! . - O59 oh in
:
ry b..ee
rn =~ } T a! i! GR a
+ = ‘a Zz ty » Po Van ow
ii »} a> Rte
j 4 T - ’ 1 xa 7 ley
P, bk k 4 H tw}
urs rn é f 4 Ne
ie 3
7 at] “ere My
i ir 4 { ‘i ; “i 5
j , Die | vet 4 Ps mi
‘ re i v + Fee
s ns 1 * i Lm 4 he ary f
7
P wise af vk 44 ty Ree i dont ? eT
i 4 Pp wu hy 4A t Jf
srs —
YP Payot
i
ee oe Pert e7
; 4 4
Ot See
,
Lig}
£ t
iF
7¥ij-
2
tp?
j >,
- €taae
at 4 . Re
Tar
: y =
—: rraiyy ey,
+5
a toa ‘ ' t
bed aah
72 NIDseeres Se
we
at ae ieeeae eaarh »
ges if glegty®t 5
by a rl te vege
4 . j ‘ ei Jtis va ,
* h, ,
i ee ie , ie mt my
ae k fies e i . » bros yh, ven ~~
a
ae
ere.© |
Song of Roland
I
During the period 1050-1200 the arts of lyric poetry and string-playing
were transformed. The troubadours emerged with their range of lyric genres
in the vernacular and their highly self-conscious approach to the art of song-
writing. At the same time, bowed instruments proliferated in many shapes
and sizes throughout Europe and the fiddle became the chosen instrument
of the Provencal courtier. These are just some of the most obvious
manifestations of a transformation in the art of being a poet or an
instrumentalist. Yet the most vivid testimony to these changes is to be found
outside the rcalm of songs and instruments. It can be sought in the narrative
poetry of the period where the change from Epic to Romance embodies so
much of what is significant in the making of the Middle Ages.'
The early French epics, of which the Song of Roland is probably the earliest
and certainly the best-known, are often approached as if they were windows
onto the eleventh century. They reveal a generally harsh and ‘uncourtly’
feudal aristocracy, constantly engaged in tragic conflicts over rights and
inheritances. In this respect these Epics have often been contrasted with the
new genre of French vernacular romance (new, that is, around 1150), for the
tales of Arthur, Tristan and other heroes, so much admired in the later
twelfth century, mirror a new and ‘courteous’ life-style. We hear less of
draughty, torch-lit castles, and more of window-seats and private chimney-
corners in well-appointed chambers.
This is a simplified view, of course, for in many ways the genre of Epic
becomes increasingly hard to disentangle from Romance as the twelfth
century progresses. It may also be an unguarded view, for we have become
reluctant to look in the ‘mirror’ of literature for a reflection of life as it was
lived in the past. Yet as far as the history of music is concerned this
interpretation of the contrast between ‘feudal’ Epic and ‘courtly’ Romance
has much to offer. Anyone who brings an interest in music to the Song of
Roland (c1080) and then, let us say, to Thomas’s Roman de Horn (c1170), is
3
Songs and instruments
We need not concern ourselves yet with the meaning of the word lai; it
suffices here to note that the /ai is clearly a song which Horn both sings and
plays. Other passages reveal that the lai was composed by Baltof, a young
aristocrat and son of the king of Brittany, to celebrate his beautiful sister
Rigmel who is Horn’s beloved. The episode implies an audience closely
interested in songs and instrumental playing.*
There is nothing new about the figure of the amateur string-player in
royal society; we find him in Beowulf, for example, and much earlier still if
we cast our net wide enough.” What is new, however, is the tacit but
unmistakable admission in the Roman de Horn that refined love is the passion
which music most readily stimulates and feeds. There are no love-songs in
Beowulf, only lays of the ancient epic heroes of the North; yet in the Roman de
Horn there is a keen sense that Horn’s musicality is part ofa disposition to be
elegant and amorous which is the heart of courtliness as interpreted by
secular individuals.° -
This is not simply a change in the ethos of songs and instruments; it is
something new in the ethos of masculinity and in the awareness of male
beauty.’ The Epic knights of the Song of Roland may have magnificent
physiques, but they are generally shown to us as seen through the eyes of
their admiring male peers in the banqueting hall, in the council chamber or
on the battlefield. Women barely figure in these places and therefore the
Epic hero’s magnificence is generally without any sexual nuances. Here is
Ganelon before the war-council of Charlemagne in the Song of Roland:®
‘He was so fine that all his peers watch him’: this is a man seen by men.
What a difference, then, if we move forward a century and encounter
Thomas’s description of Horn. Here is a man whose beauty is seen through
the eyes of a woman who desires him to distraction:
‘He has long blonde hair so that none can equal him,
he has blue eyes, large, sweet and laughing to look upon
ladies; he has a fine nose and mouth to give kisses...’
Figure 1 The seal of Bertran II, count of Forcalquier, 1168. Paris, Arch.
nat., collection de sceaux Supplément 4512 et bis. Reproduced by permission.
and the poet dwells upon the sagesse of the pagan warrior Blancandrin: ‘He
was one of the wisest pagans; he was a knight well-endowed with the
qualities of a vassal; he was a man of valour to serve his lord.’!%
In some measure the lyric art of the troubadours and trouveéres probably
arose as a transference of the eloquence which a knight was expected to
display before his male peers into the realm of leisure passed in mixed
company. In some chivalric narratives both kinds of eloquence are
associated. In Folque de Candie, for example, Thibaut is said to be amez de
dames et sages de plaidier: ‘loved by ladies and wise in pleading a case’,'* and
this because of his beles paroles, his attractive eloquence.'” The fictional figure
of Thibaut finds its historical counterpart in Conon de Béthune, the
knightly trouvére who was active in the late twelfth and early thirteenth
centuries. In 1203 we find him conducting delicate negotiations at
Constantinople on behalf of the Crusaders; he was chosen, it seems, because
he was sages et bien emparlez; ‘wise and most eloquent’.'®
Epic and Romance
I]
We have seen that the arts of string-playing and song-writing were endowed
with a new ethos in the twelfth century. These were clearly important arts in
secular society and it matters that we should have some understanding of
how they were combined.
In some respects the issues before us may seem simple enough and even
half-decided. Surely the existence of troubadour fiddlers like Guiraut de
Cabreira and Pons de Capdoill shows that the marriage of courtly song to
instruments was a natural and easy one. Perhaps it was, but there are signs
that performers were not always faithful to it. Here, for example, is a young
member of a courtly household, a vallet, performing a song by a noted
trouvere, the Vidame de Chartres (Music example 1). A king, a single
knight and a minstrel named Jouglet are listening:'’
Where did instrumental sounds and techniques fit into the genre system
of Old French and Old Provencal lyric? Practices that were appropriate in a
performance of aHigh Style song like Quant la sesons del douz tens s’asseitre may
not have been suitable for an anonymous (but no less courtly) refrain-song
intended for dancing. We cannot hope to rebuild these, or any other,
medieval traditions of performance practice. Yet I believe that we can
sometimes glimpse, as if in an aerial photograph, faint traces of the lines
which their foundations followed.
10
Epic and Romance
Example 1
e
_——— =
Pah eX z oe
6 eo Fe so, o= ° a. ao
© = ee eS
oo o fa ° z o © a6 = on Fe
-
* = o>: r= i = = SSS
ee chan - ter
o >
m'es
- tuet,
e
car
=
plus
°
ne
ee
m'en puis
5
tai - re,
— 7aaa
Ge = SS Sa a ae so =
= * =?
oa =
—,
oe oy re
:
11
a
The twelfth century
in the South
€ auziratz, si com yeu fi,
als trobadors dir e comtar
si com vivion per anar
e per sercar terras e locx.
I
Any attempt to place instrumental accompaniment within the genre system
of troubadour and trouvére poetry must take its bearings from the genre
which dominates the surviving corpus of lyrics: the elaborate love-song
which Dragonetti has termed the grand chant courtois. I shall also refer to it as
the High Style song. Throughout the later twelfth and thirteenth centuries
this kind of song, which the trouveres inherited from their predecessors in
Occitania, was the ‘classic’ form of the courtly songwriter’s art. As an
example of the prevailing poetic and musical manner of the High Style, here
is the first stanza of asong by Arnaut de Maroill:!
Example2
—= = aw. x Za aaa =
=~ eo j= ——
= eo 2 i rae
6 FE v a= ros oS e
— = = z Saat >
==
= on ones =
3. e.ls au - tres ditz e la fres -ca co - lors
6 = s ° 2 i ..
Res = ze
4 =—— a
a oe
aaa aa
os ° ee x a a
5. me do - nan- genh de chan - tar e sci - en - sa,
a —-
$= _ ° —o e eo ten = ae
Like many other melodies for High Style songs this one is essentially
rhapsodic in character. Lines | and 2 are set to the ‘same’ music as lines 3
and 4, but this is a conventional patterning in this repertory and in this
instance (as often) it is far from strict; line 2 has a ‘closed’ ending but 4 has
an ‘open’ one. There are other relationships, but the essence of the
rhapsodic musical style of the grand chant courtois is that they are disguised.
Thus there are significant resemblances between the settings oflines 1/3, 2
and 6:
13
Songs and instruments
Example3
_———— ee
fe oe = oe = ° _ > = ¥ a sy =
2 Zz aes
6 re e a = = =e oe eo .
while the music for lines 2 and 4 is closely related to the melody for line 7:
Example 4
6 zi a oe SS 6 ——
ee % = —
ee
7 —_ ——
> o —o ° ° e os = oe
Example 5
——— aS SS
SSS
———— =
SSS
S52
SSS
Whence the characteristic manner of the High Style song: neither
gregarious in impulse nor indulgent towards its listeners, it usually lacks any
choric refrains which might invite us across the space that separates us from
the singer and draw us into the song. Arnaut de Maroill maintains the
distance between singer and hearer by a great show of decorum achieved by
an accumulation of potent courtly words (verais, valen) and by strewing his
poem with connectives that give it the appearance of discursiveness and
debate (quand...car...perque...). Presented in this way the grand chant
courtois becomes a form of oration and the antithesis of a ‘performerless’
dancing song. Indeed the idea of the song as the composition of a self-
conscious artist is constantly kept in the listener’s mind and is a crucial
element of the grand chant manner:°
1D
Songs and instruments
POETRY
MUSIC
17
Songs and instruments
I]
Where did instrumental accompaniment belong in this system of contrasts?
The search for an answer begins with an aristocratic musician whom we
encountered in Chapter 1. He is the Catalan nobleman, Guiraut de
Cabreira, ‘dashing in warfare ... gracious in manners ... [and] highly
skilled on musical instruments’.
It may have been as early as 1150 that Guiraut qddpecsed a mocking
sirventes to a certain minstrel named Cabra (?Guiraut himself), playfully
rebuking him for the accomplishments he had not mastered. Guiraut’s own
mastery of the fiddle—remembered long after his death by those who te)
heard him play—sharpens the edge of what is already a pointed sirventes.*
‘Whoever taught you to bow and finger the fiddle made a bad job of it’, he
proclaims to the unfortunate minstrel in the hearing of all; ‘you fiddle badly
and sing worse, from start to finish’:°
At one point Guiraut rebukes Cabra for not being familiar with a variety
of lyric genres, including the szrventesc, estribotz, retroencha, contenson and
balaresc—a useful but rather featureless list which does not mention
particular works or name individual troubadours. Yet, in the lines which
follow, Guiraut finds a more specific charge to level against Cabra: he
cannot sing songs by the famous troubadours of the last and the present
generation: Jaufre Rudel, Marcabrun, Ebles II de Ventadorn, a certain
Anfos, and ‘en Egun’ (perhaps a mocking pun on negun, ‘nobody’):°
This suggests that some minstrels who fiddled and sang for a living
included songs by celebrated troubadours in their repertory. As we might
expect from one who was probably a younger contemporary of both
Marcabrun and Jaufre Rudel, Guiraut de Cabreira refers to their songs in
the correct way; by referring to their vers novel he is following the usage of
both Marcabrun:’
Ill
After Guiraut de Cabreira’s harangue against the unfortunate minstrel
Cabra (whose bad luck must have begun when he took on the professional
name ‘Goat’), it is a relief to turn to the roughly contemporary epic of Daurel
et Beton and two passages which may bring us a little closer to the kind of
information which we seek.!°
The story of this epic tells how Daurel, a minstrel, sacrifices his own son
to save his boy-lord, Beton, from being massacred by a French duke named
Guy. To guarantee Beton’s safety Daurel sets off with the boy in a boat, and
this stage of the narrative brings the first musical reference. When the child
cries during their hazardous and uncertain voyage, Daurel takes his fiddle
(viola) and"!
...fai.i1. lais d’amor.
In the first extract the term /ais is a suggestive one whose meanings in Old
Provengal are many and various, often amounting to little more than ‘a
song’, ‘a tune (whether vocal or instrumental)’.!? The context shows that
Daurel’s /ais is played upon the fiddle, and that it is a love-song, .i. lais
damor, which he is presumably either singing to his own accompaniment or
Example6
— —~ =
= =
a er = A SS ee ee
S25 —~ =
2 ae eye
7 aR
© # ° = ~ = =, Te ——
=—_ ©6 6
mas trop me fay m'a - ven - tu - ra do. .= J ler,
—_ ee 6 ca e ee ee
cant ieu es - gart los bes e.ls mals qu'ieu ay;
D
——— Se
que ricx dis hom que soy e que be.m vay °
ra
% re ee
#6» Fie SS
° ed e # eo owe
aad ——
a = 2 mi ° ——~? = a2 , ae
21
Songs and instruments
playing as instrumental music to soothe the troubled child. The likelihood
that Daurel’s Jais is a lyric lai—a polymorphous lyric where each sub-
division of the text has its own metrical form and musical setting—seems
remote at this date (possibly as early as cl1150). However, /ais could also be
applied to lyrics in the High Style, including the canso. Folquet de Marseilla
(d.1231) calls his song S’al cor plagues be fora huei may sazos a lais, even though
its subject matter and musico-poetic form mark it as a canso in the High
Style tradition (see Music example 6).'*
The lais d’amor played by Daurel to comfort the young Beton may be a
High Style troubadour canso sung to the fiddle, although it seems to have
been an unusual practice for a troubadour to refer to his words, his music, or
both together as a Jais. ©
In the second extract Beton’s pieces are called vers, the prevailing term for
a troubadour lyric until the later twelfth century when canso began to make
an appearance.'° Unfortunately it is not clear whether Beton is fiddling and
singing at the same time, or whether the verses in his repertory are to be
identified with troubadour songs such as those by Jaufre Rudel, since vers
seems to have been used by the early troubadours for any kind of poem.'°
On balance Daurel et Beton only seems to offer hints in whispers.
A brief passage from jJaufre (cl1170) speaks in bolder tones. In this
romance, the only Arthurian verse narrative in Old Provengal, there is a
description of a court celebration where!’
The suns are probably instrumental melodies, for so (<Latin sonus) in the
sense ‘a tune, a melody’ is attested in numerous troubadour lyrics (where
the so, or melody, of a lyric composition may be contrasted with the
words, the vers or motz).'® Perhaps the same meaning may be assigned to lais,
given the frequency with which this word carries the sense hiner whether
vocal or instrumental’ in Old Provengal.
As for descortz and dansas, there is another twelfth-century source which
joins these two terms to the minstrel’s fiddle: a partimen with stanzas by the
troubadours Raimbaut de Vaqueiras, Perdigo, and the nobleman Adémar
II of Poitiers, Count of Valentinois and Diois, 1188—1230. The first tornada
of this poem opens:'®
a2
The twelfth century in the South
This double confirmation of a link between the fiddle and the genres of
descort and dansa is striking and surely significant. Both of these lyric forms
may be said to occupy a very minor position in the corpus of troubadour
lyrics with music. Only a handful of descort settings have come down to us
with music”’ and only a few dansas, all of which are written in various forms
of Franco-Occitan and preserved as additions of the early fourteenth
century to the Manuscrit du Roi (Figure 2). In terms of survival, therefore,
this association between the fiddle and the genres of descort and dansa carries
instruments to the very periphery of the troubadour corpus.
Patterns of survival can be the result of accident or caprice, but there is
another sense in which both the descort and the dansa lie on the edge of the
troubadour art. Each one, in its own way, subverts the High Style manner
as represented above all by the canso. The descort was a polymorphous lyric in
which each subdivision of the text had its own metrical form and musical
setting. Its form therefore subverted the dignity of the canso in which each
stanza has the same form and the same melodic setting. As Baum has
argued, the descort is not simply unlike the canso; it is the anti-canso:*'
dansas of the twelfth century have passed into eternal silence, for not one
survives—neither words nor-music—but their poetic form is probably
preserved in the dansas which began to be written down in the next century
(none of which are much older than c1250). Here is the outline of that form:
If the respos were repeated at the end, then the result could be identical to
the French virelai:””
A bbaA
A glance at the list of High Style and Lower Style characteristics on p.16
is enough to show that the dansa belongs in the region of the Lower Styles.
The dansa, as its name implies, was a lyrico-choreographic form which must
often have been performed for dancing; on those occasions, at least, it will
have been performed in some kind of strict metre. When the respos was
repeated at the end ofthe stanza the dansa had a refrain (in the strict sense: a
combined musical and textual reprise); at all times the dansa had a
conspicuous musical reprise (A b a). In all these respects it is a Lower Style
form.
But what of the texts? Here we may be able to pursue the link between
instruments and unwritten repertory a little further. For the most part the
surviving dansa poems are courtly in the sense that they are love lyrics which
exploit the conventions ofliterary love whose natural home is the canso. Yet
no twelfth-century dansas seem to have survived, and there are no dansas
surviving among the works of the major thirteenth-century troubadours.”°
One explanation for this state of affairs is that the dansa began life as a
trivial, ephemeral and perhaps even popular form: one which lay too low for
the troubadours to cultivate. This would not stop dansas from being played
and enjoyed at court, but it would prevent them from becoming the vehicle
of a High Style lyric genre.
There are traces here and there of what these unrecorded dansas may have
been like. A song by the Catalan troubadour Guillem de Bergueda (d.1192/
6) claims to take its melody from a song sung ‘by the young men of Pau’,
and the verses end with a (slightly variable) refrain which may well
incorporate material from a dance-song:**
24
The twelfth century in the South
Varalalito
varalalito
deu!
varalatitondeyna.
IV
In this poem the narrator tells how he walked out one morning when ‘April
was leaving and May entering’. Low in spirits, he enters the town square of
25
Songs and instruments
26
The twelfth century in the South
place to place. In the vpinion of Raimon Vidal this was one way that an
entertainer could help to maintain and stimulate aristocratic courtesy, for
courtesy was the common culture of a network of cortz and it was the
privilege of minstrels, who travelled regularly between these courts, to
report which lords of the region were the most courteous, and which ladies
the most virtuous, as an example to all. Conversation such as this was the
minstrel’s first deployment of his saber. Next, if he saw that the company was
with him and eager for more, he turned to novas, or stories in verse which
retained the themes of courtesy and noble deeds but which represented a
move away from conversation as the company (which the minstrel had by
now turned into an audience) fell silent to listen to his material in rhyme.
Once the novas were over, the minstrel reached the most sensitive point in
the management of his audience. His next and final stage was to offer to
sing, but it was only worth proceeding to chantars if the audience had shown
itself adreitz, prims and entendens—to be clever, of fine sensibility and
discriminating. This, it seems, was the kind of audience, moulded and tested
at each stage by a skilful performer, which proud joglars sought for the
performance of songs by troubadours such as Guiraut de Bornelh and
Arnaut de Maroill. Raimon Vidal seems to regard all his skills as facets of
the same gift—the gift of eloquence. Whether it be the improvised
eloquence of conversation, or the prepared eloquence of tales and
troubadour songs, a good minstrel sees himself as one who, in all the
manifestations of his art, shows how one should speak to the pleasure and
moral profit of others. He does not emerge as a manually skilled or
dexterous individual, whether the skills be those of the knife-thrower, the
juggler or the instrumentalist.
The evidence assembled in this chapter might suggest that the High Style
troubadour canso was not generally accompanied in the twelfth century.
However, we have found no positive evidence for that view, for none of the
texts that we have so far encountered mentions the High Style canso in clear
and unmistakable terms (although a passage from Daurel et Beton may do
so; see above). All that can be said is that as early as the second half of the
twelfth century there are signs of an association between instruments and
certain song-forms (the descort and the dansa) which, in their own way, stand
well apart from the High Style canso.
28
a.
The thirteenth century
in the North
[Telamon] fu mout de grant valor.
Mout ot en lui bon chanteor,
Mout aveit la voiz haute et clere
Et de sonez ert bons trovere.
These lines from the Roman de Troie show that the courtly singer-trouvére
was a recognisable figure in northern France by cl1165. In some ways this is
a remarkably early date; we will not hear of any named trouvere for several
decades yet; there will be no written source of the trouvere art for almost a
hundred years. Yet Benoit de Sainte Maure shows that a male courtier who
was a good chanteor and trovere could expect to be admired in French society
less than a generation after the earliest known troubadours such as
Marcabrun and Jaufre Rudel.
In terms of lyric production, the half century from 1170 to 1220 was a
brilliant one in northern France. Yet in terms of information about the way
in which those lyrics were performed, it is a dark period; not until the very
end of it does the obscurity surrounding instrumental practice in trouvere
song begin to recede. Yet although the way is dark during those years, it is
not impossible to follow, for the main path must lead from south to north,
from the cortz of Occitania to the lands of magnates in France. The central
genre of the northern trouveres, the chanson, was directly based upon the
canso of the southern troubadours,' passing easily from Occitan to French in
the repertory of travelling performers who found a willing audience for these
and many other songs from Occitania which the northern barons lumped
together as ‘songs from the Poitou’, chansons poitevines. We can even see the
process at work in some of the twelfth-century epics; in Garin le Loheren,
amidst an evocation of a fierce warrior society, we suddenly encounter
minstrels who*
Vielent, notent maint bel son poitevin.
29
Songs and instruments
accompaniment as they ride, and the piece which they perform is twice
described as a son d’amour.
Gautier de Tournai’s passage, arresting though it may be, is not the only
description of its kind to survive. Here is another, this time from the
thirteenth-century epic of Hervis de Metz (Appendix 3:29):
Jean Renart’s romance of Guillaume de Dole (c1220), where over forty lyric
songs are quoted or cited, also suggests that the term son had meridional
associations for French speakers. Renart reserves it for High Style songs by
the troubadours Bernart de Ventadorn and Jaufre Rudel. His only
departure from this usage takes us back to the Auvergne, for he calls the
following High Style canso, attributed to the troubadour Daude de Pradas in
one of its three sources, a chancgon auvrignace. Here it is with the heavily
Gallicised text that Renart gives (Music example 7).*
31
Songs and instruments
Example 7
> ~¢ -e oo =- = eo a GES
—_
e =e ® = o = = a
et l'er - be nest en la sa - ne
oe
Zedont a ee
ra - ver - dis - _ sent cil ver - gier.
=
® a—=e=
Ge —e o ==c # Se
=
ons >—~
2 = ee ee
=o4= = ee
que cors me ga -_ rist et sa - ne
ure
The thirteenth century in the North
nately, there is one such source compiled by an author who had heard
chansons poitevines and who was a connoisseur of the whole range of
monophonic lyrics known in France, from little dancing songs up to chansons
by trouvéres such as the Chastelain de Couci and the Vidame de Chartres.
His decision to insert lyrics into a romance narrative of cl220 provides an
outstanding source of information about instrumental practice in northern
courtly monody, and one which suggests that performing traditions did
indeed pass northwards along the roads which linked Occitania to Fransa.
This is the romance of Guillaume de Dole compiled by Jean Renart in the first
decades of the thirteenth century. Jean Renart seems to have initiated the
fashion of inserting lyric poems into romance narrative; his poem contains
more than forty lyrics which, taken together, provide a conspectus of the
song-forms known in his day.’ They include sixteen High Style songs, six
chansons de toile (see below), two pastourelles and twenty refrain-songs
comparable to C’est la jus en la praele (see below). Since they are
presented as arising from the story—being sung by characters in the tale—
Jean Renart is committed to some narrative involvement with the way in
which they are performed.
In three cases Jean Renart refers to the presence of instrumental
accompaniment (which, in every case, is provided by a minstrel upon his
fiddle), and if we put the accompanied songs together then we begin to
discern various patterns.
The first accompanied song is a rondet beginning C’est la jus en la praele.
The context of the performance is informal and festive, for the piece is sung
at the hostel near the palaceof Conrad, Emperor of Germany, where
Guillaume de Dole is staying. Guillaume and his party return from a night
of entertainment at the palace taking with them Jouglet, a minstrel. Jouglet
is commanded to carry his viele back to the hostel in response to a request
made earlier that same evening by the daughter of the hostel-keeper who
had playfully upbraided the minstrel: ‘you have sung nothing, Jouglet, since
you came through our door’, offering to pardon the omission if he would
return with his fiddle that night. ‘I shall do so willingly’, replies Jouglet, ‘if
we are to have caroles’ (1566—7; 1799-1801). When the party arrives at the
hostel (which is clearly no country inn but a well-appointed lodging which
the Emperor judges to be appropriate for his guests) they seek out the
owner’s daughter. Once they have ascended to the upper floor where the
valets have laid out fruit and wine, the girl sings the rondet to the
accompaniment ofJouglet’s fiddle:®
A glance is enough to show that this chanconete (whose music does not
survive) is an example of what we have called a Lower Style; indeed it is
exactly the kind of lyric which was used to illustrate Lower Style
characteristics in the previous chapter: a light, informal song whose brisk
and essentially trivial character is caught, in this instance, by the diminutive
suffix in changonete. There are many features which set this poem apart from
the High Style: the refrain; the pastoral setting signalled by a formulaic,
half-narrative C’est la jus en la praele; the mention of the beloved’s name (a
gesture which is foreign to the decorum of the High Style, unless the lady be
referred to by some enigmatic name which the poet has given her); the use of
the conventionalised rustic name Perronele; the implication that the speaker
himself is of Perronele’s social class; and the evocation of an insouciant
rustic life (Perronele rinsing her clothes).
_ The second of the songs performed to instrumental accompaniment in
Jean Renart’s romance is Bele Aiglentine, a chanson de toile sung by a young
Norman squire as he rides into the fields before a great tourney. Once again
the accompaniment is provided by the minstrel Jouglet on his fiddle. He is
commanded to play by the squire (si la fist Jouglet vieler).? Bele Aiglentine
presents many of the characteristic features of the chanson de toile (literally ‘a
song to sing while sewing cloth’). It is narrative, opening as the heroine,
Aiglentine, converses with her mother whilst sewing. She is busied with
other thoughts and proves so inattentive to her work that she pricks her
finger with the needle. Her mother notices this and suspects that Aiglentine
may be troubled in love—indeed that she may be pregnant:’”
Aiglentine admits that she has fallen in love with Count Henri. ‘Will he
not have you?’ asks Aiglentine’s mother; ‘I do not know,’ replies Aiglentine
disarmingly, ‘for I have never asked him’. On her mother’s prompting,
Aiglentine goes to visit Henri in his ostel where she finds him in bed and asks
39
Songs and instruments
him if he will marry her. Henri gladly accepts, and the final verse of the
poem runs:!!
Here we have a fully narrative poem whose stanzas (except the last) end
with a refrain which calls us into the tale with an almost minstrelish
insistence:
Or orrez ja
conment la bele Aiglentine esploita.
Bele Aiglentine is characterised not only by its narrative framework but also
by its para-folkloric tone. The tone is established by the archetypal nature of
the characters (a sorrowing pregnant maid, an anxious mother, a lover),
and by the incremental repetition of the narrative technique—as when
Aiglentine’s mother calls to her at the beginning of one strophe:
This device of completing an action with the words in which the action
was commanded or predicted figures in many traditional forms of narrative
(English balladry provides a parallel); it contributes to an impression of the
world as a place where events have a momentum of their own, largely
independent of human will and choice. Yet it does so without establishing
any mature sense of tragedy or even of fate. Motivation and causality
become cryptic and the narrative can be pared down to its essentials,
leaving the reader with a stirring sense of mystery. Who is Aiglentine? When
and where does her story take place? Why does she wait for her mother to
discover her pregnancy before speaking to Count Henri since he seems quite
happy to marry her? Why is Count Henri abed when Aiglentine visits him?
36
The thirteenth century in the North
At the same time the direct speech assumes a para-folkloric tone with a
stylised and sometimes proverbial manner—as in the stanza quoted above
when Aiglentine’s mother asks to see beneath her dress:
Cele d’Oisseri
ne met en oubli
que n’aille au cembel.
Tanta bien en li
que mout embeli
le gieu soz l’ormel.
En son chief ot chapel
de roses fres novel.
Face ot fresche, colorie,
vairs oils, cler vis simple et bel.
Por les autres fere envie
i porta maint bel joel.
This is an enigmatic little song to say the least. Jean Renart describes it as
une dance composed by maidens of France to commemorate the toils of a
certain bele Marguerite. This is quite plausible since there are references to
commemorative dance-songs of this kind scattered here and there in
medieval writings from at least the ninth century on.
These are the songs which Jean Renart associates with instrumental
accompaniment in Guillaume de Dole. Several points can now be firmly
37
Songs and instruments
stated. Firstly, only three of the forty-six songs quoted or excerpted in the
romance are associated with accompaniment, a proportion of one in fifteen.
Secondly, the accompaniment is provided in every case by a solo fiddle, a
circumstance which suggests a degree of standardisation of performance
practice (since the same, solo instrument is used each time) and also betrays
a certain sobriety of taste; Guillaume de Dole provides no evidence for the
mixed bands of wind, stringed and percussion instruments which have
figured so prominently in modern recordings of this repertory.
But the most important conclusion to emerge from this survey is that in
poetic terms, at least, every accompanied song in Guillaume de Dole shows
pronounced Lower Style characteristics (the music of these songs does not
survive). There is a strong link with dance, with narrative and with refrain-
form. There is no instance in the romance where accompaniment is
associated with a High Style poetic manner.'®
This picture, if an accurate one, seems to accord with the evidence of the
Old Proveng¢al sources examined in the last chapter. There we found that
instrumental accompaniment seems to have been particularly associated
with the descort and the dansa, both of them genres which, in their own way,
subverted the High Style manner. With Jean Renart’s Guillaume de Dole,
instruments seem to lie on the periphery of the surviving corpus of trouvere
monody.
The distinction between High and Lower Styles in lyric, together with the
patterns of instrumental usage appropriate to them, is not a distinction
between what is courtly on one hand and what is uncourtly on the other. 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
primaveral ethos of most aristocratic dancing lay very close to the essence of
court-culture in the twelfth and thirteenth centuries and characterises the
scene where the Catalan troubadour Guiraut de Cabreira plays the fiddle
for dancing in the palace at Arles.'*
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 trouveres resolutely refused to do. If we have
uncovered a general principle here it is a very general one and seems to be
this: a song was less appropriate for instrumental accompaniment the more
it lay claim to self-conscious artistry.
We may be confronting a fundamental characteristic of twelfth- and
thirteenth-century string-playing here: instrumental music does not seem to
have been associated with the kind of profound creative endeavour which
demanded serious and considered attention from the listener. Time and
time again, medieval literary sources convey the impression that string-
playing was spontaneous, ephemeral and associated with very few emotion-
38
The thirteenth century in the North
39
=
Late traditions in
the South
Dansa...deu [haver] so plazent; e la
ditz hom ab esturmens...
I
The value of Jean Renart’s romance of Guillaume de Dole, explored in the
previous chapter, is that Renart’s narrative allows us to establish whether
instrumental accompaniment is meant and precisely what kind of song 1is
being accompanied. In the final analysis this is the only kind of evidence
upon which we can hope to build an understanding of instrumental
participation in courtly monody, and since so little of it survives, every scrap
counts, no matter how late or peripheral it may seem. The purpose of this
chapter is to examine two southern sources which offer explicit information
about the place of instruments in a system of lyric genres, but which are very
late for our purposes, perhaps even too late. Both are treatises on the writing
of lyric poetry in Occitan: the Catalan tract entitled Doctrina de Compondre
Dictats, probably of cl1300, and the annotations to the A text of the
celebrated Leys d’Amors, compiled in Toulouse during the first half of the
fourteenth century. As far as instrumental accompaniment is concerned, we
shall find some grounds for believing that these treatises endorse the picture
traced in Chapters 2 and 3.
The Doctrina de Compondre Dictats was probably composed by Jofre de
Foixa as a supplement to his Regles de Trobar, commissioned by Jacme II
when king of Sicily (Jofre is known to have had connections with the
Catalan court in Sicily after 1289).! The methods and treatment of material
employed in the Regles de Trobar?
Jofre’s passing comment that the dansa ‘is sung with instruments’ carries
some weight in the full context of his treatise, for no other lyric form in the
Doctrina is associated with instruments in this way. Here we seem to have a
point of contact with performing traditions reaching as far back as the last
4]
Songs and instruments
decades of the twelfth century, for the dansa is associated with the fiddle in
two Old Provengal texts produced in that period, as we saw in Chapter 2. In
that chapter it was suggested that the dansa (and the descort) may have been
distinctive among the lyric genres known to the twelfth-century troubadours
for their association with instrumental accompaniment (or instrumental
performance); I went on to propose that this was one aspect of a
distinctiveness of form and poetic content which set both descort and dansa
apart from other lyric genres. As far as the descort is concerned, Jofre at least
knows that it stands out on a limb for its melody, ‘the contrary of all other
types of song’, as he puts it,’ while the distinctiveness of the dansa seems
apparent in his treatise for exactly the reason which twelfth-century sources
have led us to expect: it is linked to instruments.
Fortunately it is possible to form a relatively clear impression of what the
dansas known toJofre were like, for it is in early fourteenth-century additions
to an important French chansonnier, the Manuscrit du Roi, that we
encounter dansas with their music for the first (and last) time (see Music
example 8).°
I]
By way of Jofre de Foixa’s treatise we come to the most exhaustive and
systematic of all treatises on troubadour poetry, the Leys d’Amors (‘laws of
love’), compiled, for the most part, during the second quarter of the
fourteenth century by Guillem Molinier.° This treatise, which exists in
several versions, was produced for the Academy of the Gai Saber at
Toulouse, an association which sought to encourage the composition oflyric
poetry in Occitan by organising poetic competitions, probably on the model
of the northern French puis, and the purpose of the Leys d’Amors is to reduce
the art of writing lyric poetry to rule so that it may be taught, propagated
and judged.
Like Jofre de Foixa, Molinier and his collaborators in the Toulouse
Academy can have had no direct knowledge of earlier traditions of
performance-practice, but this does not affect the issue of whether the
practices of their own day—aquesta prezen art—enshrined long-established
conventions—long uzatge acostumat. Our judgement of that issue (insofar as
we can hope to judge it) must rest upon a comparison of the Leys d’Amors
with the two independent sources which we have already examined: Jofre’s
Doctrina and Jean Renart’s Guillaume de Dole.
In an important section of the Leys Molinier describes the characteristic
subject matter and metrical forms associated with each genre of poem,
usually adding a brief note on the kind of melody associated with them. In
this way he works through the principal genres, or dictatz principals, a
category which embraces the traditionally significant troubadour genres.
There is no reference anywhere to instrumental accompaniment or to purely
instrumental forms, and an anonymous member of the Toulouse Academy,
42
Late traditions in the South
Example8
SSS
_—— === =e a
fatz -me bel sem - blant, gutieu —_ sui to -
SSS x
vos
eee: eS
eee
$4 Sag cape aeae TS
ota te Besaers HS =
Se
with access to the manuscript, seems to have been struck by the omission.
To remedy it he added an extensive marginal note (Figure 3) which
mentions the purely instrumental garips and also the estampida (which, he
explains, may be either a purely hele es melody or an instrumental
tune to which words have been added).’ He also mentions the bal, a form
43
Songs and instruments
This was the kind of melody required for a vers, but also for a chansos, a
partimen, a sirventes, a tenso (if it is to have music at all), a retroncha and a plang
(although Molinier does not entirely approve of the plang having a tune of
this kind).'' So it seems that the passage quoted above describes the musical
style and ethos of High Style poetry as the members of the Consistori knew
it.
‘Ethos’ seems the appropriate word in this context since Molinier is
attempting to define this musical style in terms of both its aesthetic effect
(belas passadas e plazens pauzas) and its objective properties (so noel). As every
music critic knows to his cost, the character of musical sound cannot
satisfactorily be captured in language; whence it is no discredit to Molinier
if some of his references to melody seem impressionistic, or if some of his
meaning must remain in doubt (is a lonc so a ‘long’ tune, a ‘sedate’ one, or
something else again?). Yet we can salvage this much: Molinier viewed the
performance of these major genres as a heightened form of declamation in
which singers, like trained readers, pointed the sense of their texts with
pauzas, matching the High Style of the words with a restrained and imposing
style of delivery.
So much for the vers, chanso, sirventes, tenso, partimen and plang. With the
pastorela this musical register changes: the melody of this genre should be
‘not so slow as that ofa vers or chansos, but it should have a tune which is a
little more fast and lively’.'* This fits well with the light tone, outdoor setting
and semi-narrative content of the pastorela and its northern equivalent, the
pastourelle.'°
At the other end of the spectrum in the Leys we find the dance-related
genres, most of which seem to be associated with instruments. In a
particularly revealing passage the anonymous annotator records that the
tune of a bal should have more short notes and be more likely than that ofa
dansa (so mays minimat e viacier), whence the bal is ‘more apt for singing to
instrumental accompaniment than the dansa’.'* This suggests that the
quality which made a song more apt for accompaniment in the view of the
Toulouse Consistori was a pronounced gaiety and dance-like character,
quite in order in certain dictatz no-principals but foreign to the decorum of the
High Style and principal genres.
Turning back to the Doctrina de Compondre Dictats we can now see that
Jofre de Foixa reveals something similar. Writing of the plant, which must
45
Songs and instruments
treat d’amor o de tristor, Jofre says that such a poem may be set to ‘any kind of
melody you like, save that of the danga’.'° This surely implies that the music of
the danga—the only genre with which Jofre associates instrumental accom-
paniment—was somehow distinctive and unlike that of any other lyric
genre.
Here, then, is the aesthetic spectrum of the Leys d’Amors (including the
annotations) showing the position where instrumental accompaniment and
performance seem to be located:
f.41v ANNOTATOR
mays minimat e INSTRUMENTS
viacier [que dansa] HERE
e mays apte per Bal
cantar amb esturmens
que dansa
46
Late traditions in the South
III
A pattern has emerged from Guillaume de Dole, the Doctrina de Compondre
Dictats and the A text of the Leys d’Amors, the most explicit vernacular
sources that bear upon the problem ofinstrumental accompaniment. With
the Leys d’Amors this pattern has formed itself into a musico-aesthetic
spectrum in which instruments are associated with songs (or instrumental
pieces) whose music is of a predominantly joyful character, and which have
a refrain (or a musical reprise of some kind) often linked to dance.
These conclusions help us to refine some ideas about the aesthetic status
of instrumental sound which were floated at the end of the last chapter. It
seems that during the twelfth and thirteenth centuries instrumental sound
had its own aesthetic colouring—at least when considered as a potential
accompaniment to songs. It was not a colouring which could be painted
onto any kind of lyric. This sense of propriety—of performing decorum—is
not difficult to account for. At this time the apprehension of instrumental
sonorities was in a ‘pre-composer’ phase, for instrumental colours did not lie
on the composer’s palette waiting to be combined for whatever expressive
ends he desired; nobody composed in that way during the Middle Ages (or
so I assume). Instrumental sound therefore had its own, closely-defined,
expressive range according to the kinds of music which, by convention, it
was directed to perform, and according to the kinds of occasion which, again
by convention, it was called upon to adorn.
During the period that we have been examining, this expressive range
seems to have been strongly influenced by the ethos of vivacious dance-
music. According to the annotator of the Leys d’Amors it was because the bal
had a rather more vivacious tune with more short notes than the dansa that it
was ‘more appropriate’ for instrumental accompaniment. Since a high
proportion of medieval instrumentalists were professionals whose livelihood
depended upon their ability to rouse a courtly company to festiveness, and
especially to dance, it is apparent that much instrumental-playing of the
time must have been gay, tuneful, and closely linked to dance. Virtually all
of the medieval dance-music which survives, whether in the form of textless
(and therefore presumably instrumental) pieces, or of songs which could be
sung for dancing and accompanied, is characterised by an instant tuneful-
ness achieved by conspicuous, short-range melodic patternings (see Music
example 9).'°
Only one piece of instrumental music survives from before the thirteenth
century and it bears out these observations. According to a famous account
in the Razos (explanations of how certain troubadour songs came to be
composed), Raimbaut de Vaqueiras composed his lyric Kalenda maya to the
tune of an estampida played at the court of the marquis of Montferrat by two
fiddlers (Appendix 3:11). Fortunately, Raimbaut decided to write a High
Style canso to fit the fiddlers’ melody. The combination ofa High Style text
and a Lower Style melody is part of the wit of the piece—for it is a
ay
Songs and instruments
Example 9
jos SS
SaaS a ==
soil SSS
Se
5aoe ec SS
Example10
= eser b =
re o — = —— =sere eee
= 4 4
oe 9 ==: ee = Se
$= eo
§ 2.non truepque.mplaya pros do-na ga-ya tro c'unir-nel mes-sa-tje n'a - ya
———
2 ee Ss
8 eo
O26 aX
é eo o oo eo ey.
‘ oe oe —
5.qu'ieua-ya, e.m tra-ya vas vos,do-na ve - ra - ya,
—————— a wee ee
8 ——,
48
Late traditions in the South
49
pol
Paris ~.
Bonus autem artifex in viella omnem
cantum et cantilenam et omnem formam
musicalem generaliter introducit.
I
It is time to introduce two Parisian authors of the thirteenth century with
much to offer: the music theorists Johannes de Grocheio and Jerome of
Moravia. I have kept these two celebrities backstage until now for they are
exceptional writers who are reporting the musical customs of an exceptional
city.
Viewed in the context of this book, Paris is exceptional for many reasons.
It is a city, not a court; a hive of ambitious clercs packed into tenements and
hostels, not an assembly of courtiers and knights in a chamber or hall,
‘around the bright fire... amiable and agreeable’.' Above all, Paris is full of
men who can lire et chanter—read from script and sing from musical
notation—the two basic skills of every clerc and the foundation for all of his
ambitions for lucrative office. In later life these skills, acquired in childhood,
perhaps on a bench in a parish school, could help to nurture a taste for the
savant forms of music which reigned in Parisian musical circles, the motet
and the Latin conductus, both monophonic and polyphonic. As Johannes de
Grocheio says, these were forms for Jitterati and divites—for men made
learned by study or made rich by worldly success; trouvére song, in
contrast, was for magnates whom it would be inapt to call litterati and vulgar
to call divites; in Grocheio’s scheme of things they are reges, nobiles and
principes terre (Figure 4).
It is this contrast between courtliness and the world of clerical literacy,
both verbal and musical, which provides the theme for our prolegomenon to
Johannes de Grocheio and Jerome of Moravia. In Chapters 1—4 of this book
I have suggested that in the predominantly notationless world of the
troubadours and trouveres before c1250 there existed a courtly tradition of
instrumental usage. It was courtly because it rested upon a collective moral
and aesthetic apprehension within aristocratic society that it was courtois to
50
Paris
Oh inepurer auanace ma
Tene ame tle Mlemewlone
atdier: A cethedamcoondte- wna
ter nule riche tame: ong h
tehlemene: erOn aferrment
qn aon ae renomiek: aie
me Lier eryprenc- waligilede cs
mix instrumental sound with some kinds of songs but vilain to mix it with
others.
This austere ethic of instrumental usage—for an ethic it deserves to be
called—was vulnerable to abuse by the ignorant who had not the taste to
appreciate it, but it was also under threat from the learned and literate who
had the manoeuvrability of mind to bypass the tradition and come up with
something else. This manoeuvrability was already there in the act of
recording courtly songs with neumes and staves, for once a lyric has been
written down (as happened to many troubadour and trouveére songs after
c1250) it ceases to be an event. It becomes an object and can therefore be
objectively perceived. Any moorings which may have tied it to a kind of
occasion, or a kind of performance, become loosened.
The emergence of a literate tradition of troubadour and trouvére song in
the second half of the thirteenth century may therefore have exerted a
profound effect upon performance practice in that repertory, for musical
notation enabled songs to be lifted out of the behavioural conventions in
which they had become embedded. So it is important to look at some of the
ways in which string playing interacted with literacy, both verbal and
musical, in the Parisian milieu which nourished Jerome of Moravia and
Johannes de Grocheio.
Il
‘Music’, writes Jehan des Murs, who attained the academic title of magister
at Paris in the 1320s,”
his generation, that musical composition and notation had been trans-
formed to the point where a new art, an ars nova, had arisen. Yet his words
cut deeper than that, for the ‘newness’ he senses involves a recognition that
a new kind of urban community has developed, one which forces the literate
man to widen the range of human activity which he judges to be
consequential.
This sense of the city as a place where a wealth of music comes together
animates Johannes de Grocheio’s De Musica, probably written in Paris
around 1300. It is strongest in what might be called Grocheio’s ‘urban
morality’, for in his view music contributes to the stability of the city and
anything which serves the civitas cannot be bad in itself, although it may
have evil applications.’ Sometimes this philosophy leads Grocheio into
opinions that are surprisingly liberal for the thirteenth century and wildly at
variance with contemporary teaching from the pulpit and confessional.
Consider, for example, Grocheio’s view of the public ring-dances, or coreae,
which were so popular amongst the young men and girls whose musical
initiatives are praised by Jehan des Murs. In the eyes of most contemporary
moralists these coreae were services in Satan’s church, but Grocheio, in
defiance of some of the most respected churchmen of his day, takes the
opposite view: ‘these dances keep young men and girls from vanity and are
said to have force against the passion of erotic desire.’* Even the simplest
and most popular music, it seems, had a role to play in maintaining
civilisation on the banks of the Seine.
As we shall see later in this book, Grocheio has much to tell us about
instrumental practice, and it is probably because many clercs shared his
background of urban experience that we begin to find an increased
awareness of musical instruments in many Latin works of the thirteenth
century: sermons, tracts on the Vices and Virtues, Bible commentaries and
manuals of confession, to name only a few. This new awareness of
instruments is apparent in this kind of vocabulary which these clercs employ.
For many centuries monastic writers had been content with a repertoire of
instrument-names derived from the Bible or from classical literature (cithara
and psalterium, for example), and these words were luminous with the
splendour of Christian history and prophecy. Psalterium evoked Mankind’s
long journey towards the Incarnation, foretold by David as he sang in
psalterio, while cithara spanned the history of Man from Tubal in the first
generation after the Flood to the Revelation of St John where the sound of
citharae is heard amidst the marvels that signal the end of all earthly music.
From the thirteenth century on, however, Latin writings begin to fill up
with a new generation of words which could not be found in the Bible, in
Ovid, in Isidore of Seville or in any other standard and revered source.
Indeed it was precisely this lack of literary charisma which recommended
these new words; they were useful because they were current in vernacular
speech and were directly tied to instruments in daily use. Grocheio gives us
viella, and the highly exotic guitarra sarracenica. Jerome of Moravia has viella
a
Songs and instruments
and rubeba, while other theorists have liuto and cistolla. It is hard to imagine
an eleventh- (or even twelfth-) century Latinist comparing the seven words
of Christ on the cross to the ‘seven notes of a viella tuned up to breaking
point’: .vij. note vielle que usque ad rupturam cordarum fuit tensa, but it does not
n isac)
eca
sareeare
i:
een
“lees od
om awa wa me “s
on
Paris
Isidore says in Book III of his Etymologies that the psaltertum has a
soundbox in its upper part, and also that its form was like that of the
Greek letter which is called Delta, so it is clear that the psaltertum of
the Ancients was not the same in form as the one now in use.
Whence it should also be noted that, in the realm of modern
instruments, all kinds of vielle (some of which have more strings than
others) have a soundbox above which is rested against the left
shoulder. While these strings are sounded by means of a haired
stick, the sound is modulated below by touching the strings with the
fingers ofthe left hand so that they sound now higher, now lower.
The psalterium [of the Ancients] is comparable to the viella in two
respects: firstly, that it has its soundbox in the upper part, and
secondly, that its sounds are modulated by the touch offingers
below. Yet these two instruments are also to be distinguished in
three aspects: firstly, in the shape and form [of the psalterium of the
Ancients] which, according to Isidore, was like the Greek letter
Delta which looks like this A, yet I do not remember ever having
seen [a viella] of this form amongst those in modern use; secondly,
55
Songs and instruments
lauder dum.
56
Paris
Ill
Jerome of Moravia’s treatise points us to the most obvious way in which
musical instruments came increasingly to impinge upon the lives of clercs:
they were increasingly played by clercs. Jerome’s chapter on the tuning of the
rubeba and viella is embedded in a compilation which, according to his own
testimony, is for the ‘friars of our order or of another’. We can only conclude
that there existed—at least in Paris—a population of well-educated and
musically literate friars who wished to learn the rudiments of the fiddle from
Jerome’s last chapter just as they wished to learn the rules of Franconian
notation from the Ars cantus mensurabilis which 1s one of the treatises included
in Jerome’s compilation. Some of them may even have done so because of an
interest in Bible commentary such as we found in Petrus de Palude’s
treatise on the instruments of the psalms; at least one thirteenth-century
friar was of the opinion that the ideal commentator would be one who not
only knew the properties of the instruments mentioned in scripture but who
could also play them.’ Others may have been attracted by the savant ethos
which had begun to gather around string-playing in the wake ofAristotelian
studies. According to this, the first genuine alternative to the courtly ethos of
string-playing which medieval culture evolved, the skill of playing upon
instruments was regarded as a rational and teachable technology: a scientia.
The Dominican Albertus Magnus, the leading scientific personality of
thirteenth-century Paris and the Germanies, deals with this point in several
places while commenting upon Aristotle’s Ethics:'°
Here the intricate art of fiddling is well on the way to being unravelled by
rational thought. Albertus, following Avicenna, and sharing with him a
rigorous, Aristotelian terminology, classifies viella-playing as a scientia: a skill
which rests upon a knowledge of principles, and since principles, by
definition, are ‘intelligible things’ (intelligibilia), fiddling can be taught.
Hence Albertus’s careful choice of words: ‘the science of arranging strings
teaches how to adjust...’. Yet a scientia is more than a set of teachable
principles; it is also something to practise and master: a usus. Hence, once
more, Albertus’s careful use of terms: ‘the other science uses ...’. Here, once
again, is writing without which Jerome’s treatise on fiddles seems unimagin-
able; Jerome imparts what Albertus calls the scientia compositionis chordarum,
the skill of measuring (i.e. tuning and stopping) strings into high, middle
and low notes. It was because the viella had been drawn into this special
constellation of ideas that Jerome thought it worthy of treatment, specifying
three tunings, itemising the stopped notes, mentioning an advanced
technique, and even giving a moment’s attention to repertory.
The most likely place for this constellation to have been formed is among
the students, friars and priests of later thirteenth-century Paris. The
attraction of string-playing for young university students seems to be
everlasting, and it is no surprise to find Gilles li Muisis (1272-1352)
reporting that he had seen the students of Paris returning from the schools
58
Paris
and playing chistolles as they went.'' Yet there could be more to this
musicianship than the natural ebullience of the young. Konrad of Megen-
berg, a magister at Paris in the fourteenth century, rules that stringed
instruments ‘are sensibly classed amongst modest activities and philosophi-
cal pastimes when intervals from study are given’,'” which suggests that a
more serious notion of string-playing was sometimes involved. Konrad’s
description ofstring-music as worthy to be classed among the ‘philosophical
pastimes’ (philosophicis solatiis) shows how far instrumental skills had
become immersed in a savant ethos owing much to the recovery of
Aristotle’s philosophy in the thirteenth century.
What fragments these may seem, however, beside the reminiscences of
Henri Bate, a student at Paris around 1266-70 and later a distinguished
theologian. In the Nativitas magistri Henrici Mechliniensis Henri gives this
lavish and detailed account of his musical interests while a young student:'*
instruments are the most appropriate ones for young boys to learn, on the
other he aligns himself with the doctrine (so important in thirteenth-century
theology) that it is legitimate to relieve the tedium of earthly activities and
duties with some kind of play—in this case, play of instruments, ‘not
inimical to the student life, especially amongst the young’. We also notice
how, as in Johannes de Grocheio’s De Musica, ideas which originated
independently of Christian beliefas strictly formulated by theologians could
succeed in stifling any cleric’s doubts about the moral legitimacy of secular
music and its pleasures; for Grocheio, it is a view ofthe city as an organism,
derived from Greek philosophy; for Henri de Malines, it is the imagery of
courtliness as projected in a thousand romances. With these thoughts
in mind, neither shows any reluctance to speak highly of dancing, for
example, despite the passionate opposition of countless sermonisers and
preachers, among whom were some of the most distinguished literary men
of the age.'*
IV
Henri de Malines allows us to draw the Parisian evidence for instrumental
practice in monophonic song together. As a literate fiddle player active in
Paris during the later thirteenth century, he is probably just the kind of
musician who might have encountered Jerome of Moravia’s treatise on the
fiddle, and whose views about performance practice may be embodied in
Johannes de Grocheio’s De Musica. We are unlikely to find a better
candidate for the kind of man whom Grocheio means when He speaks
admiringly of what a ‘good fiddler’—a bonus artifex in viella—can do.
One thing is certain: Henri, and many clerc fiddlers like him, would have
been musically literate to the point of being able to read plainchant
notation. The Paris schools stood on the highway to success in ecclesiastical
or secular office but they were staging posts along a road that began in the
parish schools of the towns and villages where young clergons learned their
alphabet, the psalms and some antiphons, together with staff-notation when
their teachers had some proficiency in cantu. Gautier de Coinci portrays such
an education in his Miracles de Nostre Dame; a poor woman’s son is put to
school at a tender age, ‘to the honour of God and our lady’, and ‘soon he
knows how to sing and soon how to read’ (Tost seit chanter et tost seit lire).'°
The ability to draw words from script and to elicit song from musical
notation was regarded as the characteristic talent of the cleric and the pair
of verbs lire et chanter is found throughout medieval French literature in this
context.
The task of learning to sing from plainchant notation would have
involved clergons like the young hero of Gautier’s story in study of solmisation
and the gamut, for solmisation was the scaffolding for most musical literacy
in the later Middle Ages and was ‘very familiar even to boys only just setting
61
Songs and instruments
Jerome of Moravia’s chapter on the tuning and fingering ofthe rubeba and
viella leaves us in no doubt that solmisation had found a secure place in the
primary stages of learning the fiddle. He describes the intervals set between
the strings of these instruments and then inventories the notes produced by
stopping each string within the resources of the first position. In every case
he employs the alphabetic letters of the musica recta gamut to do so, and with
the rubeba (the first instrument to be described) he uses the hexachord
denomination as well: the first string makes C fa ut, with the application of
the first finger D sol re, and so on.!'®
The scales of the rubeba and viella as Jerome presents them are a perfect
marriage of manual technique and literate theory. He stresses that each
finger of the left hand must be put down ‘in a natural fashion’ (naturaliter)
and his implied teaching is therefore that the physical structure of the hand
welcomes the musica recta gamut, for that is what his stoppings produce. He
does not discuss /ficte musice, but it is obvious that they will be produced by
deploying the fingers in a fashion which is not ‘natural’: by ‘twisting’ them,
‘bending’ them, or drawing them up towards the pegbox:”°
62
Paris
ut re ml fa sol la
Example 11
Se
the c-clef establishes the second line as fa and the f-clef does the same for the
fourth line. To play from this notation a fiddler needed to know where to
find a fa on his instrument; this is what Jerome tells him. In the first viedla
tuning, for example, there are three stopping points that may be solmised as
fa and the player can build upon that knowledge to assign the correct
pattern of tone and semitone steps to the notation:
63
Songs and instruments
(The lateral bourdon string has been omitted since it is not stopped by the
fingers. )
ut sol re ut la sol re
r G . dd
first finger la mi re la mi
second finger fa ut
second finger
fourth finger la mi re
Literate string-players equipped in this way would have been able to play
from any of the surviving trouvere chansonniers.
Our best understanding of the repertory of Parisian fiddlers comes from
the music treatises of Johannes de Grocheio and Jerome of Moravia, so often
mentioned in what has gone before. In his account of tunings used on the
viella, or five-stringed fiddle (Figure 7), Jerome makes several direct and
indirect references to repertory in his descriptions of tunings | and 2:°!
Tuning|
dGgd'd'
Et talis viella, ut prius patuit, vim modorum omnium
comprehendit.
And such a viella as just described encompasses the
material of all the modes.
Tuning 2
dGgd'g'
. . necessarius est propter laycos et omnes alios cantus,
maxime irregulares, qui frequenter per totam manum
discurrere volunt.
64
Paris
Figure 7 The viella. New York, Pierpont Morgan Library, MS 638, f-17r.
French (Paris?), thirteenth century. Reproduced by permission.
65
Songs and instruments
Example 12
yeaa CLEAN —- z = Fe ~
zi os i = =
cho - rus no - vae JC tom oe ie meee - lem
Fe = ——— 7 ae
é = e z | 03-5
8
no - vam me - li duly =) «ce We=1 di - nem
= 7 EN —~\
O— Os
G Sse =? —— a=: 7 == =
8
pro - mat co. — Tens cum so - bri - is
6 z a = SS ————— 4
pas - cha - le fes - tum gau - di - is.
With Jerome’s ‘secular and all other kinds of songs... which frequently
66
Paris
wish to run through the whole hand’ we are on less solid ground, for no
secular music has survived from Jerome’s day which exploits the full gamut
of two octaves and a sixth (an exceptional compass for any medieval piece,
whether monophonic or polyphonic, until the second half of the fifteenth
century). Troubadour and trouvére songs, for example, rarely exceed a
twelfth, and the same may be said of Ars Antiqua motets (most of which
appear to have been written for three voices of approximately the same kind;
a singer who wished to sing tenor parts one moment and triplum parts
the next would find himself required to cover no more compass than he
might expect to exploit in performing a monophonic song). Jerome’s secular
music which wishes to run through the whole gamut of two octaves and a
sixth therefore seems impossible to identify in terms of contemporary
compositions and the answer presumably lies with performance practice;
perhaps the fiddle-postludes, or modi, described by Johannes de Grocheio
were virtuosic improvisations that exploited the whole compass of the
viella.*®
We shall return to Jerome of Moravia; for the moment his source has run
dry and it is Johannes de Grocheio who now offers the fullest information
about fiddle repertory in late thirteenth-century Paris. Grocheio’s De Musica
refers several times to the central genre of the trouvere art, the High Style
chanson, using the terms cantus coronatus (for the most sublime examples of the
genre) and cantus versualis (for those specimens which fall short of the very
highest standards ofliterary and musical excellence).*’ These passages leave
us in no doubt that the cantus coronatus was customarily performed by the
best fiddlers, and Grocheio even describes a performing technique which
these viellatores used when playing trouvére songs:“® ‘
cantilena corresponds to our distinction between the High Style and the
Lower Styles. Cantus is a generic term in Grocheio’s usage, and it embraces:
The cantilena register embraces Lower Style songs, all of which have a
. - OC
reprise or refrain of some kind:”°
This is a long list of forms and we may begin to wonder whether the best
fiddlers of thirteenth-century Paris set any limit to their repertory. At first
sight Grocheio’s words suggest that they did not: ‘a good fiddler generally
plays every kind of cantus and cantilena, and every musical form.’ Yet it would
be hasty to proceed from here to the view that the expert viellator played
every kind of music. It is striking, for example, that.Grocheio discusses
musical instruments in the section of his work devoted to ‘monophonic,
popular or un-learned music’ (musica simplex vel civilis vel vulgaris); this might
be taken to imply, as Gushee has pointed out, that instruments did not
perform in polyphonic forms such as the motet and polyphonic conductus—
forms which Grocheio treats under the separate heading of ‘composed,
regulated, rule-bound or measured music’ (musica composita vel regularis vel
canonica vel mensurata).*° Since there is a third category of music in Grocheio’s
treatise (‘Church music’, or musica ecclesiastica) in which instruments were
probably not used (at least as a general rule), then Grocheio’s statement
that good fiddlers played ‘every musical form’ may actually embrace less
than a third of the musical forms discussed in his treatise.
Be that as it may, the key to forming a proper understanding of
Grocheio’s evidence surely lies with the question of literacy versus illiteracy,
and the kinds of performing mentality bred by these two conditions. At the
beginning of this chapter we suggested that when trouveére and troubadour
songs began to be written down, as they were in great numbers from the
mid-thirteenth century on, then they became objects which could be
objectively perceived in isolation from any performing conventions that may
have clung to them when transmission was largely or exclusively oral. As
68
Paris
V
If this is an accurate picture of performing traditions in Paris, how did these
literate viellatores accompany High Style trouvére songs? The question is out
of place here since it belongs with later chapters on performance practice,
yet we have spent so long in the Parisian milieu of Jerome of Moravia and
Grocheio that it becomes irresistible. Surely Paris, with its vigorous
polyphonic traditions, supported by an ancillary literature of music-
treatises, is exactly the place where we might expect to pick up an echo of
the way a bonus artifex in viella performed a trouveére song.
I believe that there is such an echo in Jerome of Moravia’s chapter on the
rubeba and viella, although it is one that has often been distorted. At the close
of the chapter Jerome describes what seems to be an advanced playing
technique in these terms:*!
The exact meaning of this passage will probably never be clear, for Jerome
is saying too much in too few words. What is meant by replying in the first
harmonies to any note from which any melody is woven?
At some time before 1306 the owner of the only complete surviving copy
of Jerome’s treatise, Pierre de Limoges, decided that these lines about an
advanced viella technique were too cryptic to be of use. Like many medieval
bibliophiles, Pierre was happy to scribble over his books and he felt that
Jerome’s explanation of the advanced fiddle-technique required an exten-
sive annotation. This is what he made ofit:*
Example 13
(relative pitch)
-@-@
= =
Se SS eee
oe
Viella
y EES be be *.
Bourdon
string
melodies in which the drone (if that is now the right word to use) was
sounded here and there as the tune threaded its way in and out of the
required relationships with the bourdon. Yet, as example 13 shows, there
are several notes in the fiddle’s compass which cannot be sounded with the
bourdon ‘in the first harmonies’ (unless we expand what is meant by ‘first
harmonies’ to the point where it embraces thirds and sixths; in that case the
rules of the technique would permit an almost constant drone—presumably
the device which the ‘advanced’ way of playing is intended to supplant).
The difficulty here is that some of the stopped notes on the fingerboard
cannot be sounded in the first harmonies with the bourdon, yet Jerome
requires the skilled player to reply to any note from which any melody is
woven with the bourdon ‘in the first harmonies’.
One way out of this impasse is to assume that Jerome is describing a
playing-style analogous to the contemporary vocal practice of ‘fifthing’.
Sarah Fuller has shown that four music treatises from the thirteenth and
fourteenth centuries contain the same core of basic rules for creating a
second voice, largely in parallel fifths over a given melody, and these rules
describe a note-against-note technique in which the singer who is fifthing
must know the octave or fifth of any note in the melody he is accompanying
so that he may sing in parallel fifths above it (and supply occasional passing
notes).°* With this basic knowledge he should then proceed to follow a few
simple directives to negotiate his way into fifths at the start of phrases:**
Example 14
in the first harmonies. This means, I suggest, that the fiddler should know
how to ‘fifth’ a given melody, played or sung by another musician, upon his
instrument.’ Jerome’s advanced fiddle-technique would therefore have
evolved as a form of improvised note-against-note counterpoint over
another melody. This interpretation helps to explain Jerome’s insistence
that this is a ‘serious’ and ‘advanced’ way of playing (these are not epithets
one might readily apply to the selective use of the bourdon envisaged by
Pierre de Limoges); it is difficult and specialised because it involves literacy-
based musical devices (albeit simple ones) taken from the realm of the
discantor.
Ata time when a great deal of instrumental playing must have been based
upon drones (as Jerome’s fiddle-tunings suggest), this ‘advanced’ technique
required the viellator to study the learning of the discantor. He needed to know
contrapuntal formulae like the ones listed above and to consult the
Guidonian hand in order to locate the octave and fifth of every note in the
melody. This technique must also have involved the use of musical notation,
for Jerome records that it is more easily accomplished by those who know
the ‘second hand’ (presumably the Guidonian hand transposed up a fifth)
or the normal hand, which implies that players consulted notated melodies
and then built up their fifthing-style accompaniment using the hand as a
mnemonic to tell them where the semitones should lie.
If this interpretation is correct then we can reconstruct a literate style of
accompaniment which may have been used in Parisian circles (and surely
elsewhere?) around 1300 to accompany High Style songs of the trouvére
tradition:*°
Example 15
viella
— 1 =
SG "eo, © oe ee oe = oa ao = ae,
oe
33 Se oe Pith Te
== SSS
= =
FY a ESS ar Baa2 Ls 7
=9: a £ = = fetes
72
Paris
There are two striking things about this style of accompaniment. The first
is that it looks like a tidied version of an illiterate, heterophonic technique of
doubling a melody in parallel fifths. No doubt this is exactly what it is, just
as the vocal technique of fifthing was a slightly savant version of the kind of
parallel singing in fifths, octaves and twelfths described in the 1270s by Elias
of Salomon.*’ In other words there must have been illiterate fiddlers who
could produce something quite like Jerome’s advanced technique, and this
is indeed what several sources suggest. ‘Many succeed in the art of fiddling’,
says Albertus Magnus, ‘who cannot explain the rational basis of harmoni-
ous accord,”*® and Engelbert of Admont makes the same observation,
hemming it with a citation from Aristotle: ‘there are many who, without
scientific mastery, accomplish the things which such mastery accom-
plishes.’*? However, only the advanced and literate fiddlers, playing ex arte
as well as ex usu, would have been able to produce contrary motion such as
we see in example 15 on a rational basis, enjoying the satisfation of being
able to explain their procedures with technical terms and concepts.
The second striking feature about the style of accompaniment shown in
example 15 is that Jerome describes it as an advanced form of playing; yet,
in vocal terms, it is quite rudimentary. Amongst the discantores who had
proceeded to the performance of motets and polyphonic conducti the
technique of fifthing was regarded as material for the novice. This
discrepancy between vocal and instrumental practice suggests that the
artistic horizons of voices and instruments were differently placed in the
thirteenth century, even amongst literate players.
It is in this last respect, perhaps, that Jerome’s advanced technique is of
most historical interest. The purpose of the technique is clearly to create a
style of fiddling regulated by literacy-based concepts about consonance,
dissonance and contrary motion. This can only have happened because
written vocal polyphony was fast becoming the respected form of musical
endeavour and fiddling was trying to catch up and imitate some of the
savant techniques of conductus and motet. Indeed the aim of the technique
seems to be to turn a monophonic song into a kind of rudimentary two-part
conductus. With Jerome, in other words, we see an art-instrument coming
under the sway ofvocal polyphony for the first time in the West; this was the
force that was eventually to produce the art of string-instrument intabu-
lation in the late fourteenth and fifteenth centuries and to shape the whole
course of instrumental music in the Renaissance.
Example 15 may therefore represent a late thirteenth-century—and
particularly Parisian—style of performance for three reasons. Firstly, in the
fact that accompaniment is used at all in a High Style song of the trouvére
tradition. Secondly, in the use of a mildly savant and literacy-based
technique of counterpoint as a basis for accompaniment. Thirdly, in the
question of rhythm. Example 15 uses a non-mensural notation in ac-
cordance with the way in which the song has been preserved. Yet it is well
known that there are some trouvere sources in which High Style songs are
ie
Songs and instruments
Example 16
es
viella (reconstruction) 3
eee ee
eS SS
polyphony, not by using a High Style in words and music as had been the
custom in the monophonic tradition. (Grocheio had some idea of what a
High Style meant in terms of trouvére song, but it is revealing that he
overdoes things in his attempts to capture its ethos; when he places the High
Style song in a half-romantic world of princes, kings and noblemen he is
using his imagination as much as his judgement to describe a repertory all
but dead in his lifetime.) Grocheio’s ears must have been full of the sounds
of polyphonic forms such as the motet, perhaps the most admired form of
music-making in thirteenth-century Paris. The friction of the motet, with its
multiple texts and strong dissonances, was enough to scour away the old
monophonic decorum, for when viewed in terms of that decorum the motet
takes on a highly subversive appearance: it joins a new High Style in music
to the old Lower Styles of poetry.
There is a good deal of poetry in the motet corpus which would be quite at
home in a High Style trouvére song. This, for example:*”
Now I clearly see that I must confess to her who has kept
me injoy for so long, like a true lover who always
intends to do good, and is constant and discreet, ought
to feel, unless I want to lack all honour and, like a
miserable wretch, love slander....
There is too much chiming of rhymes here, perhaps, for the true High
Style chanson of the trouvéres, but the diction and sentiment would do well
enough for that genre. To a very large extent, however, motet poetry in the
vernacular draws upon the idioms of the pastourelle and the simple refrains
often associated with rondeaux. It presents us with a mass of Lower Style
characteristics: diminutives in -ete (amourete, flourete, joliete, pucelete, praélete,
and many more); short lines with chiming rhyme-sounds (‘Je sui joliete,/
doucete/et plaisans/jone pucelete...’); the exclamations which express
strong passion but leave no room for feeling to be explored (“Dieus! hé! plus
Wirai2’); the direct and desperate address to the beloved (‘lady, with clasped
hands I beg your mercy’); allusions to stereotyped narrative situations,
especially those involving marriage (‘I would sing out of sheer pleasure for
the one that I have loved, but she has a new husband who has stopped me
75
Songs and instruments
from seeing her’), and the easy pastiche of the pastourelle manner:*!
76
20.
‘The carole
They serve the Devil with their voices as
they sing, with their hands as they lead one
another, and with their feet as they move in
a circle. Whence their dance is called the
Devil’s mill.
I
The last five chapters have followed a chronological path from the first Old
Proveng¢al sources of the twelfth century to the music treatises of Grocheio
and Jerome of Moravia, compiled in Paris towards the end of the thirteenth
century. The aim of the three chapters which follow is to investigate some
special repertories which do not press to be included in a chronological
pattern but do demand separate treatment so that their distinctiveness may
emerge.
These repertories are special in various ways. The first comprises the
public dance-songs, or caroles (Latin coreae), which were sung in popular
festivities as well as at court. Although the musical remains of these dances
are relatively scarce, a great deal of evidence survives pertaining to the way
in which they were presented and performed, and in the eyes of contempor-
ary social commentators the carole simply was secular music in the North, not
the highbrow forms of trouvere song or the savant forms of conductus and
motet. The carole therefore presents a particularly striking case of the iceberg
problem which meets us everywhere in medieval music: the forms which
dominate our view of secular music-making in twelfth- and thirteenth-
century France are those which survive, but the view of contemporaries was
dominated by material which has left relatively little trace.
The second repertory comprises monophonic conducti. The texts of these
songs, which are in Latin rather than the vernacular cultivated by the
trouveres, range over many subjects including religious devotion and social
satire and they cannot be placed within the scheme of High Style and Lower
Styles which has guided us in Chapters 1—5. Furthermore, the monophonic
conductus is a largely northern and—in some measure— Parisian repertory,
77
Songs and instruments
I]
When the moralists of the thirteenth century turned their gaze upon secular
music their vision was filled with the spectacle of public dances, or caroles
78
The carole
(Figure 8). These dances seem to have flourished in every region oflife, from
the village churchyards to the aristocratic halls, and churchmen inveighed
against them throughout the thirteenth century. So abundant are the
references to these caroles and coreae, and so meagre are the moralists’
references to élite music such as courtly monody or the motet,” that coreae
deserve to be placed at the centre of our picture of secular music in
thirteenth-century France.
Beyond the walls of aristocratic residences and castles these dances,
celebrated during spring and summer,’ were public festivals in a world
where almost all the daylight hours were given over to manual labour. To
judge by the quantity of harangue directed against them from the pulpit and
confessional they loomed large in the lives of country people and city-
dwellers, just as they did at court during the great festivals of Christmas and
Pentecost.* In both the popular and courtly manifestations of the corea
(which should not, perhaps, be too sharply distinguished), the dancers
prepared themselves in advance with fresh clothes and floral garlands for
their foreheads; once the corea was under way and the performers, linking
hands, began to move in a circle to the left,? the dance became a public
event where there could never be too many participants or spectators (‘the
more who participate’, comments an anonymous preacher, ‘and the more
who watch, the merrier it is’).° In the eyes of many contemporary moralists
the result of all this dancing and crowding was to produce a grotesque
parody of the Church’s liturgy in which the singers who led the dance were
anti-priests:’
In a corea the Devil has a priest who sings and a cleric who
serves, and he [the Devil] causes almost all the hours to be
sung through the streets and roadways. And just.as a priest
changes his clothes when he must celebrate, so they who
must lead coreae do... and in place of the Office of God
they celebrate the Office of the Devil, to which many
79
Songs and instruments
more people come than the Office of God and spend more
time because the Devil’s Office is longer (and at one
time or another can last the whole night or more). They
serve the Devil much more devotedly and keenly than
God....In this they sin against the sacrament ofordination,
because they busy themselves in just such a liturgy to the
Devil as clerics perform to God, as has been said. And
through the music [of the women who dance coreae] the
music of the Church is brought into disdain, for when they
should be at Vespers they are present in coreae.
Mainamain...
devant le tref, en un pré vert,
les puceles et li vallet
ront la carole commenciee.
Une dame s’est avanciee.
vestue d’une cote en graine,
si chante ceste premeraine:
Here Jean Renart has captured the ethos of the thirteenth-century carole. His
dance is the very epitome of the primaveral quality of jeunesse, for it is
performed in a green pasture by maidens and young men; it expresses the
elegant and carefree movement of people who, in Jean Renart’s imagin-
ation, are held in an idealised state of candid and insouciant youthfulness.
This emphasis upon youth traversed all classes of society where caroles
were performed. As manifested in these dances the courtly cult of jeunesse
drew nourishment from festivals in the towns and villages where the public
dances were mostly performed by mulieres et iuvenes—by women and young
80
The carole
men.” The primaveral ethos of the caroles also attracted the young students
in the schools. ‘{I1 was] a merry and amorous leader of coreae and master of
dances in wooded places’, says the theologian Henri Bate; ‘such things are
not inimical to the student life, especially amongst the young.’!?
To judge by a wealth of literary testimony, coreae were usually danced to
songs which the dancers provided for themselves. The writings of thir-
teenth-century preachers and theologians are laden with bitter invectives
against these cantus and cantilene which drew young husbands away from
their wives and seduced serving girls away from their households.!!
Ill
Some degree ofinstrumental involvement in coreae seems to be suggested by
pictorial sources (Figure 8) and by several passing references in the writings
of thirteenth-century theologians. Just as the Israelites danced to instru-
ments when they had crossed the Red Sea, argues Honorius Augustodunen-
sis, so the faithful ‘still use musical instruments in their ring-dances’
(Appendix 3:48). In a comparable passage Guillaume d’Auvergne mentions
how dancers perform coreae ‘making a clamour and springing...to the
sounds...of musical instruments’ (ibid., 45).
Here we seem to be dealing with luxurious evidence: pictures, references
in romances, harangues from the pens of moralists and passing references in
theological treatises and encyclopedias. Yet none of the specific sources
cited so far has clarified the question of whether the songs performed in
coreae were accompanied by instruments, or whether there were certain
kinds of corea which were performed to the music of instruments alone.
Depictions of these dances, for example, are not usually endowed with
sufficient detail to reveal whether any of the participants are singing as they
move. Similarly, it is uncertain whether the coreae mentioned by Honorius
Augustudonensis (‘which the faithful still [perform with] musical instru-
ments’) are accompanied dance-songs or just dance-pieces played upon
instruments.
At this point Johannes de Grocheio’s evidence is of great interest. He is
our principal witness to the musical character and performance practice of
the thirteenth-century corea; the striking feature of his account is that he
draws a clear distinction between purely vocal and instrumental coreae.
Grocheio’s term for the pieces performed during these dances is ductia, no
doubt because of the Latin idiom coreas ducere, ‘to lead [i.e. to dance] coreae’.'*
He distinguishes a vocal and an instrumental form. From his descriptions
we learn that the vocal ductia began and ended with a refrain and that
between the refrains there lay poetic material with some lines sharing the
forms and rhymes of the refrain.'’ These details are confusing at first, and
there are more: from Grocheio’s description of the rondeau (rotunda vel
rotundellus) we may deduce that the added material in the vocal ductia had
81
Songs and instruments
different melodic material from the refrain. Bewildering though these details
may be, they gradually resolve into a description of a virelai like this one:!*
In the light of Grocheio’s description of the vocal ductia the crucial lines here
are 3—4: S’amor venist a plesir/que me vousissent sesir; in accordance with virelai
form, this couplet introduces new melodic material and new rhymes not
found in the refrain. This seems to be the kind of piece which Grocheio had
heard ‘sung in coreae by young men and girls’.
Grocheio’s instrumental ductia, however, displays a different musical
form. Its function is exactly the same as that of its vocal counterpart (‘the
ductia excites the soul of man to move ornately .. . in coreis’),'” yet while the
vocal type appears to have been cast in the mould of a virelai, the
instrumental ductia consists of paired melodic sections with open and closed
endings, so following this form (compare Music example 9):
open
closed
open
closed
open
Oe
Sains closed etc.
Thus Grocheio’s account of corea music reveals a fissure between vocal and
instrumental usage. According to his report, vocal and instrumental corea
melodies were identical in name and function but were quite distinct in their
form (and therefore, presumably, in the choreographies which were used
with them).
If we widen our field of view to take in the references of moralists and
preachers we find further sign of a divide between coreae performed to songs
on the one hand and performed to instrumental music on the other. For
example, Albertus Magnus legislates that coreae are not evil in themselves,
82
The carole
La escoutoient bonement :
.I. conteor, qui lor contoit
Une chancon et si notoit
Ses refrez en une viele,
Qui assez iert et bonne et bele.
This chanson with its refrains may be a rondeau or some similar form. This is
suggested by certain features in the poet’s description of the scene, for the
minstrel performs En mi ... d’une praierie which is near la rive de mer; both of
these phrases are key registral terms in the thirteenth-century repertory of
rondeaux and simple refrain songs. Compare the following incipits of lyrics
quoted in Jean Renart’s Guillaume de Dole:”*
This is not to claim that the author has scattered clues to the nature of the
minstrel’s song here and there in his account of its performance, but only
that his imaginative apprehension of this musical scene is impregnated with
poetic formulae which suggest that the chanson is a monophonic refrain-song,
perhaps a rondeau or virelai. These few lines may be our only guide to the
way in which such lyrics were performed as solo songs with instruments
during the thirteenth century.
84
ue
The monophonic
conductus
‘This boy knew how to play a fiddle ... he was familiar with (and
willingly sang) all kinds of monophonic songs in diverse languages. ...’
I
According to his own account, Henri de Malines, theologian and graduate
of the University of Paris, was ‘most willing to sing all kinds of monophonic
songs (cantiones vulgarium) in diverse languages’ when he was a young man.!
Since Malines lies midway between Brussels and Antwerp it is likely that
some of these songs in diversis linguis were in Flemish; others were probably
in French, the universal language of courtliness, and some perhaps in Latin,
a language which Henri wielded to good use throughout his distinguished
career as a theologian. In the minds of such clercs as Henri—men who
mingled Latin and vernacular in their sermons, produced French trans-
lations of Latin treatises, and naturalised a host of Latin words into the
vernacular—the distinction between Latin and vernacular lyric cannot
always have been a firm one. Johannes de Grocheio, for example, basing his
views upon Parisian custom, classifies all non-liturgical monophonic songs
together as musica vulgaris (compare Henri’s cantiones vulgarium) and does not
even raise the issue of language; what matters in his eyes is whether a piece
is savant (i.e. polyphonic), or whether it is intended for use in church.
Indeed Grocheio provides clear evidence that Parisian musicians of c1300
assimilated the High Style trouvere song in the vernacular to Latin song;
some Parisians apparently called the trouvere productions simplices conducti,
‘monophonic conducti’.” In adopting this terminology the Parisians were
assimilating a genre whose history has little connection with their city—the
trouvere chanson—to one whose story gathers around the banks of the
Seine—the monophonic conductus. It was an obvious assimilation to make,
for both forms were characterised by their monophonic and predominantly
85
Songs and instruments
syllabic or mildly melismatic melodies, their usually strophic form and their
use of rhyme. The resemblances between the two forms would have been
conspicuous to clercs who enjoyed both kinds of music—men such as the
‘masters and students’ perhaps, who, according to Grocheio, admired
trouvere songs in the High Style ae amongst whom there must have been
many connoisseurs of Latin song.”
Since we have Grocheio’s authority that High Style trouveére songs could
be associated with the fiddle in late thirteenth-century Paris, it seems likely
that Latin monophonic songs may have been performed in the same way
since both could so easily be bracketed together as simplices conducti. This is
only an inference, of course, but a few fragments of evidence suggest that
monophonic conducti may have been associated with instruments as early
as the first decades of the thirteenth century. As early as Le Roman de la
Violette (probably 1227—9) the term conduis is used to denote music played by
jongleurs upon their fiddles,* and although this is a flimsy basis for a
generalisation about the performance of the monophonic conductus reper-
tory it may at least be said that the word conduis undoubtedly passed into
Old French from contemporary Latin usage and may therefore have
retained some specialised meaning.
A more intriguing passage is contained in Gautier de Coinci’s Miracles de
Nostre Dame where Gautier gives his version of the famous miracle whereby
the Virgin caused a candle to descend upon the fiddle of a minstrel who
went from church to church singing her praises (Appendix 3:12 and 37).
Having told this story Gautier takes the opportunity to admonish the
ecclesiastical singers of his day:?
This passage implies that there was some kind of music, involving voices
and instruments, which was ostensibly devotional but failed in that object if
the musicians performing it were defective in faith or conscience. This music
is unlikely to encompass the secular pastoreles, sonnés and chanconnetes which
Gautier so often condemns as trivial and unworthy of educated men.
However, it may well have included the
86
The monophonic conductus
which Gautier deemed proper musical fare for gatherings of learned men.°
The Notre Dame conductus repertory, both monophonic and polyphonic,
incorporates a large hoard of devotional conduis, including 39 conduis de Nostre
Dame (to borrow Gautier’s phrase), 16 pieces in praise of various saints, 43
for Easter and 62 for advent and Christmas.’ The monophonic items may
represent the music which Gautier had heard sung and accompanied by
musicians who (in his judgement) sometimes forgot their devotional
purpose and revelled in their artistry. Gautier would probably have
considered instrumental accompaniment appropriate for conduis de Nostre
Dame (or for any devotional conductus), to judge by the zest with which he
tells the story of the minstrel at Roc-Amadour who sang to the Virgin Mary
and accompanied himself on his fiddle (Appendix 3:37).
To return to the artistic parity between the trouvere chanson and the
monophonic conductus, it seems likely that (at least in Parisian circles)
Latin conducti would sometimes have been performed in much the same
way that their vernacular counterparts were treated by a bonus artifex in
viella. An instrumental accompaniment based upon the practice of fifthing,
such as we reconstructed for trouvere monody in Chapter 5, certainly looks
at home when placed above many conducti:®
Example 17
viella reconstruction
s asec
anaeSi =
$ pttop ty
Mors vi-te pro-pi - ti-a Sex - ta pas -sus fe - ri-a;
— r=p-f—tr
Mor-tis a mi- se - ri-a
itp
Nos e - rex 7 it.
Ses: fees
ee ——s ee
Di - e Chris-tus ter - ti-a Re - sur-rex - it.
87
Songs and instruments
I]
So far we have dealt with the conductus in the general terms which are all
that the fragmentary state of the evidence allows. However, our last musical
example carries us to a collection of pieces within_the monophonic corpus
where the likelihood of instrumental involvement in some performances
seems strong: the Latin rondeaux in the eleventh fascicule of the Florence
manuscript.
There are sixty of these pieces, most of them in some kind of rondeau form
and provided with bold and ingratiating musical settings.’ For the most part
their poetic forms and musical style are closely related to those of vernacular
dance-songs such as we meet in Jean Renart’s Guillaume de Dole. Compare
the following song, for example, with C'est la jus en la praele, which Jean
Renart describes as being performed to a fiddle (see above, pp. 34—5):!°
Nicholaus inclitus,
Laudet omnis spiritus,
Factus est divinitus
Presul cum letitia;
Laudet omnis spiritus
Gubernantem omnia.
Example18
SSS
SSS SS
SSS SS SSS
SSS
SS Lau - det om - nis spl a=
5 Gu -
PaGaL Saez =
ber - nan - tem
88
The monophonic conductus
A further link between the world of the secular corea and these Latin
rondeaux is established by the frequent references in the Latin poems to
‘floral joy’, ‘floral festivities’ or ‘new flowers’:'”
89
Songs and instruments
and which they have got from their lovers, are like
garlands of manifold triumph which the Devil has
won through them ... just as bold knights are
accustomed to place crowns of flowers on the
heads of their horses in tournaments.
Such floral garlands placed upon the forehead were regarded by many
moralists as a sin against the sacrament of confirmation, the sign of the cross
being replaced by the sign of the Devil. One anonymous casuist even breaks
into the vernacular to express his hatred of them:’*
...faciunt contra sacramentum confirmationis quia
in fronte signum crucis suscepunt tanquam empte
Christi sanguine, in choreis vero, signo crucis abiecto,
signum diaboli pro eo posuerunt, scilicet signum
venalitatis in capite quod gerlond dicitur.
Clerics, students and other Jitterati able to read and write Latin also took
part in these coreae. In his reminiscences Henri de Malines candidly admits
that in his younger days he was an enthusiastic corearum ductor, or leader of
coreae, while a sermon preached at St Victor in 1230 by Pierre de Bar-sur-
Aube records how newly-elected Masters of the University of Paris arranged
great festivities in which their friends would ‘lead coreae through the streets
and highways’.'”
But it was not only in Paris that clerics and men of (some) Latin learning
showed their appetite for coreae. In the middle years of the thirteenth century
Odon Rigaud, archbishop of Rouen, visited many ecclesiastical foundations
in Normandy and the records he kept reveal how popular were coreae with
men who, in his judgement, should have known better. At St Yldevert
(Seine-Inférieure) he found ‘clerics, vicars and even chaplains conducting
themselves in a dissolute and scurrilous way on certain feast-days, especially
that of St Nicholas, leading coreae through the streets and making le vireli’,
while at the priory of Villarceaux (Seine-et-Oise) he found members of the
community dressing up ‘in secular clothes... and leading coreae with secular
persons’.'°
It is possible that some of the lighter Latin rondeaux in the Florence
manuscript were intended for performance in contexts such as these. Some
of them may be pious contrafacta of secular dance-songs intended to provide
90
The monophonic conductus
literate men whose appetite for coreae could not be suppressed with material
which would not pollute their throats, but many of them are more spirited
than spiritual and surely reflect the ebullience of ayoung student population
(‘let the labour of study cease!’) for whom the distinction between Latin and
vernacular lyric was not always conspicuous.
The performing-traditions of the vernacular coreae are therefore our best
guide to the performance of these Latin rondeaux. I tentatively suggest that
these songs were sung unaccompanied when performed as populous dance-
songs (the picture which opens the rondeau fascicule in the Florence
manuscript does not show instrumentalists, for what that may be worth).
Yet they are likely to have been often accompanied when they were
performed as solo songs—customarily, perhaps, by a single instrument.
v1
oO
The Roman de Horn
and the lai
What leaf-fring’d legend haunts about thy shape?
l
In Chapter | we encountered what is probably the most striking reference to
accompanied performance anywhere in medieval literature. It is a passage
in the Anglo-Norman Roman de Horn (c1170) where the hero sings something
called a lai to the harp. It is worth looking at again:
92
The Roman de Horn and the lai
These lines are tokens of an ardent imagination. Horn checks the tuning
of the harp by plucking the strings melody-wise (chanter) and chord-wise
(organer).? When he has finished these testing flourishes (Quant ses notes ot
fait) he begins to retune it, making the strings give out completely different
notes (...si la prent a munter/E tut par autres tuns . les cordes fait soner), and then
sings the /ai composed by Baltof, son of the king of Brittany, ‘just as the
Bretons are versed in such performances’. After this he repeats the music of
what he has just sung upon the harp (Apres en lestrument . fet les cordes suner/
Tut issi cum en voiz . avert dit tut premier...).
However we may choose to interpret it, this remarkable description of
something called a /ai being performed by a courtly amateur is one of the
most detailed references to vocal-instrumental delivery that the Middle
Ages have left us; it is an important witness to performing techniques such
as the elaborate tuning-preliminary (which seems to be half preparation and
haif performance in this case) and the alternatim use of voice and
instrument. Yet it is the nature of the music being performed which has
brought these lines from the Roman de Horn to the pages of countless
musicological books and articles,” for by calling Horn’s piece a Jai the author
seems to steer us directly towards one of the most impressive and rewarding
repertories of medieval song. For the troubadours and trouveres a /ai was a
specific lyric form, of most ambitious design, in which each subdivision of
the text had its own metrical form and musical setting. Two verses of the
anonymous French Lai d’Aélis of the thirteenth century will show the
expansiveness of /ai design:*
Example 19
ee ee
I
ee
¢ En sos - pi - rant trop de par - font A - ten - drai le
a
t Ss @ eo ZS
—
Zs
= ri =x@: = ra LA os
aa ING
o# oe oS aaa aS
ZA fe» — se =, zo * 6
& K'en mon cuer font lor fon -de - ment, Et li pen sers
aS Fea = a = _
N a Qi YM er
eA, a —, oe 7
——— SS SSeS
&
me font gas-ter A - mors mon tens. Nuit et jor sos-pir et
= —
a = as —o os = =p —~ se @ rd 72
oe = e
&
plor quant me por - pens; sos- pi - rer ce - le me fait a
zs ———- — = = a = oo = ama
—
8 -
cui je pens. Diex m'o-troit ke ce ne soit sor son def - fens!
= &
= =
ee
Mo - rir quic se de li n'ai se - cors par tens.
94
The Roman de Horn and the lai
II France deboinaire,
De ta grant franchise
Ne porroit retraire
Nus en nule guise.
Coment porroit faire
Mes cuers nul servise
Ki te peust plaire?
Ice me devise:
Ne te puis plus taire
Le mal ki m’atise;
Ne m’i fai contraire:
Je aim sans faintise.
Example 19 (contd)
uf
% =
ey
oe
—— = > _= = 2 14S SS SE
— 2
= 2 oe = o = _ = =e ae = ;
fA dl Co -
o
nul
oo
ser -
SS
vi - se.
<——_
% a = . = ro = =e =o: = qe
fs
—— 2
=e eee =e Fe ra = 2 ©: = a as
Vm
FG # ts. ear = ra y_meeeeeen oe oe z ai]
Aan? m'i fai con -trai - re; Je tiaim sans fain - ti - se.
At first sight the harping passage of the Roman de Horn seems such a vivid
and detailed piece of writing that the idea of questioning its verisimilitude
barely arises in our minds, and part of its apparent fidelity to contemporary
life lies in the detailed information it seems to provide about the
performance of lyric /ais such as the Lai d’Aélis.
Yet a search through Old French literature soon reveals that this harping
99
Songs and instruments
episode is not unique. Indeed there are passages in other romances which
are so similar to the musical episode in the Roman de Horn that the extract
quoted at the beginning of this chapter begins to look like a complex literary
convention. The key elements in this convention are
(1) the courtly hero/heroine who plays the harp and who
(2) sings/composes works called ais
(3) as part ofa tale richly endowed with what might be called
‘Celtic mystique’—the charisma of Arthurian Britain and all
ancient Celtic realms (especially Cornwall, Brittany and
Ireland).
‘Fresne,’ said Galeran, ‘I have tried out my skill with a new Jai and I
am very keen to teach it to you at once’.... ‘Begin,’ said Fresne,
‘then I will harp and learn the /ai on my instrument.’ Then he began
to play, and she listened, studying the way he cast his fingers on the
strings. When he had listened to the notes he tuned them with his
tuning key so that they were perfectly accorded. The words and the
music were sweet, and he sang and played the Jai until she knew
both the words and the tune; then she tuned her silver-stringed harp
to the lai.
Taking these references together we can begin to see the harping episode
96
The Roman de Horn and the lai
in the Roman de Horn in a fresh light. The passage where Horn sings a lai to
the harp—at first sight such an arresting and seemingly idiosyncratic piece
of writing—now appears to activate a complex narrative motif in which
there are a number of stable elements: the musician is always a protagonist
or an important person in the story; he or she is a courtly amateur; the
instrument which they play (in a most accomplished manner) is the harpe;
the pieces which they perform are called dais; the stories in which these
protagonists appear are fraught with the romance of the ancient Celtic
realms of the North.
I]
The history of this complex motif deserves a study to itself and lies beyond
the scope of this book (although we shall glance at it towards the end of this
chapter). What matters here is that the term /ai is part of the motif. Does
this reveal anything about the performance of pieces such as the Lai d’Aélis
(Example 19)?
It can be said at once that in all cases where a romancer gives examples of
what he means by the /ais which his heroes perform, the pieces presented are
never polymorphous lyric works such as the Lai d’Aélis. Two romances
compiled during the first decades of the thirteenth century will suffice to
demonstrate the point: Guiron le Courtois and the Tristan en prose. These two
works continue the hero-harpist tradition which we encountered earlier in
the Roman de Horn and in Galeran de Bretagne, for they both contain scenes
(shrouded in Celtic mystique) where something called a az is performed to
the harp by some virtuoso courtly amateur. In Guiron le Courtois the poetic
texts of these pieces are inserted into the narrative, and in some manuscripts
of the Tristan en prose these lyric insertions may even appear with their music.
A case in point is provided by a passage from Guiron le Courtois. Here the
author tells how the first ever lai came to be composed. It seems that
Meliadus, king of Leonois and the father of Tristan, has returned home to
his kingdom where he languishes for love of the queen of Scotland. The
extract is well worth quoting in extenso:
King Meliadus did not forget his love, and when he was back in
Leonois he was so unsettled that he did not know what to say for
himself. He was so deeply in love he thought he would die. He
composed songs about his love which he sang night and day, and it
was this which gave him most comfort in this affair. What shall I
say? He suffered for a long time from this love which he did not dare
confess to any man in the world. Eventually, he composed a poem
about his love which was more wondrous and subtle than anyone
had ever composed before, and he set this poem to music such that it
might be sung to the harp (et sor celui dit trowe chant tele que len puet
chanter en arpe), for there was no man in all the world at that time who
o7
Songs and instruments
98
The Roman de Horn and the lai
The knight, who wished for further entreaty than that of the lady
Orgayne alone, replied that he had no skill in singing. All the ladies
then entreated him, and when he saw that they (and the knights who
were there) pressed him so intently, he replied: ‘Then give me the
harp’, and they passed it to him at once. When he had settled
himselfhe began to look upon the queen of Scotland so that she
noticed his attention. When he had tuned the harp to the best ofhis
ability, and according to the music that he wished to play, he began
his song and his Jai at once...
Several manuscripts of Guiron le Courtois give the text (but not the music)
of this Jai: it is a long love-poem in monorhymed quatrains, beginning:
This poem is very similar to many of the so-called /azs inserted in various
manuscripts of another great prose romance, the Tristan en prose. The
majority of these dais, some of which have survived with their music, are
composed of monorhymed quatrains with a melodic scheme aabc (the
melody being more or less exactly repeated for each stanza).'” In one
instance a /ai given the title Lay voir disant is prefaced by a picture of aharper
playing the piece to King Mark at Tintagel (Figure 9). The first stanza
runs:
HM
Example
20
2S Ss
& ant me sui de di - re te = U
—_—————
& : re
que je me sui a- per - che - u
— ee! o> Ss a
re ee $e
pour ch'ai mon lay? (ora. “= (men /= ite = u.
Figure 9 A harpist performs the Lay voir disant before Mark, king of
Cornwall. From a thirteenth-century copy of the Tristan en prose. Paris,
Bibliotheque Nationale, MS fr.776, f.271v. Reproduced by permission.
of the lyric /ai is that each subdivision of the text has its own metrical form
and its own melody, but the stanzas of the ‘Arthurian’ /ais are built on the
opposite principle: they are isometric and are all set to the same melody.
These ‘Arthurian’ lais are something of a mystery. They have often been
compared with the Ambrosian hymn,'* and when they are supplied with
music they are usually laid out in the manuscripts like liturgical hymns in
an antiphonal: with music for every verse (Figure 9). These dais might also
be compared with some of the earliest narrative poems in the French
language. The eleventh-century Passion of Clermont-Ferrand, for example,
is composed throughout in quatrains of octosyllables, and was clearly
intended to be sung, for the first verse is provided with musical notation:!°
100
The Roman de Horn and the lai
It is possible that such stanzaic, narrative poetry lasted into the twelfth
and thirteenth centuries amongst professional entertainers singing songs of
the lives of saints and of the deeds of secular heroes. Perhaps this is the
background of the ‘Arthurian’ Jais.!°
As a coda to Guiron le Courtois and the Tristan en prose there is a passage
from the thirteenth-century romance of Sone de Nausay which also offers a
description of a lai being accompanied on the harp by a courtly amateur,
and where the text of the composition is also given. This poem seems to
reinforce our picture, for while it is not an ‘Arthurian’ Jaz, it is definitely not
a lyric Jai either:'°
The song performed in this extract, Gentilleche et pités, pries pour mi (not
known from any other source) is a somewhat anomalous piece of work. It
has a polished surface that catches many reflections of the High Style
trouvére song: the isometric stanzas of decasyllabic lines; the illusion of an
intense subjectivity (A lui siervir mon cors abandonnai) and exaggerated
sensibility (...qu’en ses bras me pasmay); yet the resemblance to the classic
song-form of the trouveres fades as the text unrolls to more than twice the
length of a normal trouvere chanson.
We have seen that each of the harping episodes in the Roman de Horn,
Galeran de Bretagne, Tristan, the Tristan en prose and Guiron le Courtois activates
a complex narrative motif: the hero-harpist performing something called a
lai in a scene charged with Celtic mystique. There seems no reason to doubt
that these passages embody details of contemporary musical practice (the
harpist tuning-up before his performance, for example), yet it is far from
certain whether the details add up to a whole that reflects reality in the same
direct way. These episodes are not so much snapshots of contemporary life
as oil-paintings, touched and re-touched during many generations.
Yet this leaves a good deal to be explained. Why do we find these detailed
harping episodes in Anglo-Norman and Old French narratives, and why are
they so detailed?
Here we confront some wide-ranging questions about the way medieval
narrators conduct their dealings with music. To begin with the obvious,
these fictional tales must first be read on their own terms as stories; and
stories, as Tolkien pointed out, are like soup in which many ingredients,
102
The Roman de Horn and the lai
some new, and some ancient, lie simmering together.'’ Indeed there are
reasons for believing that the harping passages we have examined are a
mixed broth of ancient and modern in which the stock is provided by a story-
pattern already in existence by c500 AD: the story of Apollonius of Tyre.
III
Shakespeare’s description of the Apollonius story as one that
is a fitting introduction to his own version of a tale which had enjoyed more
than a millennium of popularity before Pericles was launched on the theatre-
goers of Elizabethan London. It first comes into view as a Latin prose-tale,
the Historia A pollonii Regis Tyri, probably composed c500.'® This Historia was
much read in the monasteries of Dark Age Europe and by c1000 had already
been translated into English prose. The story tells how Apollonius, prince of
Tyre, is forced to flee his homeland and is eventually shipwrecked and cast
ashore at Cyrene. He dines with the king, and after the meal Apollonius
plays the lyre so well that the princess of that land asks her father if she may
take music-lessons from their guest. Here is the description of that
performance as it runs in the Old English version of the Latin narrative:'”
for instrumental technique (‘Apollonius took his tuning-key and then began
to stir the harpstrings with skill’) and that same interest in the manner ofthe
performance (‘[he mingled] the sound of the harp with joyful singing’). In
the Old English version of the Apollonius story, as in the Roman de Horn, the
musician is a courtly amateur who astonishes his royal listeners with a harp
(hearpe in the Anglo-Saxon and harpe in the Roman de Horn).
A closer look at the resemblances between the story of Apollonius and the
Roman de Horn leads to a discovery: in many respects they are the same story.
In both tales the hero leaves his native land in a boat, having been driven
from his homeland; he sets off at the mercy of the waves with no idea where
he will come aground. He then arrives in a foreign land and impresses his
foreign hosts with his accomplishments. Eventually he has a chance to
amaze a royal company by the excellence of his string-playing. Every
listener is lost in admiration as the hero performs, but it is the king’s
daughter who is most struck by his talents and she asks that he may become
her music-teacher.
We can make still more of these resemblances if we turn to the story
which, in many ways, is the fons et origo of the harping mystique which runs
through French chivalric literature: the tale of Tristan. Here again we find
many of the same narrative motifs as have now become familiar from the
Roman de Horn and the story of Apollonius:?”
If the resemblances between the Historia Apolloniit Regis Tyri and certain
elements of the Roman de Horn are striking, the parallels between the tale of
Apollonius and the medieval legend of Tristan are even more arresting. In
an important article Delbouille has shown that the Apollonius legend was
well-known in twelfth-century France and that it exerted a powerful
influence upon the development of literary fiction in the vernacular.”!
It is universally accepted that many elements in the Tristan story are of
Celtic—perhaps Cornish—origin,”* yet there can be little doubt that many
details of the Tristan story are derived from the tale of Apollonius, including
the detail of Tristan’s skill as an instrumentalist, an element for which no
convincing Celtic analogue has yet been found.”’ The result of this
confluence of story traditions was therefore a hero-harpist charged with the
potent Celtic mystique which bewitched medieval listeners and readers for
three centuries. As a glance at the story of Apollonius shows, this complex of
story motifs allowed for some passing description of performance practice
(‘Apollonius took his tuning-key and then began to stir the harpstrings with
skill’). This set of story motifs is the generating cell of every /ai/harp passage
in Old French fiction, and the history of the set—how it originated and
came to exist as we find it—probably lies with the history of stories rather
than with the history of music and performance practice. There are non-
musical explanations for the emergence in twelfth-century French literature
of (1) a protagonist who is a courtly amateur string-player, (2) whose
chosen instrument is the harp, which he plays in a virtuosic way and (3)
whose story unfolds in a Celtic world.
Those ‘non-musical’ explanations probably lie with what might be called
‘metaphorical’ realism: the realism which emerges when a network of story
motifs which a culture has inherited from the past becomes a means of
confronting current interests or anxieties. The motif of the hero-harpist was
successful, I suggest, because it provided the means for mediating contem-
porary interest in the figure of the trouvére; many of the harper-heroes in
Old French literature are composers (Horn is perhaps the most striking
example).
105
Songs and instruments
107
(“A i Sar 7 ty rtheta RG
: y) Pe aia gute Fp he -
fcoscted ab ao bf ifed: sed panning ee he m~
ee | ae We 1 le ave ad <ergePreiy,. ane
tS ered eisatirtns Mt ds
drench nate
Pe ER ee Vibcasi
—t- Chapel Ugiag srepreepenelt-eig) sartynal
cube tes > Sai Lhe dwopiy vee pales saath vibe beeline nn
1a On oy are? er + seengesfi mm <9 e whe aw hieee hedcena iesbabe
vy if
~~ y 9 ¢
7
y + . 2 al “y
» = Te :
* ti os 4 ‘ 7 a? A
sd Zz ray? S
« ‘ o ome 7 x
7 . . iM.e = + a 1 Ao
bing : ‘ ; A 4) —_
. mae Oe: : : - t, (Pete &a
. - : . A aan -
oh, ze ery rs oe > 7 ee rye he
—_—— Y ys 4 i - :
no ete RG w if roy bested one oe 4
‘ é 4 “hc as _
rs
-
'-4 13) Says - —,
-
4 yawn » ai)
ll : .
J ‘* ~, Tae
PART 2
PERFORMANCE PRACTICE
Our foreheads felt the wind and rain.
Our youth returned, for there was shed
On spirits that had long been dead,
Spirits dried up and closely furled,
The freshness of the early world.
il Aa) ed 4)
18 (Webitie GIO)
7 rani atria nti
ny ens? “W /
wt 4 wide wet ee
bey
Open-string instruments:
tunings and techniques
...a small kind of tinkling which symbolised
the aesthetic part of ayoung lady’s education.
I
In Part | of this book I attempted to reconstruct an artistic tradition for the
use of instruments in courtly monody. The purpose of Part 2 is to explore
certain technical characteristics of instruments which bear directly upon the
character of the accompaniments they would be able to provide—especially
tuning-patterns and the use of drones or other heterophonic techniques.
It seems advisable to include every major kind of medieval instrument
in this survey and yet the evidence for the use of plucked instruments in
any kind of troubadour or trouvére song seems very slight. Time and time
again it is the words viella, viele and viola which appear in sources that refer
to the accompaniment of the voice and it would be no injustice to the
evidence to devote the whole of this part of the book to bowed instruments.
Nonetheless it would be hazardous to deny the possibility that open-
string instruments participated in the performance of monophonic songs
and I have therefore included them here.
One of the first thirteenth-century theorists to bring his Latin down to the
level of wood and strings is the Englishman Amerus, whose Practica Artis
Musice was completed in 1271. Amerus writes that a semitone!
111
Performance practice
112
Open-string instruments: tunings and techniques
: a ie ae re memos Me dr1d1cte-
ry Sirtird eae
7 ' ee — 3 ral.
ut fru: oe =
frnitns 2ubidue
abicty come [5 ee
fe Rito” ig ee digtTOy-4-08 sep ;
ofc; reprimiac
nus-t adhe DucateG ; v1 pomunt Smoar
Diltiecs firme tont-
Here is the familiar motif of the harper-hero who discards the tuning
which a previous player has used; we remember Thomas’s Roman de Horn.
Yet the author of the Tristan en prose is even more determined than Thomas
to stress the close relation between the hero’s tuning-preliminary and the
uniqueness of his musical skill; Tristan does not just re-tune the instrument
artfully, he does so a sa maniere et a sa guise: in his way and after his fashion.
On another occasion it seems that two adjustments are involved. Tristan
arrives with Hector before the house of Bréhus and is led before a young girl
who sings beautifully. Tristan sings a lai which he composed himself,
preparing for the performance by setting the harp ‘according to the music
which he wished to perform’: selonc le cant quil uoloit dire.° Tristan announces
the piece and then ‘he begins it when he has tuned the harp another time’: 2/
le conmenche adont quant il a la harpe atempre autre fois.’
In the Vienna manuscript of the Tristan en prose we have the melodies for
most of the Jais performed in these scenes, but unfortunately they do not take
us further forward. The music of the lai which the harper-messenger
performs for Iseut (A vous, Amours, ains c’a nului) reveals nothing about the
messenger’s use of scordatura since it is a simple diatonic melody which
might be played on any set of strings set to a major scale.? Nor does the
music reveal what may have been involved when Tristan tuned a sa maniere
et a sa guise; the piece which Tristan re-sets for (La u joe fui dedens la mer) can
be played with exactly the same diatonic arrangement as the piece which
has just been performed (Apres chou que je vi victoire).
Yet while they reveal little in precise terms, these passages from the
Tristan en prose are telling insofar as they suggest a willingness on the part of
aristocratic readers in thirteenth-century France and England to scrutinise
the opening moments of a hero-harpist’s performance and to imagine him
setting the instrument in an artful and individualistic way: ‘... he begins to
tune the strings above to those below’ (commence a acorder les cordes desus a celes
desous); ‘he accords some strings to the demands of others so that those
below reply to those above in true sound and true accord’ (acorder les unes
cordes a la raison des autres si que celes desous respondent a celes desus en droit son et en
droit vers).'° These references may not tell us much about the tunings used by
thirteenth-century harpists, yet the way in which the romancers consistently
associate exceptional harping-gifts with scordatura and individual settings
(rather than with rapidity of execution, for example, or with complexity of
embellishment) is most suggestive and may well be a window onto
contemporary harping.
So may a remarkable passage in the Lumiere as Lais (‘Light to the Laity’),
a religious poem in Anglo-Norman which was completed at Oxford in 1267
by one Pierre of Peckham." Pierre’s attention turns to harping when he
compares the condition of Man living in charity with the well-tuned strings
114
Open-string instruments: tunings and techniques
P15
Performance practice
If the player of a harp arranged in this way wished to play a melody with
what we would describe as a minor character, and also wished to play it
from the fourth string of his instrument upwards (perhaps in order to
accompany a singer who wished to perform at that pitch), he would move
from tuning (1) below to tuning (2):
(1) strings SS oi a a) Oe
For the harpist (who would not perform an infinite number of adjust-
ments such as this, but only those which suited his particular needs and
116
Open-string instruments: tunings and techniques
II
Once tuned, the strings of an instrument become a trellis to shape the
florescence of the player’s invention. The harp and psaltery offered the
medieval player an elaborate lattice of strings with considerable scope for
interlacing his music. It is possible to imagine a medieval harpist or
psaltery-player producing drones with the lower strings, for example,
having perhaps assigned a particular course, or courses, to that function
(compare the special drone-pipe of the thirteenth-century organ, the
bordunus organorum).*' Psaltery-players would have come into their own with
the technique of flourishing a melody with parallel intervals in the manner
of improvised organum; once they had tuned the unison pairs of their double-
courses into octaves or fifths they could produce thickened volutions of
sound from their standard playing-technique.
This is all speculation, of course; yet there is scattered and fragmentary
evidence that all of these techniques may have been exploited by medieval
harpists and psaltery-players. We begin with the word bourdon. When used
in connection with parts of instruments in the thirteenth century its
meaning usually lies in the area of ‘something which produces an unvarying
118
Open-string instruments: tunings and techniques
note, drone’. Thus the bagpipe’s drone was termed bourdon and the drone-
pipe of an organ was a bordunus organorum; the lateral, unstopped strings of
fiddles were borduni.”* In some cases the term borduni lingered well into the
Renaissance; we find the bottom course of the lute named bordon in late
fifteenth-century Spain, for example,”* and the lowest string of the dulcimer
named bourdon in seventeenth-century France.”*
So it may be significant that the lowest strings of the harp appear to have
been called bourdons in thirteenth-century French (or Anglo-Norman, to be
precise). Let us turn back to Pierre of Peckham’s Lumiere as Lais. Pierre
relates that the harp may be tuned in various ways with differing patterns of
tones and semitones; yet however diversely it is tuned, he continues, there
will always be a need for the octave, fifth and fourth:?
Kar saunz ices ne purra mie
En harpe estre sun de armonie.
Les essais e les burduns
De ces treis unt ausi les suns.
The terms essais and burduns are presumably intended to cover the whole
string band from high to low, and as all known usages of bourdon-words
display the idea of deepness, the burduns are presumably the lower strings
and the essais the higher ones.”° Everything which we know about the
application of bourdon-words to components of instruments in Pierre’s
lifetime suggests that the lowest strings of the harp acquired the name burdun
because they were (or had once been) used for drone-playing.
A few crabbed words squeezed into the margin of a copy of Alain de
Lille’s Anticlaudianus may help us trace these harp-bourdons a little further.
Alain tells how Natura desires to make a perfect man and how Prudentia is
sent to Heaven as an ambassador to obtain a soul for him. Journeying
through Air and amidst the Music of the Spheres, she hears the ‘celestial
cythara’, sounding so softly ‘that it performs the function of a string lying
lower’ (.. neruique iacentis/Inferius gerit illa uicem edie. WAt, this. point a
fifteenth-century copy of the text in the library of Balliol College, Oxford,
has an important annotation:
In accordance with prevailing medieval usage this cithara with its bordon may
be a pillar-harp.”?
If harpists used drone-strings (at least in the twelfth and thirteenth
centuries) then two references to open-string playing from that period gain
something in force and clarity. In his Topographia Hibernica of 1185-8 Gerald
of Wales gives a striking account of the string-playing skills of the Irish. The
musicians of Ireland, Gerald relates, ‘play quite freely the tinklings of the
thinner strings along with the duller sound of the thicker one’ (sub obtuso
grossioris chordae sonitu, gracilium tinnitus licentius ludunt).°° This reference to ‘a
thicker [string]’ in the singular may be nothing more than a stylistic flourish
adorning the simple thought: ‘they make the deep strings accompany the
higher ones’. Gerald’s writing is full of such plain earthenware as this, fired
to a high glaze by rhetoric. But in this case his meaning may be made of
finer matter for he had almost certainly heard Irish harping,*’ and his
‘thicker string’ with its ‘deep sound’ might conceivably be a drone-bourdon
accompanying the melody in the treble.
There may be an echo of such a bourdon-technique in an unpublished
psalm-commentary now in the Bodleian Library at Oxford, and probably
dating from the twelfth or thirteenth century. Here an anonymous
commentator steps outside accepted traditions of describing the Psalmist’s
cithara and records that ‘... two hands are used in playing the czthara; one
hand continually (iugiter) plucks the lower strings; the other hand plucks the
higher strings, not continually, but at intervals and in turn’ (non iugiter sed
vicissim et interpollatim).** There is no escaping the firmness of the author’s
distinction between lower strings which are plucked continually and higher
ones which are plucked intermittently. Perhaps the lower strings are drones,
plucked constantly to accompany the higher strings which are touched ‘at
intervals and in turn’ as they are used for melody.
A drone, firmly planted, can provide a player with the trunk of a
technique, but it is melody that ramifies his art. It will never be known how
medieval harpists and psaltery players executed their tunes, what graces
and ornaments they used, what tricks of phrasing they employed. Here and
there a poet, busy about some other matter, allows us to glimpse a little of
what has been lost in oblivion; the author of Galeran de Bretagne, for example,
mentions a melodic style (or playing technique?) called ‘Saracen notes’,**
but nothing is known of them. However, amongst these poetic references a
few fragments of evidence suggest some use of parallelism and improvised
heterophony. The Sermones of the twelfth-century German satirist Sextus
Amarcius incorporate a luxuriant description of a performance upon the
chelys (?lyre)** where Sextus even notices the material from which the strings
are made (see Appendix 4:8). Fortunately his eye for detail also extends to
the performer’s technique: as the musician delivers a programme of
narrative songs (or declaimed poems?) he ‘repeatedly adjusts the melodious
strings in fifths’ (Jlle fides aptans crebro diapente canoras).*° This sounds like a
tuning-prelude, but ‘adjusts’ may not be the best translation for aptans;
120
Open-string instruments: tunings and techniques
Il
We have been concentrating on the harp where the material is richest. But
what of the members of the zither family—the psaltery, the rota, the medius
canon and the rest?
The psalterium was of passing interest to some of the music-theorists since
it had a string for every note and therefore lay within the conceptual field of
traditional music-theory. The Speculum Musice of Jacques de Liege, for
example, contains several passages in which various matters of musical
mathematics and acoustics are explored through the psalterium (‘if the string
ofa psaltery is touched with a quill in a deliberate or accidental way ...’).1°
Some theorists may even have compiled treatises on the proportions of
psaltery uss for Engelbert of Admont seems to refer to such essays in his
treatise.*’ No such work is known to survive. The closest we come to a
122
Open-string instruments: tunings and techniques
Figure 11 The rota. Note the double tow of tuning-pegs and the position of
the player’s hands. The double string-band of the rota is mentioned in the early
fourteenth century by Petrus de Abano. Twelfth-century sculpture from
Surgeres.
123
Performance practice
medius canon 64
rota (pie app
unspecified 28
unspecified 44 .
44 reduce to 11 notiores
64 reduce to 16 notiores
124
Open-string instruments: tunings and techniques
psaltery in more than name.*° The drawing displays one striking feature: the
lowest string, which stands on bridges of its own, is tuned a fourth below the
next string. In the text Mersenne calls this low string a bourdon. Now many
Gothic illustrations of psalteries, from all periods and places, show players
with one finger extended over the lowest course (or courses) of their
instruments;*’ this posture suggests that they may be plucking just such
bourdons as are shown in Mersenne’s diagram of the psalterion.
There is one more detail in Mersenne’s account of the psalterion which
may be a relic of medieval practice. He records that the courses may
sometimes be tuned in octaves together with fifths and fifteenths ‘to
augment the harmony’.*® If such parallel harmony was acceptable in the
dulcimer-playing of seventeenth-century France (and Mersenne seems to
have admired the psalterion), it is not likely to have been shunned in the
Middle Ages when the technique of parallel organum was still cultivated by
some liturgical singers.*? It is tempting to build on Mersenne’s evidence and
try a very tentative reconstruction of a possible Gothic psaltery-tuning after
this fashion. As the player’s quill plectra elicited a ringing sound from the
metal strings, and as the melody moved in parallel fifths over a drone
(which may itself have been compounded from unisons and fifths), the
psaltery’s sound-picture would have been a colourful one and a world away
from.the silvery thread of melody which we may hear in our mind’s ear upon
first seeing images of psalteries in medieval manuscripts and carvings.
125
10
Jerome of Moravia and
stopped-string
instruments
For wise he was, and many curious arts,
Postures of runes and healing herbs he knew.
I
Jerome of Moravia’s chapter on fiddles opens with the rubeba, an instrument
whose Arabic name and single pair of strings suggest a relationship with the
modern rabab of Morocco. In Jerome’s lifetime we find this instrument
depicted in one of the celebrated Cantigas manuscripts, probably produced
in Seville c1275. At this date the rubeba may only just have reached as far
north as Paris (for the term rubebe first appears in Parisian vernacular
sources around 1270)' and Jerome seems very keen to define the rubeba (‘a
musical instrument played with a bow ...’) as if itwere unfamiliar to his
readers.
According to his account the two strings of the rubeba were tuned a fifth
apart—another link with the Moroccan rabab—allowing the player to
produce a full octave without using the weak fourth finger (although Jerome
allows that it may be used to produce a ninth note). He is quite explicit that
once the little finger has been employed the rubeba can ascend no further and
the technique of changing position is nowhere envisaged in his chapter.
The viella or five-stringed fiddle comes next (Figure 7), a much more
versatile instrument which has not one tuning but three:* :
] D G d
2 D G g
“i r [a eb c c
126
Jerome of Moravia and stopped-string instruments
l d Ge mid dilwends
2 d Canoe aa
8 G Gad Amie
In tuning | the d-course, which Jerome calls the bordunus, runs to the side of
the fingerboard and cannot be stopped by the fingers.
Tuning |
The syntax of the first tuning is perhaps that the G and g strings form an
octave pair while the two d' strings make a unison double-course:
PLO Gores
Viewed in this way the tuning looks like a simple fifth g-d' which has been
heaped-up (1) by doubling the g an octave lower, and (2) by filling in that
octave step with d. It is an accretion ofideas and devices that suggests a vital
and generative playing-tradition. The tradition stipulated five strings for the
viella (the number mentioned by Elias of Salomon); in this first tuning the
five-string set has been formed into three courses, each of which is the focus
ofa separate idea: a unison pair, an octave pair, and a single string running
to the side of the fingerboard which may be plucked with the thumb as well
as bowed. Each of these ideas represents a distinct design upon sound and it
is this which makes Jerome’s treatise so engaging; few other writings take us
to that part of the medieval string-player’s mind where sounds were
imagined and then sought after.
What did thirteenth-century fiddlers hear in their mind’s ear which
brought them to this tuning? A comparison with the modern violin may be
instructive here. The four courses of a violin do not represent different
designs upon sound, but one design: each course is single; each is planned to
blend in sonority with the others and each stands the same distance away
from its fellow. In addition, the violin’s tuning is not chordal (a pile of fifths
produces cacophony) but is designed to produce a single line passing
smoothly from string to string without any disjunctions in the quality of the
sound—only a pleasing rise in brilliance as the melody moves to higher
strings and an increasing warmth and vibrancy as it falls to lower ones.
The fundamental contrast between the modern violinist and the thir-
teenth-century viellator is therefore that the violinist generally thinks in
terms of pure monophony whereas his medieval predecessor cultivated
devices that would produce both melody and heterophony: auxiliary noises
clustering around any tune he played. This distinction, a measure of the
127
Performance practice
great difference between the thirteenth-century fiddler’s ear and the modern
violinist’s, is clearly embodied in the tunings of the two instruments: the
viella chordal, the violin melodic. The accordatura of the viella points to the
thirteenth-century fiddler’s distinctive tendency to think of his art in terms
of simultaneously sounding strings.
The next important contrast is that whereas the’violinist seeks a certain
homogeneity and continuity, the medieval fiddler who used Jerome’s first
tuning was apparently looking for disjunctions. This tuning unloads three
different colour-making devices upon five strings, and each device is used
once only: unison pair, octave pair, and bourdon. As the player crossed from
course to course, from Gg to d'd'’ and back again, his melody would
comprise a series of parallel octaves one moment and a chain of reinforced
unisons the next. The mixture of harmonics in the tune would be constantly,
and abruptly, changing. And there would be a further disjunction in that all
the notes played upon the double course Gg would set up an octave-
ambiguity leaving the ear uncertain as to where the melody was located,
high or low. Once the player shifted to the top course d’d’ this ambiguity
would be suddenly (but only temporarily) resolved as the octave doubling
dropped away, leaving two strings sounding in a powerful, perhaps even
strident unison.
To these disjunctions would be added a third: the percussive, plucked
drone of the bordunus.
The purpose of these devices was presumably to create an impression of
density, changeability and abundance with just five strings: to confuse the
ear’s awareness of what was actually happening by setting up the kind of
brouillard sonore, ‘sonorous mist’, so beloved of non-Western string-players in
many traditions and®
Tuning 3
At this point it will be helpful to jump to Jerome’s third tuning, which
128
Jerome of Moravia and stopped-string instruments
GG d ec
Example 21
More than any other information we have about the tuning of medieval
instruments, Jerome’s third accordatura highlights the distance which
separates modern notions of string-playing from medieval ones. It shows
how the viellator’s tendency to think in terms of simultaneously sounding
strings could go so far as to allow the melodic scope of five strings to wither
down to the scope of only one string: a fifth. Self-accompaniment, through
heterophony, is the priority.
129
Performance practice
Tuning 2
Jerome’s second tuning is an exception to all this, and it may be some
confirmation of the interpretations we have offered so far that Jerome also
saw it as an arrangement requiring some special comment. In this tuning
the strings produce the same notes as in tuning | save for the top string,
which lies a fourth higher:
d Gg d' g’
This second tuning, says Jerome, ‘is necessary for secular and all other
kinds of songs, especially irregular ones, which frequently wish to run
through the whole hand’.® As we have already seen above (p.65), this is a
problematic remark, for it may be doubted whether there are any pieces of
thirteenth-century music in existence which run through the whole gamut, a
compass of two octaves and a sixth.
Now we move from uncertain ground to the most dangerous terrain. After
the strange landscape of tuning 3 (GG d c’c’) with its small shoot of melody
in a bed of drones, tuning 2 (d Gg d’ g’) seems to bring us home to country
that we recognise. It has a wide compass (two octaves and a fifth when the
top string is fingered) and therefore appears to have been shaped by the
concern for playing melody through a generous compass which has
characterised much art-bowing since the Renaissance; it puts us in mind,
perhaps, of a round-bridged viella where all strings may be played
individually in the interests of clear, smooth melody, as on a violin.
Yet is this necessarily an accurate view of Jerome’s tuning? Consider the
arrangement and ‘pitch’ of the strings. This second disposition has often
been presented as a continuously ascending series Gd gd’ g’, while Jerome’s
use of gamma to denote the lowest string of the set (nominally equivalent to
G) has prompted the view that the bottom of some thirteenth-century
fiddles actually sounded in the region of modern G at the bottom of the bass
clef—or even lower.’ But this is probably a false picture of Jerome’s
instrument. To take the pitch question first, if Jerome had called the lowest
string anything other than gamma he would have been unable to record the
stopped pitches of tuning 2. That tuning spans two octaves and a fifth; a
tone less than the musica recta gamut. If Jerome had started mid-way in the
gamut, for example, he would have exhausted the notational signs available
to him before he reached the ceiling of the instrument’s compass.
As for the order of the courses, Jerome says that the first four strings
produce the same notes as in tuning | (d Gg d’ ...) save that the top string is
tuned to g’ and the d bordunus is allowed to run over the fingerboard so that it
may be stopped; he is quite explicit about these details and so it seems that
tuning 2, like tuning 1, is re-entrant.®
What of the G and g’ strings? Do they form an octave course as they seem
to do in tuning 1? In all probability they do. The open strings of tuning 2
130
Jerome of Moravia and stop ped-string instruments
span two octaves(G—g"), more or less the limit of what seems to have been
attainable with medieval gut on instruments such as the viella where all the
strings must be of the same length.” Much further than this and players
would have had to endure constant breakages at the top together with poor
tone and tuning instability at the bottom. It seems likely therefore that the
tone of the G in tuning 2 d G gd’ g’ would have been only just satisfactory,
and in this tuning, where all the strings were used melodically, would
probably have required the reinforcement which it also receives in tuning |
(where it is doubled at the octave) and in tuning 3 (where it is doubled at
the unison). We seem to have been led away somewhat from the idea of this
tuning as a vehicle for melody or even sonority.
When the viellator ran through the compass of his instrument it would
have sounded like this (relative pitch):
Example 22
(relative pitch)
strings
=e
=o =
3
SIERO E. ee
eo #£
ra a oe °
—
2 2
l d Ggd'd'
2 d Gg d g’
Although the first four strings of each arrangement are tuned in the same
way, and while the replacement of the lateral d bourdon used in tuning 1
back on the fingerboard for tuning 2 would be only a minor operation, a
fiddler who wished to pull the top d’ string of tuning | up a fourth to g’ for
tuning 2 would encounter problems of string-breakage and tuning instabil-
ity. This does not prove, of course, that players did not sometimes create
tuning 2 out of tuning 1, but it might have made their lives easier to carry a
second viella.
II
A crucial determinant of abowed instrument’s technique is its bridge.'” If it
is round, as on a violin, then the player can bow each string individually; if
13]
Performance practice
it is entirely flat, he must bow all the strings at once. Whence an insuperable
problem: we know very little about the bridges of medieval bowed
instruments.
Bridges are often shown in pictorial sources, but it is very doubtful
whether this material will ever provide enough accurate and trustworthy
evidence to serve as a basis for reconstructing thirteenth-century fiddle-
technique. The problem is not simply that medieval artists must often have
been ill-informed about technical minutiae such as fiddle-bridges; it is also
that a flat-topped bridge can be turned into an effectively round one by
cutting string-slots of graded depth.'' This simple fact vitiates much of the
iconographical evidence which has been brought into the debate concerning
round and flat bridges on medieval fiddles. (Iconographers will find this a
harsh judgement, but they will be the first to admit that very few pictorial
sources are precise enough to incorporate such a detail as this.)
Jerome’s evidence throws a little light on this problem for his chapter
clearly points to the use of heterophonic, self-accompanying devices
(especially the drone) which would be well-served by having all the strings
lying in a single plane or, perhaps better, in a slight arc. Tuning 3 GG dc'c’
seems hardly intelligible unless we interpret it as a drone-block GGd with a
melody running above it, while tunings 1 and 2 are completely concordant:
each forms a consonant chord without a third, suggesting that the bow is
intended to touch several strings at a time. Tunings like these, mixing
fourths and fifths, are not the most advantageous ones for melodic playing
(compare the violin’s neat pile of fifths), but they are ideal for the kind of
polychordal execution where the player uses only the ‘home key’ of his
instrument (in this case g, to use modern concepts and terminology) and is
therefore always safe to fill out his melody by brushing adjacent strings. I
suspect that many thirteenth-century fiddlers played in this way, and it is
doubtful whether many of them can have shared the modern notion of a
‘tune’: a melody abstracted from its accompaniment and from the reflexes of
eye and hand necessary to play it on a particular instrument.
What happened when thirteenth-century viellatores played was presum-
ably something like this: the five strings were disposed in a slight arc so that
players could sound all five at once by applying some pressure to the bow, or
choose them individually or in pairs by bowing delicately; with the five
strings arranged in this way an appreciable number of combinations would
be possible and melodies would be embedded in a constantly changing
‘drone’.
Il
We have lingered long with the viella, but what of other fingerboard
instruments: the lutes, gitterns and citoles? A few scraps ofevidence suggest
that they may have been tuned and played in ways comparable to the
132
Jerome of Moravia and stopped-string instruments
practices of viellatores. We have already seen that the author of the Summa
Musice records that fingerboard instruments ‘are tuned in the consonances of
octave, fourth and fifth, and by putting down their fingers the players of
these make tones and semitones for themselves ...’.!* This certainly sounds
like the kind of tuning which Jerome describes, and the advantages to a
medieval lutenist or gittern-player of a tuning mixing octaves, fourths and
fifths to produce some kind of accordatura are obvious. Everything we have
said about the fiddler’s wish to turn his instrument into a heterophonic, self-
accompanying resource might also be applied to players of plucked
fingerboard instruments.
Here the evidence of certain sixteenth-century lute-duets, where one lute
plays a constant drone to accompany the florid melody of another, is most
suggestive. These pieces may well be a relic of medieval practice'* and the
same might be said for the drone-tunings which some of them employ. In
the light of these pieces a tuning such as dad’ g' (which accords with the
evidence of the Summa Musice and of some sixteenth-century lute sources)'*
seems a possibility for Gothic lutes, gitterns and citoles. By stopping the top
course a tone above the nut, the player would produce a full accordatura of
dad a’. The gap of a fourth between the top two courses breaks the
accordatura but is none the less ideal for drone-playing; it is comfortable for
the player to have a finger or two on the fingerboard when the full drone is
produced—to hold down the drone, as it were—rather than to find his left
hand suddenly redundant every time a sweep of all four strings is required.
133
wisp
Conclusions: voices and
instruments
The lai
As far as can be discerned, the references to accompanied /ais in romances
such as the Tvistan en prose do not refer to the polymorphous lyrics which
musicologists (following an established usage in both Old and Middle
French) now call /ai (Music example 19). In those cases where it can be
established what a romancer means by Jai (i.e. when the song in question is
quoted, either as a song with music or simply as a poem without notation),
the Jai is usually a strophic song in quatrains of a kind scarcely found outside
the romances (Music example 20).
Dance-songs
Refrain songs which could be performed for dancing such as the dansa
(Music example 8) and rondet (example 5) seem to have been closely
associated with accompaniment both in Occitania and France. However,
instruments do not appear to have been much used when songs of this order
were performed for the populous corea.
The pastourelle
The genre of the pastourelle in both north and south seems to occupy an
interesting middle position between the High and Lower Styles. The
pastourelle is almost invariably narrative, and shares a great deal ofits lyric
registration with simple dance-songs (the outdoor setting with rustic girls
and shepherdesses; diminutive suffixes in -eéte, and so on). As we have seen,
the Leys d’Amors associate the pastorela with a melodic style which should be a
‘little more fast and lively’ than that of the chief High Style forms, the vers
and canso, and the music of some French pastourelles is full of the
conspicuous melodic patternings and short-range repetitions which we have
associated with instruments; the celebrated pastourelle by Moniot d’Arras, Ce
fut en mai, for example, which tells in its opening stanza how the poet heard a
viele in the woods, is effectively an estampie with two puncta in its musical
form. This is an exceptional piece, perhaps, yet there seems good reason to
believe that pastourelles may often have been accompanied.
Monophonic conducti
In all probability the performing traditions of these pieces were never
closely involved with the decorums of the High Style song. Accompaniment
may well have been used on a regular basis.
135
Performance practice
Techniques of performance
Most accompaniment seems to have been provided by solo instrumentalists.
In Jean Renart’s Guillaume de Dole, for example, accompaniment for the
voice is almost invariably provided by a solo fiddle, and there is no firm
evidence that mixed groups of string, wind and percussion performed in
courtly monody. Even references to small ensembles of related (or of the
same) instruments are rare (see above, p.30, and Appendix 3:5 (last
reference) for two notable exceptions). In the unique manuscript of the
Doctrina de Compondre Dictats the dansa is said to be accompanied ‘by
instruments’ (ab esturmens), yet even here there is doubt, for it seems that the
scribe originally wrote esturment in the singular.’
This conclusion may be widened momentarily into a proposition that
much—perhaps most—monophonic music of combined voices and instru-
ments was provided by solo musicians in the twelfth and thirteenth
centuries. To turn from Old French literary evidence to the account books of
the English royal court, for example, is to encounter a similar picture of solo
minstrelsy. As Richard Rastall has pointed out in an important study:?
Alternatim
In addition to techniques of simultaneous participation it is apparent that
solo singer/instrumentalists sometimes organised their performance on an
alternatim basis—not surprisingly, since these musicians had only three
colours at their disposal: voice alone, instrument alone, and voice and
instrument together.
There are references to such alternatim performance scattered here and
there in medieval literature and we must go far afield to find them. The most
famous, needless to say, is the passage incorporated in the Roman de Horn of
c1170 (see above, p.5), where Horn sings and then performs exactly what he
has sung upon his harp. At about the same time Gerald of Wales describes
how a band of horsemen travelled with a fiddler and singer going before
them ‘who replied to the notes of the song on his fiddle in an alternatim
fashion’ (cantilenae notulis alternatim in fidicula respondentem).’ This is reminis-
cent of a passage in the Old French romance of Claris et Laris where a
minstrel sings a refrain-song and doubles the refrains upon his fiddle (see
above, p.84). A further reference to alternatim practice—and a most
striking one—appears in a thirteenth-century gloss on Martianus Capella;
here a procedure adopted by minstrels (joculatores) is described: ‘while they
play their instrument, they stay silent; and while their instrument is silent,
they sing’ (dum instrumentum suum tangunt, silent; et dum silet instrumentum suum,
cantant).*
There is ample evidence that such alternatim procedures sometimes took
the form ofpreludes and postludes. Johannes de Grocheio reveals that when
fiddlers performed High Style songs they were accustomed to tack a
postlude on the end which they called modus.” This is a well-known
reference, but there is another, less familiar allusion to this practice in a
treatise on the virtues and vices, Le Somme le Roi (1279), compiled by friar
Laurent at the request of Philippe III; here Laurent compares his own
prologue to the Lord’s Prayer to une entrée de viele, a fiddle prelude.° The use
of such preludes on the harp is abundantly attested in the 77stan en prose.’
137
Performance practice
Tessitura
To judge by pictorial sources (a treacherous source of information in this
context, admittedly), most stringed instruments of the twelfth and thir-
teenth centuries must have played in a tessitura very roughly equivalent to
that of the modern violin or viola. The history of instrumental compass from
c1300 to c1500 is one of expansion downwards—the development being
particularly clear in the case of the lute which began its life in the West with
four courses (thirteenth and fourteenth centuries), then changed to five
(most of the fifteenth century) and then to six (the end of the fifteenth
century). Given what is known of the string materials available to medieval
lutenists, it seems unlikely that this increase in string numbers was
accompanied by a significant movement upward in pitch for the top strings;
it may be that the top strings of a fourteenth-century lute lay in much the
same pitch-range as the top strings of a sixteenth-century lute of comparable
string-length (and equipped with comparable strings). Roughly speaking
we may suppose that the prevailing tessitura of much twelfth- and
thirteenth-century string-playing was probably ranged upwards from the ¢
at the bottom of the viola, perhaps (and I think this more likely) only from
somewhere in the region of g at the bottom of the violin.
Turning now to the human voice we find that the music theorists of the
thirteenth and fourteenth centuries usually give (more or less) two octaves
as the compass of the voice.® Putting aside the delicate (and probably
unanswerable) question of whether the falsetto voice was cultivated in the
performance of troubadour and trouvere song,’ and restricting our attention
(as the theorists do) to the male voice, we may propose something in the
region of F—f’ as roughly corresponding to the two-octave span which the
theorists give as the compass of the vox humana.
If these are tolerably accurate assessments of vocal and instrumental
tessitura then it is plain that male voices and stringed instruments (which
we have placed at around violin or viola pitch) overlapped for about an
octave: roughly speaking, ff". This means that many stringed instruments
would possess many notes which lay too high for voices to reach—and this
is what certain theorists say.!° A tentative conclusion would therefore be
that any accompaniment which stringed instruments offered to the perform-
ance of lyric songs by male voices is likely to have lain in the same tessitura
as the voice (our crude calculations show that ‘fundamental’ drones would
have been feasible) or to have lain above it—perhaps doubling at the fifth
(as in Jerome of Moravia’s advanced fiddle technique) or at the octave.
138
Appendix I
Terminology of musical
instruments
39
Appendix 1
Palsgrave renders three English words (croude, fyddell and rebecke) with a
single French term (rebecqg). One explanation is that all of these nouns were
employed willy-nilly in the sixteenth century: hence the proposition that
speakers of the time had carried ‘confusion to the point of ridicule’. Yet
there is another explanation: that Palsgrave believed, and was right in
believing, that the single French word rebecg would serve the Englishman in
all situations where, at home, he might have used either croude, rebecke or
fyddell. This workmanlike approach to definition is almost universal in
popular language dictionaries and phrase-books of all periods. Here is a
table comparable to the one based on Palsgrave, compiled from a modern
authoritative dictionary of French:
Here the single French word culotte(s) is equated with four English words:
‘shorts’, ‘tights’, ‘breeches’ and ‘knickers’. Certainly the table could be
improved: un short is now acceptable French usage and ‘knickers’ would
perhaps be better rendered as ‘culottes de femme’. Yet what the list gives is
enough for many purposes and there is no confusion amongst English-
speakers as to what the words ‘shorts’, ‘tights’, ‘breeches’ and ‘knickers’
mean. In the same way there is no feeling amongst French-speakers that
culotte is a vague or imprecise term. It is easy to imagine how a Frenchman
140
Appendix 1
equipped with this table alone might have difficulties in an English clothes-
shop and become very confused indeed, but that confusion would result
from a mismatching of verbal resources between English and French, and
not from deficiencies in the resources of each language.
The second reason why medieval instrument-terminology often appears
confusing is that some influential writers have dealt lightly with some
important distinctions: of language (English is not the same as French;
vernaculars are not the same as Latin); of date (an eighth-century word is
not the same as a thirteenth-century one); and of usage (learned, bookish, or
Latinate usage is not the same as day-to-day spoken usage). Let one
example from a classic English work suffice. In his Old English Instruments of
Music Galpin reproduces a drawing of a lyre from an Anglo-Saxon psalter of
cl1030—50 now in Cambridge University Library (Galpin’s plate 38). Here
are three references to the illustration from his text:
in an Anglo-Saxon Psalter of the early part ofthe
eleventh century ... the bowed Cruit is seen in the
hands of Asaph
Crowd—early 11th century (Cambridge)
... Asaph is playing on the bowed Rotte (the Crwth or Crowde)
This is a remarkable muddle in which cruit, crowd (and crowde), rotte and
crwth are all used to denote the same instrument. C7ruzt is an Old Irish word.
Crwth is its Welsh cognate, not recorded before the thirteenth century. Rotte
is recorded in northern European Latin as early as the sixth century and rote
is its Old French relative, not recorded until the twelfth century. As for
crowd(e), this is a Middle English borrowing of Welsh crwth not recorded
before c1300. None of these words can be shown to have existed in Anglo-
Saxon.
Finally, the considerable diversity of both the instrument-names (as they
are recorded in texts) and of the instruments (as they are shown in pictures)
goes far to confuse us. This richness in the material has often been
emphasised—indeed it has been over-emphasised and the time has come to
redress the balance. As for the names, once we have discounted variations of
spelling within one language (thus, for example, Middle English fithele,
vydele, fethele, and so on) and gathered cognates together from various
languages (fithele, vielle, fidula, vyhuela, viola, for example) it is the consistency
of the names, and not their diversity, which strikes us. Many of them can be
reduced to a stable skeleton of two or three consonants present in virtually
all the recorded European languages of the Middle Ages, but fleshed out
with vowels in different ways in the different tongues. The ‘fiddle’ words
listed above, for example, all seem to be generated from something like this
(where asterisks denote the variable vowels, and brackets enclose any part
of the skeleton which may be dispensed with):
f/v *(d) *1*
141
Appendix 1
Se Lee (a)
f/v*"(d)ael =
whence Middle English sautrye, Old French psalterion, Old Castilian salterio,
Middle High German psalterium, and so on. Most of the major instrument
terms of the later Middle Ages appear to have been international ones.
It is often supposed that the typological diversity of medieval instruments
would have made accurate use of terminology impossible, but there seems
little to recommend this view. A modern analogy may be helpful. Since the
1930s ‘guitars’ have proliferated in almost every conceivable shape and size
(makers of electrified instruments have been particularly inventive);
‘guitars’ have been built which are arrow-shaped, rectangular and lyre-
shaped as well as figure-of-eight-shaped in the traditional design; they have
been strung with metal and plucked with artificial fingernails or with a
plectrum, and strung with bare and wound nylon and played with the
fingers; ‘guitars’ have been built in both 6 and 12 string forms; equipped
with round soundholes, with f-shaped soundholes like a violin, with flat
bellies, and with curved bellies like a cello; some types have been built with
electronic pickups. Although epithets are sometimes used to distinguish
these various types of guitar (‘folk’, ‘jumbo’, ‘electric’, ‘Spanish’, ‘acoustic’,
and so on), all these instruments are readily identified by speakers of
modern English as ‘guitars’, without any sense of confusion. Could not the
same thing have happened in the Middle Ages?
But the most important point is that when we assemble the sources that
reveal the names which specific instrument-types bore in the Middle Ages
we find evidence of considerable consistency 1n usage. In the remainder of
this appendix I have assembled some of the most detailed and revealing
sources (principally drawings of instruments with their names written by
them). What emerges from this list is that there were certain pan-European
traditions of nomenclature which lasted throughout the later Middle Ages.
Identifications
In this section the labelled illustrations inventoried above are listed by
number.
Fiddle names
The available evidence shows that Old French viele, and its Latin offspring
144
Appendix 1
viella, were principally used to denote bowed instruments.'? The names viele
and viella are usually associated with five strings (3 [p.20], 12 [p.79], 13 and
14 [pp.20 and 26]), although Amerus, in a general reference to fingerboard
instruments, mentions ‘four or five or less’ (12 [p.79]). Source | is the only
text to mention the possibility of equipping a viella with up to seven strings
(see above, p.57). The body-shapes of these instruments reduce to two main
types: the ovoid, where there is a firm distinction between neck and body,
and the piriform, where the body and neck blend into one another. It seems
likely that the names viele/viella straddled this morphological boundary (and
also the boundary between the tri-chordic and pentachordic traditions of
stringing). It is possible, however, that the piriform instruments often
attracted the name gigue (see below). This leaves the tri-chordic octoform
fiddles played in the lap, very common in pictorial sources before c1300. In
source 4 (Paris MS), such an instrument (and a piriform fiddle) is labelled
viola. It would seem, therefore, that the viele, viella, viola complex covered
most (perhaps all) bowed instruments before 1300.
Rebec names
Although the term rebec is recorded before 1300" it is primarily a late-
medieval word. Old French texts show forms closer to their Arabic parent
rabab. Jerome of Moravia describes the rubeba as a bi-chordic bowed
instrument (13) and Abano lists the rebeba as a two-stringed instrument (15,
Particula 19, problem 3). The guiding factor in the use of names built upon
r—b stems would seem to have been smallness. Jerome of Moravia’s
introductory remarks imply that the rubeba is smaller than the viella—and
not simply because it has only two strings (13, p.88, lines 4—5). In the
fifteenth century Gerson writes of the biblical symphonia that ‘some think the
symphonia to be the viella, or rebecca, which is smaller’,!? while Tinctoris
describes the rebecum as ‘very small’ (valde minus) in his De Inventione et Usu
Musicae.'© The term rubebe does not appear in Old French sources until
c1270,'’ and Jerome of Moravia seems to have regarded the rubeba as an
instrument which might be unfamiliar to his readers. The word therefore
belongs to the very end of our period and the sources suggest that it
generally denoted a two-stringed bowed instrument, played in the lap.
Gigue names
Renaissance usage points to bowed instruments, and the same meaning is
clearly implied for Middle High German gyge in Der Busant (Appendix 4:26).
Abano’s statement that the ziga has four strings points to a fingerboard
instrument. In source 4 (Sloane MS), the words Giga vel lira are written
above a pillar-harp (only giga appears in the Paris manuscript). However,
lira is written in a later hand in the Sloane manuscript and may be an
145
Appendix 1
Ss:
aip jee pec
on
e S 3hPane
aphended
pibs
meet WebOth
Oe thstf
ass
aatSt putedokt
en XY Pe
ee eae i We ry
ua in. bot bene ¥ ! Oe
attempt to supply the more common (or the correct) Latin term for the
harp. Giga might then belong to the hurdy-gurdy nearby, which has no
label, or to the gittern, which is also unlabelled. The latter suggestion
accords eit the gloss Hec giga: getyrne in a fifteenth-century English
Nominale.'® As Laurence Wright has shown, the gittern was a piriform
146
Appendix 1
short lute so there is no difficulty in assuming that the term often (usually?)
denoted piriform bowed instruments during the Middle Ages.
Lute names
There is almost no medieval evidence apart from Amerus (12, p.97). There
can be no doubt, however, in view of usage in virtually all the vernaculars of
Renaissance Europe (and in medieval/modern Arabic) that terms such as
Old French luthz generally denoted a plectrum-plucked short-lute with a
vaulted back and turned-back pegbox. As Wright has shown, the gitterns of
the Middle Ages were very similar to the lute in form, but smaller. It is
tempting to believe, therefore, that size may have been one of the factors
defining the use of lute names in the Middle Ages, a ‘lute’ before 1300 being
generally larger than anything one would call a ‘gittern’.
Gittern names
Wright has convincingly demonstrated that these names referred to short-
lutes, generally with a vaulted back and sickle-shaped pegbox.
Citole names
In the same article Wright demonstrates that these names were customarily
applied to plectrum-plucked short-lutes, often shown with holly-leaf shaped
bodies and ‘thumb-hole’ in the neck.
Psaltery names
There are numerous thirteenth- and fourteenth-century sources in which
the Latin word psalterium and French psalterion (in various forms and
spellings) are associated with ‘pig-snout’ psalteries (1: see Figure 6), 2, 3, 5
(some manuscripts) and 10 (a large body of illustrations). Earlier manu-
scripts of source 5, the Psalterium Decem Chordarum of Joachim of Fiore, show
trapezoidal instruments with single, central soundhole.'” This seems to be
the form required by Joachim’s allegory and therefore represents one kind of
psalterium that he knew. Pig-snout instruments are not often found in
pictorial sources before 1200, whereas trapezoidal instruments are common
(Figure 14).?? This suggests that in the twelfth century, psalterium, and its
vernacular scions, was usually applied to trapezoidal instruments, but in the
thirteenth century came to denote at least two types of instrument when the
pig-snout form (whose origins are obscure) was disseminated.
147
Appendix 1
Rotta names
Steger has drawn attention to the Moissac cloister sculpture of shortly
before 1100 where a triangular harp-zither is labelled rota. Two new pieces
of evidence can now be adduced, both of which support Steger’s contention
that this was the kind of instrument which (at least in the Gothic period)
bore forms of the name rota. Source 4 (Paris manuscript) shows what is
probably a triangular zither, labelled rota. This has almost certainly been
copied from an earlier manuscript or a pattern-book and its relation to
fourteenth-century usage might be questioned on those grounds. However,
Petrus de Abano states that the rota he knew had two string-bands—a
description which fits the structure of triangular harp-zithers as revealed
with particular clarity in the romanesque sculpture of France (see Figure
LA}:
Harp names
Harp names have changed their meaning since late Antiquity. The
philological and archaeological evidence suggests that the lyres used by
some Germanic peoples during the Dark Ages bore the name harp before the
advent of the pillar-harp in the medieval West.*! The name ‘harp’ must
have been slowly grafted onto the pillar-harp, perhaps from the eighth
century on (the chronology of events here is almost completely obscure). In
some areas of Europe where the old northern lyres continued to be used into
the Gothic period the name ‘harp’ may often have been retained—as
suggested, for example, by certain Scandinavian carvings which interpret
the harpa of the Gunnarr legend as a lyre.””
The meaning of farpa in an eleventh-century text such as the Ruodlieb, for
example, therefore remains uncertain; both lyre and pillar-harp are
possible, and we can only speculate how many other possibilities there may
be. By the Gothic period, however, the evidence from France (where
plucked lyres do not appear to have been used later than the twelfth
century) points directly to the pillar-harp as the instrument denoted by Old
French harpe (the illustrations in certain manuscripts of the Tristan en prose,
for example.)
LATIN TERMS
The histories of Latin terms such as cithara and lyra in the Middle Ages are
complex and have drawn forth some very detailed research.”* Here it is only
possible to draw the outlines of these histories.
148
Appendix 1
Cithara
In Classical Latin the term cithara, a borrowing from Greek, was reserved for
various forms of lyre (and, perhaps, of harp). This is the usage transmitted,
for example, by Boethius’s De Institutione Musica, where it is clear that the
cithara is an instrument with a string for every note. Cithara could also be
used in a general sense, as revealed in Isidore of Seville’s discussion of
stringed instruments in the Etymologiae (3:22). There psalteria, lyrae and other
instruments are described as different species of cithara.
These two traditions persisted throughout the Middle Ages. It is
tempting to believe that many medieval authors who use the term cithara
(and especially poets) have no more specific meaning in mind than ‘a
stringed instrument’ (for the general sense see source 8). Yet during the
Gothic period there existed a powerful tradition which associated cithara
with the word ‘harp’, in its various forms in different vernaculars, and with
the pillar-harp. In England this tradition can be traced as early as the
eighth century and from 1200-1500 it flourished (see, for example, sources
1, 3, 6 and 7,) although it had to compete with lira (sources 2 and 4 (British
Library MS)).
Yet there are clear signs that this ‘open-string’ tradition was not the
only one which influenced the use of cithara in Medieval Latin. As early as
the fourth century there are references to citharae with four strings.”* This
seems rather a small string-total for a Late Antique lyre, so these citharae
may be fingerboard instruments.
The history of fingerboard instruments in Dark Age Europe is so obscure
that it is impossible to trace this use of cithara into the Gothic period with
any precision. What is clear, however, is that by 1200 plucked instruments
were in use in the West bearing names ultimately derived from Latin cithara
(e.g. gittern and probably also citole). With the increasing acceptance of
vernacular-derived terminology in medieval Latin the form (h)arpa became
an acceptable—although far from universal—way of referring to the pillar-
harp; as a result cithara, so often associated with harps, sometimes became
attached to gitterns and citoles whose vernacular names were so similar to
cithara. This is what seems to happen in Abano (15) where the cythera is said
to have four strings, and in source 2, where the chytara (sic) is shown as a
citole.
Lira
Although this word is found in Latin sources throughout the Middle Ages, it
does not seem to have enjoyed anything more than a sporadic existence in
Old French. The ‘open-string’ tradition which exerted such an influence
upon the use ofcithara in the Middle Ages also controlled the senses of lira to
149
Appendix I
some extent, whence the word is sometimes found in association with pillar-
harps. Two manuscripts of c1200, both illustrated in the region of Alsace,
show the lira as single-string bowed instruments (7 and 8). These two
sources are closely related.”° One of them is the celebrated Hortus Deliciarum
of Herrad of Hohenbourg and it has long been recognised that some
illustrations in this (now destroyed) manuscript show signs of Byzantine
influence. It may be no coincidence, therefore, that the bowed instruments
in these pictures are strikingly similar to the modern Pontic lira. There is
very little evidence to suggest that this use of lira was widely disseminated
during the Gothic period. The same may be said of the usages whereby lyra
(in various forms) was applied to the hurdy-gurdy and the lute; I find no
evidence that these traditions were established during our period.
Symphonia
During the Gothic period, symphonia and its vernacular offsprings was a
general name for hurdy-gurdies (regardless of size or of whether played by
one or two men).”°
Organistrum
A general name for hurdy-gurdies (regardless of size or of whether played by
one or two men) but generally confined to areas of High and Low German
speech.”’
150
Appendix 2
Selective typology of
musical references in
French narrative fiction
to 1300
The aim of this appendix is to establish a context for the many brief extracts
from long narrative works which will be presented in Appendix 3. Appendix
2 describes some of the most important genres of musical reference in Old
French fiction and is based upon the following texts:
Romances
Amadas et Ydoine
Athis et Prophilias
Attila
Beaudous
Le bel inconnu
Blancandrin et l’orgueilleuse d’ Amour
La chastelaine de Vergi
Le chevalier au lion (Yvain)
Le chevalier de la Charrete
Li chevaliers as deus espees
Claris et Laris
Cligés
La confrere d’amours
Le conte du Graal (Perceval)
Le court d’amours
Durmart le Gallois
Eledus et Serene
Eracle
151
Appendix 2
Erec et Enide
Escanor
L’estoire del Saint Graal
L’estoire de Merlin
Fergus
Floire et Blancheflor
Floriant et Florete
Galeran de Bretagne
Gautier d’Aupais
Gille de Chyn
Gligois
Guillaume d’Angleterre
Guillaume de Dole
Guillaume de Palerne
Guiron le courtois
Hunbaut
Ille de Galeron
Ipomedon
Jehan et Blonde
Joufroi de Poitiers
Kanor
Le lai d’Aristote
Le livre d’Artus
Le livre de Lancelot del Lac
La manekine
Meraugis de Portlesguez
Les mervelles de Rigomer
La mort le roy Artus
Narcisus
Partonopeu de Blois
Peliarmenus
Philomena
Piramus et Tisbé
Prose Tristan (read in Vienna 2542)
Protheslaus
Robert le Diable
Le roman d’Auberon
Le roman de Laurin
Le roman de Silence
Le roman de Thebes
Le roman de Troie
Le roman de la Violette
lo?
Appendix 2
Epics
Aiol
Aliscans
Ami et Amile
Anseis de Carthage
Anseys de Metz
Auberi le Bourgoin
Aye d’Avignon
Aymeri de Narbonne
La bataille Loquifer
Boeve de Haumtone
Brun de la Montagne
La chanson d’Aspremont
La chanson de Godin
La chanson de Guillaume
La chanson des quatre fils Aymon
Le charroi de Nimes
La chevalerie d’Ogier de Danemarche
La chevalerie Vivien
Couronnement de Louis (verse redactions)
Doon de Maience
Doon de Nanteuil
Doon de la Roche
Les enfances Guillaume
Les enfances Renier
Les enfances Vivien
L’entree d’ Espagne
Fierebras
Floovant
Florence de Rome
Folque de Candie
Garin le Loheren
Gaufrey
Gaydon
Gerbert de Mez
Girart de Roussillon
Girart de Vienne
Godefroid de Bouillon
Gormont et Isembart
Gui de Bourgogne
Hervis de Metz
Hugues Capet
Huon de Bordeaux
Jehan de Lanson
153
Appendix 2
Jourdain de Blaye
Macaire
Maugis d’Aigremont
Moniage Guillaume (verse redactions)
La mort Garin le Loherain
Les Narbonnais
Otinel
La prise de Cordres et de Sebille
Prise d’Orange (verse redactions)
Raoul de Cambrai
Le roman d’Aquin
Le roman du comte de Poitiers
Le siege de Barbastre
Tristan de Nanteuil
Voyage de Charlemagne
Yon
1 The feast
Tet Essentially a listing of the musical (and other) entertainments offered
at some courtly function,
12 usually a feast, and therefore in the hall,
1.3 and in this context often clearly signalled by some formulaic reference
to the termination of the meal, often with Quant... or Apres...
Quant les tables furent levees...
Quant les tables ostees furent...
Quant cho vint apres mangier...
Apres disner i eut...
Apres mengier...
ie The occasion of the feast is often a royal marriage (under the influence
of Brut lines 10543ff and Evec et Enide, lines 1983ff).
35 The doings of professional entertainers, especially instrumentalists
and singers, loom very large in these lists of entertainments at feasts.
Many such lists are mainly strings of instrument-names (a rhetorical
procedure acknowledged in Geoffrey de Vinsauf’s Poetria Nova
(c1200)).
However, many lists also include references to the doings of courtiers,
primarily dancing (see 2.1—13), tale-telling, and (for the men) the
playing of chivalric sports (such as fencing and jumping). In this case
154
Appendix 2
la otssiez...
la peussies oir...
1:9 The total effect of the entertainment and its music is often expressed by
the formula
grant joie/noise (de)mener
1.10 The formulae in 1.8 may introduce references ranging from a couplet
to a dozen lines. There are other formulae allowing narrators to signal
the presence of music in a single line, including:
(et) VERB et VERB (et) VERB cil jongler
Cantent et notent, vielent chil jongler Hervis de Metz 569
Cantent et harpent, vielent cil jongler — Hervis de Metz 7929
e cantent et viélent et rotent cil jugler Voyage de Charlemagne 413
e cantent et viélent et rotent cil geugler Voyage de Charlemagne 837
2 The carole
Apres disner i eut vieles,
Muses et harpes et freteles,
Qui font si douces melodies,
Plus douces ne furent otes.
Apres coururent as caroles.
Ou eut canté maintes paroles. Jehan et Blonde 4761—6
4 Singing on horseback
Boefs si en mounte le palefrei corser...
Tretot en chantaunt comence a chivacher
4.] Outside the CAROLE (2.1—13) and the SINGING PARTY (3.1-6)
the courtly amateur rarely sings except when riding (for the singing of
lais to the harpe see 5.1—9). The hero or some other important character
(generally male) sings having just mounted his horse, or
sings during the course of a journey on horseback.
His song is commonly described as a son.
He may be holding a hunting bird.
Sometimes he is alone
and sometimes with others who sing with him (in which case the genre
may blur with the SINGING PARTY, 3.1-6).
By singing, the hero expresses his fine state of mental and physical
health (in epics before the late-twelfth century)
and also his courtliness and amorousness of bearing (in romance, and
in many of the later epics).
There may also be a powerful suggestion (at least in epic) that the
singing protagonist is enjoying his last moments of light-heartedness
before some disaster, and even that his singing is an expression of a
false sense of security.
musician plays (which may or may not be filled), and narrators seem
to have been free to fill the slot as they wished.
6.8 A special sub-genre of references to courtly, amateur instrumentalists
centres upon the private musician/retainer of Charlemagne in epic
tradition (Chanson de Guillaume and Aye d’Avignon).
7 Courtly accomplishment
Many epic and romance heroes are praised for their accomplishments
and education which are often itemised in detail.
| In both Epic and Romance chivalric skills and sports (jousting,
fencing, leaping, etc.) predominate in the lists of male accomplish-
ments, and the main arts of peace are chess and draughts.
72 It is exceptionally rare for a male to be praised for an ability to sing or
to read. References to a hero having mastered the ‘vii ars’ are also very
rare, though rather more frequent in connection with women.
re) Outside of 5.1—9 (and of texts relating to Aristotle’s education of
Alexander) it is almost unknown for a male to be praised for
instrumental skills; exceptions are Florimont, which is related to the
Alexander material, the Roman de Horn and Eracle where, in both cases,
the ability to play the Aarpe is presented as a skill cultivated by the
nobility of the past.
(pe In accounts of female education and accomplishment both singing and
playing are sometimes mentioned, and perhaps the seven Liberal Arts.
159
Appendix 3
Literary references
relating to the
involvement of stringed
instruments in French and
Occitan monody
Most of the following references are taken from Old French and Old
Provencal literature, together with certain Latin writings produced in
France and Occitania during the time of the troubadours and trouveres.
Since Catalonia shared in the troubadour culture of Occitania, several Old
Catalan sources are included (items 15 and 19-21). One Old Spanish text
has been included (item 4), since it offers the fullest description of a
performing ensemble which occurs time and time again in the following
texts: fiddle and voice.!
The list is restricted to passages bearing upon the performance of lyric
repertory; and references belonging to the lai/harp complex are omitted (see
the selective typology in the previous appendix, section 5.1—10, and Chapter
8).
The ideal reference for our purposes is one which specifies that a certain
kind of song is being accompanied by a certain kind of instrument. A corpus
of such references would provide a solid basis for investigating the
performance practice of twelfth- and thirteenth-century music, but unfortu-
nately very few of the passages gathered here provide clear information of
this kind. Most of them are indecisive for one or more of the following
reasons:
1 It is impossible to establish what kinds of songs are being performed.
2 It is impossible to determine whether (a) instrumental or (b) accom-
panied-vocal performance is being described.
3 When the texts mention voices(s) and instruments(s) together they are
not sufficiently explicit in delineating relations of space and time for us
160
Appendix 3
161]
Appendix 3
Sometimes the list may be built around verbs rather than nouns (‘the one
does A, the other does B...’), as in Flamenca (item 5):
One plays the lai of Cabrefoil on the viola, and the other
the lai of Tintagoil; one sang the /ai of the Fins Amanz,
and another the Jai which Yvain composed. One brought the
arpa, another the viula; one plays the flautella, and
another whistles; one brings the giga and another
the rota; one gives out the words and another puts the music
to them.
In other words these lists are, in some measure, formulaic: they are
composed of pre-set pieces of language, the common property of poets
working within Old French epic tradition, designed to present a conven-
tional scene or action within fixed metrical constraints.’ (See Appendix
2:1.1-10.)
It is a measure of the simplicity and syntactical poverty of many narrative
references to music that punctuation is rarely a crucial issue in determining
the meaning of a passage. Indeed the only instance in this appendix where
choice of punctuation appears to be crucial is Gerbert de Montreuil’s Le
Roman de la Violette (item 33), where Gerbert offers a passage which stands
quite outside the ‘distributive list’ tradition.
Amidst the ambiguities and uncertainties certain general conclusions
may be drawn:
1 The evidence is overwhelming that the preferred instruments for
accompanying the voice in twelfth- and thirteenth-century France and
Occitania were bowed instruments. This suggests at least a measure of
standardisation and convention in performance practice: the same type
of solo instrument seems to have been used, time and time again. There
is no firm evidence for the accompaniment of any medieval courtly
monody by the bands of mixed string, wind and percussion instruments
which have figured so prominently on recordings and on the concert
platform during the last twenty years. The Old Spanish Libro de A polonio
(item 4) suggests that such fiddle-accompaniment was a highly-prized
and subtle art.
2 There is abundant evidence for the accompaniment of dance-songs. See
items 2 (the Provengal dansa); 3 (?again the dansa); 9 (again the dansa).
Item 13 presumably describes instrumental music since there is no
mention of singing.
3 There is also good evidence for the instrumental accompaniment oflove-
songs (whose identity cannot be precisely established from the texts).
See items 22 and 29.
163
Appendix 3
Anon Jaufre
Anon Canso de la Crosada
Anon Libro de Apolonio (Old Spanish)
Anon Flamenca
ND
OC
OS
D Arnaut Vidal de Castelnaudary Guilhem de la Barra
>
Troubadour lyrics
7 Peire d’Alvergne Deiosta.ls breus jorns
8 Guillem Ademar Chantan dissera si pogues
9 Senher n’Aymar, chauzes de tres baros
Ramon Llull
19 Libre de Contemplacio en Deu
20 Libre de Meravelles
21 Libre d’Evast e d’Aloma e de Blanquerna
Verse narratives
22. Anon Ami et Amile
164
Appendix 3
Music theorists
40 Elias of Salomon Scientia Artis Musice
4] Jerome of Moravia Tractatus de Musica
42 Johannes de Grocheio De Musica
Petrus de Abano
49 Expositio problematum Aristotelis
The numbers in square brackets after the narrative extracts refer to the typology of
literary material presented in Appendix 2
165
Appendix 3
Verse narratives
1 °?12c Daurel et Beton
x
Source: A.S. Kimmel, ed., A Critical Edition of the Old Provencal Epic Daurel et
Beton (Chape! Hill, 1971). See above, pp.20ff., and add the following:
See also 1941ff (where the piece sung by Daurel to Beton’s fiddling appears
to be in the style of epic rather than lyric).
2 ?cll170 Jaufre
166
Appendix 3
(1]
Several stanzas of this Old Spanish romance contain what may be the fullest
account of fiddle-accompanied singing in medieval literature. The text
leaves us in no doubt that simultaneous participation of voice and
instrument is taking place, but reveals little about the kind of music which is
being performed. As always, this is primarily a problem of terminology;
although the following passages have received much attention (from
Spitzer, Devoto and Artiles among others) ,° our knowledge of musical
terminology in Old Spanish is too fragmentary to permit much more than a
guess at the author’s meaning in most instances. Yet the text is valuable as
an indication that the author thought of accompanied singing as a subtle
and highly nuanced art. Devoto’s study considers almost all previous
literature on the subject and takes full account of the mass of scholarly
conjecture which has accumulated around these passages.
The Libro de Apolonio is ultimately based upon the Historia Apollonii Regis
Tyri of c500. It exists in two recensions, RA (c500) and RB (somewhat later).
Passages of the Latin which shed light upon the interpretation of the Old
Spanish text are given below. The most important musical episode of the
Latin romance is set at the court of King Archistrates of Pentapolis when
Apollonius is entertained by the singing and playing of the king’s daughter.
In RA the princess sings and accompanies herself:
Puella vero iussit sibi afferri liram. At ubi accedens cepit, cum nimia
dulcedine vocis cordarum sonos, melos cum voce miscebat.
The girl commanded the /ira to be brought to her. And when she
took it she mingled the melody of the string-music with her singing
with surpassing sweetness ofvoice.
Apollonius is dissatisfied with the girl’s playing and asks for the
instrument. Then he performs for the company; in the words of both RA and
RB: ‘he mingles his modulated voice in song with the strings’ (Miscetur vox
cantu modulata cordis). The author of the Libro de Apolonio probably worked
from a version of the story in which the wording ofthis episode left no doubt
about the simultaneous participation of voice and instrument: he had only
to follow it to produce a clear reference to instrumental accompaniment. Yet
this hardly detracts from the interest of the following passages. They are full
of technical terminology (none of which appears to have any precedent in
167
Appendix 3
The girl prepared herself and they made room for her; she tuned
the vihuella well to a natural accord, let fall her mantle and
was left in her gown; she began a laude, none had ever heard the
like.
Lexical problems: the reference to tuning the vihuella ... en un son natural is
tantalising (for previous conjectures as to the meaning of the phrase see
Devoto, op. cit., pp.305ff). Forms of the adjective natural occur as a
description for music twice elsewhere in the poem (427b and 495b, the
former in a reference to fiddling; see below). In each case natural forms the
rhyme-word, as here, which may suggest a relatively weak meaning for the
term (‘pleasing’, perhaps?), but some technical sense may be involved, at
least in the above stanza. A natural tuning might conceivably be an
accordatura of perfect intervals akin to the ones described by Jerome of
Moravia.
The term laude (Devoto, op. cit., pp.297ff) presents an intractable
problem. It may mean little more than ‘song’, although this sense is not
recorded in the available dictionaries of Old Spanish (such as they are). A
connection with Italian /auda (and, for that matter, with Old French Jai) is
possible.
168
Appendix 3
forms the term also appears in the Libro de Buen Amor of Juan Ruiz (where it
is also associated with the fiddle) and in the Libro de Alexandre:'°
In various places this table of forms seems to show the influence of Old
Catalan balar, ‘to dance’ (whence, presumably, the deballadas of Alexandre
2118b), and of Old Spanish bailar, ‘to dance’ (cf. Old and Middle French
balade). It is tempting to translate the debailadas of the Libro de Apolonio as
‘dance-like melodies’, or something similar. This has often been proposed
(see Devoto, loc. cit., and Spitzer, p.370).
Puntas is a term that is elsewhere used for the music ofinstruments in Old
Spanish,'! but not, perhaps, in a technical sense (compare Grocheio’s use of
punctum to denote the sections of the instrumental stantipes and ductia).'”
Both those ofhigher and of lower rank spoke of her. The girl
and the viuela accorded so well together that all those who saw
her were astonished. She then showed her skill in even more
excellent ways.
169
Appendix 3
188 He raised his face a little towards the girl. She was a
little taken aback with embarrassment. He drew the bow in even
and similar strokes, and the princess was beside herself with
UR io
170
Appendix 3
... She took a fine viola that was well tuned and went to the
market to earn money by fiddling.
She began to perform words and music that were ‘natural’ and
very sweet ....
When she had well entertained the people with her viola and had
sung very much to their taste ....
Specific details from the HISTORIA: in both RA and RB the princess plays
the lira; the former mentions herfacundia sermonis, the latter her facundia oris.
Lexical problems: the pairing of viesos and sones in 427a recalls a troubadour
usage of the twelfth century (vers e./ so, ‘words and the music’), whence the
translation offered here. See also Devoto, of. cit., pp.295ff. On natural see the
note on 178b above. In the following the musician is again Tarsiana,
daughter of Apolonio:
Specific details from the HISTORIA: at this point in the Latin Tarsiana sings
a song whose text is given in full, but there is no reference to string-playing.
Lexical problems: the puzzling term natural makes another appearance. ‘In
a natural hexachord’? It is perhaps unwise to press for a technical intepret-
ation of this sort (see note on 178b above).
[6]
5 ¢cl250 Flamenca
These are the opening lines of a remarkable passage in which the author
draws up a long list of heroes from Classical, biblical and Arthurian story,
all of whom are supposed to have figured in the songs sung by the minstrels
at the wedding of Lord Archimbaud and Flamenca. The mention of lais of
172
Appendix 3
Cabrefoil, Tintagoil and Fins Amanz is surely designed to make the scene
recede into that mysterious and remote ‘Celtic’ past where so many of the
favourite stories of the twelfth and thirteenth centuries are located, and to
give the passage a distinctively narrative, ‘French’ feel.
This long passage ends:
Is there some relation between ‘the vers of Marcabrun’ and ‘the hum of the
viula players’?
The romance of Flamenca also contains a reference to the performance of a
dansa at court by two hundred fiddlers. No mention is made ofsinging (lines
716ff).
[1]
6 1318 Arnaut Vidal de Castelnaudary: Guilhem de la Barra
Source: P.Meyer, ed., Guillaume de la Barre, SATF (Paris, 1895), line 635.
A reference to minstrels with lors bals e lors esturmens. These bals are surely
identical with the bals described in nearby Toulouse by Guillem Molinier in
the Leys d’Amors (see item 17): instrumental dance-tunes to which words
could be added to make an instrumentally accompanied dance-song.
[1]
Troubadour lyrics
7 21149-21168 Peire d’Alvergne: Deiosta.ls breus jorns
‘Audrics will know how to fiddle this vers, I think, which is by [Peire]
d’Alvergne.’
Audrics, through this song let every churl know that [Peire]
d’Alvergne claims that a man without a love-service to perform is
worth no more than a withered ear ofcorn.
The most that can be said is that one Provengal scribe of c1300 was
prepared to let a song by Peire d’Alvergne leave his desk with a call for
accompaniment contained in its final tornada.
1890), pp.114—16) who tentatively proposed that the mysterious nullet might
be an error for *viulet, ‘small fiddle’ (he compares Middle French violete,
‘fiddle’). The opening lines of this tornada might then be translated:
‘Peironet, learn to sing this song with a small fiddle pleasingly and with a
clear voice ...’. Yet Middle French violete does not appear to be recorded
before the fifteenth century (Appel’s example is taken from the early
sixteenth-century chronicle of Philippe de Vigneulles), and viulet is not
attested in Old Provengal. K. Almqvist (Poésies du Troubadour Guilhem Adémar
(Uppsala, 1951), pp.116—17) retains the MS reading, construing nullet as
the name of a second minstrel (‘Little Nobody’, as it were) who is being
addressed together with Peironet.
Elias Cairel
fl. first decades of the 13c.
According to his Vida (version of MSS AIK): Elias Cairels ... fetz se joglars
... Mal cantava e mal trobava e mal violava e peichs parlava, e ben escrivia motz e sons,
‘Elias Cairel ... became a minstrel .... He sang badly and composed badly
and fiddled badly and spoke even worse, and wrote words and melodies
well.” MS H omits the reference to the fiddle and gives a different
assessment of Elias’s abilities (Boutiére/Schutz, p.254). The idea that Elias
was a minstrel may have been developed from his tenso N’Elias Cairel, de
l’amor (see H. Jaeschke, ed., Der Trobador Elias Cairel (Berlin, 1921), p.134).
175
Appendix 3
Pons de Capdoill
Pons appears in documents up to 1220.
According to his Vida Pons sabia ben trobar e violar e cantar ‘knew well how to
compose and to fiddle and to sing’. There is a passage in a poem by Pons
which may have given rise to this reference (Max von Napolski, ed., Leben
und Werke des Trobadors Ponz de Capduoill (Halle, 1879), p.52) where the poet
tells his lady that no music of instruments, including viwlas, ‘counts for
anything compared with the solace you can bring’.
Perdigo
fl. late 12c, early 13c.
According to his Vida Perdigo saup trop be violar e trobar ‘knew very well how
to fiddle and how to write songs’. This information is probably based upon
the partimen listed here as item 9.
Only 3 of the 101 troubadours commemorated in the Vidas are said to have
been fiddlers, but as the compilers of the Vidas often had little (or no)
information beyond what they could glean from a literal-minded reading of
the poems this is not a revealing figure. A third of the troubadours with
Vidas are said to have been joglars, and since instrumental skills were
fundamental to the livelihoods of most musical minstrels it is likely that
many troubadours could play an instrument of some kind (for the case of an
aristocratic troubadour who played the viola, unknown to the Vida
compilers, see item 13).
LP ise ha20s
The Razos, probably composed at about the same time as the Vidas,
explain the circumstances in which various troubadour songs were com-
posed (or the reasons why they were composed, which often amounts to the
same thing). One of the most famous is devoted to Kalenda maya by
Raimbaut de Vaqueiras (Music example 10), a poem apparently devised to
fit the melody of an estampie played by two French players of the viola
(aquesta stampida fu facta a las notas de la stampida qe.l jo[g]lars fasion en las
violas). The Razo text neither says nor implies that Raimbaut’s song was
performed to instrumental accompaniment.
An extract from a story telling how the Virgin caused a candle to light upon
the fiddle of a minstrel who played devoutly before her image at Roc-
Amadour. The story also appears in Gautier de Coinci’s Miracles de Nostre
Dame (item 37) and in the Cantigas de Santa Maria (Cantiga 8) of Alfonso el
Sabio. A related story ofa similar miracle at Arras crops up here and there,
the most interesting instance from the musical point of view being the
anonymous Dit des Taboureurs of the thirteenth century.'”
Source: Shelagh Grier, The Otia Imperialia of Gervase of Tilbury: a critical edition
of Book III, with introduction, translation and commentary (D.Phil. Thesis,
University of York, 1981), 1, pp.232—-4.
archbishop of Arles 1191-1202, and his marriage brought him the palatium
where the events which he describes here took place. Gervase’s mother-in-
law was probably the source of the story. Gervase was appointed marshall of
Arles by the emperor Otto IV (addressed in the parentheses below), but the
date of the appointment is unknown. He probably settled there in 1190. The
flamboyant protagonist of this story, Giraldus de Cabreriis, is probably the
Catalan troubadour Guiraut de Cabreira.!® >
Gervase does not say whether Guiraut (or the ladies for whom he played the
viola) sang as he played dance-music. In view of the conclusions reached in
178
Appendix 3
Treatises on poetry
15 end 13c_ 2Jofre de Foixa: Doctrina de Compondre Dictats
This definition recalls the story of Raimbaut de Vaqueiras (item 11) and the
bals as described in the Leys d’Amors (see above, p.43). Barberino is
describing the practice of putting words to a pre-existing instrumental tune.
Ramon Llull
19 ¢l272 Ramon Llull: Libre de Contemplacio en Deu.
Source: Ramon Llull, Obres Essencials, 2 vols. (Barcelona, 1957 and 1960).
Llull spent many years at the University of Paris and must have known the
monophonic song-culture of both northern France and Occitania, together
with the largely Provengal-derived art of the Catalan troubadours.
180
Appendix 3
[In the following translation those Catalan terms whose precise meaning is
not obvious are inserted in square brackets after the English equivalent
which I have chosen for them. |
2 But, as we may now see, Lord, in our time all the art of
minstrelsy is changed, for those who apply themselves to playing
upon instruments, to dancing and to composing neither sing, nor
play their instruments, nor compose poems or songs save on the
subject of lust and the vanity of this world.
love and goodness are blessed, for they preserve the art of
minstrelsy as it was first established.
5 If Mankind could only beware of, Lord, the evil which ensues
from minstrels and composers and how their songs and instruments
are wretched and useless things, then these minstrels and
composers would not be so readily welcomed and accepted as they
are.
6 Through the instruments that the minstrels play and the new
poems which they compose and sing, through the new dances that
they devise and the things which they say, your goodness is
forgotten, Lord....
23 Since minstrels, Lord, through the art and skill which they
possess, can harmonise the music, dances [ball e les voltes] and
songs [lais] which they perform on their instruments with the music
which they imagine in their hearts, how does this wonder come
about that they do not know how to open their hearts to praise you
a)
Like the moralistic diatribes of every age, this one generalises and
exaggerates its subject. The language and tone of these passages do not
suggest that Llull is engaging with a specific musical milieu (that of Paris,
for example, or of Parma where he compiled the Libre de Contemplacié en Deu).
He twice refers to trobadors (section 5) and uses atrobar and trobar to denote
the art of composing songs (sections 2, 6 and 13), but this Provencal-derived
terminology does not establish that he is referring to such courtly or
sophisticated milieux of southern France and Catalonia as may have
fostered songs of the Old Provengal tradition during the last decades of the
thirteenth century. Llull mentions composition in these extracts because he
is concerned with the moral responsibilities of those who make music—both
of those who compose and of those who merely play. Within this scheme
trobador seems to mean little more than ‘composer’ in the most general sense.
182
Appendix 3
Llull makes little attempt to distinguish joglars from trobadors. He does not
seem interested in drawing a moral distinction between them, and while he
employs trobador in the sense ‘composer’, he also speaks of joglars ‘who
compose and sing new poems’ (joglars ... per les novelles raons que atroben e que
canten ...). He is content to bundle together all the skills of musical
entertainers—composing, singing, playing instruments, dancing—and his
references to ‘minstrels and composers [with] their songs and instruments’
are perhaps too casual to be truly informative.
20 Libre de Meravelles
Chapter 76 (p.221) tells how a cleric passed a tavern where there were many
rascals and wretches ‘who were drinking in the tavern, singing, dancing and
sounding instruments’ (los quals bevien en la taverna, e cantaven, e ballaven, e
sonaven estruments). He entered the tavern and began to dance with them,
singing a song about the Virgin Mary. Llull gives the text: a strophic song of
two stanzas (inc: A vs, dona verge santa Maria) with one tornada and
apparently without a refrain.
Verse narratives
Few of the following texts can be accurately dated and they are therefore
presented here in alphabetical, rather than chronological, order.
22 Ami et Amile
Source: P.F.Dembowski, ed., Ami et Amile (Paris, 1969), lines 2325-6. See
also above, pp.31ff and items 24 and 38.
Lubias and twelve of her knights go to hear mass at Mont St Michel with a
minstrel riding before them:
184
Appendix 3
24 Doon de Nanteuil
At a feast:
That day many a rote and vielle was tuned, and many a
changon poitevine was uttered.
The line Li uns chante, li autres note (‘one sings, another plays an instrument’ )
seems to have been a stereotyped formula; compare Flamenca (item 5) line
607, and Joufroi de Poitiers:'’
[1]
Source: A Wallenskéld, ed., Florence de Rome, 2 vols (Paris, 1907 and 1909),
2 eDal2b.
Milon, son of the king of Hungary, and one of the persecuteurs of Florence of
Rome, is entertained whilst in prison:
[elements mostly 1]
[6]
Minstrels from many countries sing and play their vieles etc.
[1]
There was great rejoicing that day in the noble palace; the valets
both sang and played there aplenty ....
[1]
The following passage has often been cited as firm evidence that the ability
to accompany oneself on an instrument was regarded as an exceptional skill
during the Middle Ages.'®
The immediate narrative context of the extract is as follows. The hero of
the romance, Gerart de Nevers, returns to Nevers disguised as a minstrel to
see how things stand now that the traitor Lisiart is installed there. Gerart is
shown into the hall and called upon to fiddle. He is wet and exhausted from
travelling and so replies that he would gladly warm himself and eat before
he performs. ‘To the devil with your hesitation!’ sneers Lisiart. Hearing this,
Gerart springs forward and begins tuning his viele, complaining that a
minstrel’s life is a hard one, for the more cold and miserable he is, the more
often his masters command him to sing and sit in a draught. In Buffum’s
edition his plaint continues as follows:
188
Appendix 3
It is this version of the text which has been cited as evidence that self-
accompaniment (or, at least, self-accompaniment with a viele) was regarded
as an advanced skill in the thirteenth century. Yet why should Gerart say
(whether to the company or himself) that he cannot sing and fiddle at the
same time? A glance at the Old French text shows that choice of
punctuation may have a decisive effect upon the meaning of this passage,
and since the two earliest manuscripts of the romance have no punctuation
at this point, we are free to experiment.” A period after apris in the second
line alters—indeed reverses—the sense:
Here the implication that it is difficult to sing and fiddle at the same time
has vanished; when Gerart says he must do something ‘which I am not at all
taught to do’ he now seems to be saying in a covert way that he is not a
minstrel but a count, forced for a moment to act as a minstrel— Chou dontje
ne sui mie apris.
189
Appendix 3
And there is one further possibility. One of the four MSS of Le Roman de la
Violette (Buffum’s MS C of the early fifteenth century) gives a text in which
the third and fourth lines of the extracts quoted above are transposed.’
Here is Buffum’s text with the transposition made:
Source: J.Misrahi, ed., Le Roman des Sept Sages (Paris, 1933), lines 697-8.
[2]
Source: E.Faral, ed., Mimes Francais du XIII’ siécle (Paris, 1910), p.101.
Ge suijugleres de viele;
Si sai de muse, et de fretele,
Et de harpe, et de chifonie,
De la gigue, de l’armonie:
Et el salteire et en la rote
Saije bien chanter une note. 29—34
Source: V.F.Koenig, ed., Les Miracles de Nostre Dame par Gautier de Coinct, 4
vols (Geneva and Lille, 1955-70).
Ma viele
Vieler vieut un biau son
De la bele
Qui seur toutes a biau non,
19]
Appendix 3
It is tempting to construe the first two lines of this poem as a call for
instrumental (specifically fiddle) accompaniment. But there is another
reason why Gautier may have wished to name the viele twice in these eight
lines. In the preface to the first book of the Miracles he proclaims:
Here, once again, is Gautier invoking his viele, and we scarcely need the
signal provided by the word lire in the first line of the extract to realise that
Gautier is drawing upon more than a millennium of Latin literary tradition
in which literate poets present themselves as string-playing bards. Whence
Statius:
Now I tune my lyre only for the singing of Aonian arms and
of the sceptre fatal to both tyrants .... (Thebaid, 1:33—4)
192
Appendix 3
One passage in the Miracles reveals how imagery such as this lay close to
the surface of Gautier’s mind. Having recounted the miracle of the candle
which descended onto a minstrel’s viele at Roc-Amadour (cf. item 12) he
emphasises the importance of inner as well as outward effort in the per-
formance of devotional music by ‘our priests, cantors, clerics and monks’:
He takes his viele again and raises his eyes towards the
statue [of the Virgin]; he sings and fiddles so well that
there is no sequence or kyrie that you would have heard more
gladly ... he began once more his song and his music . . . his
mouth sings and his heart rejoices.
After a feast:
39 =Macaire
[6]
Music theorists
40 1274 Elias of Salomon: Scientia Artis Musice
In describing the musical forms used by the Parisians of his day Grocheio
distinguishes two kinds of music made by the human voice, cantus and
cantilena, each with three sub-divisions:
CANTUS
cantus gestualis
cantus coronatus
cantus versualis
CANTILENA
rotunda vel rotundellus
196
Appendix 3
stantipes
ductia
What are these vocal forms which any good fiddler can generally perform?
To begin with the cantilena group:
CANTILENA
rotunda vel rotundellus
stantipes
ductia
Each genre in this group has a refrain (Responsorium vero est, quo omnis cantilena
incipit et terminatur). Within the group the terms rotunda and rotundellus seem
to have both a general and a precise sense: Grocheio says that some people
use these two words to denote any cantilena, whereas ‘we’ (presumably the
Parisians of his day) confine the terms rotunda and rotundellus to a refrain-
song having the same melody for its refrain as for its verses: a description
which fits the rondeau. The stantipes, to judge by its name, is an estampie.
The ductia is described as a choric dancing-song, light and brisk.
The identification of the three kinds of cantus poses a delicate problem.
CANTUS
cantus gestualis
cantus coronatus
cantus versualis
The cantus gestualis is certainly the epic narrative, or chanson de geste. But what
of the cantus coronatus and the cantus versualis? Van der Werf has proposed that
Grocheio’s account of these types is so imprecise as to be almost useless, but
this is a harsh judgement.” Let us look again at what the theorist has to say:
These definitions are clear and coherent if we think in terms of the register of
the songs involved. A survey of Grocheio’s description of song-forms
suggests that two registrations underlie what he has to say, and I shall call
them the cantilena register and the cantus register. The cantilena register is the
lower, for it is associated
(1) with the refrain (all of Grocheio’s cantilene have refrains);
(2) with dance (although Grocheio only mentions dancing with reference
to the ductia, we know from other sources that the rondeau and estampie,
Grocheio’s other cantilene, were often used for dancing, though this may
not have been Parisian practice);
(3) with the entertainment of the young (this is specified for all Grocheio’s
cantilene).
198
Appendix 3
In contrast the cantus register is higher (at least as far as it embraces lyric
rather than the narrative cantus gestualis, or epic), because it is headed by the
cantus coronatus, a kind of song which Grocheio validates in almost every
possible way:
(a) its music and poetry are excellent;
(b) it deals with delightful but demanding subject-matter (de delectabili
materia et ardua);
(c) it is normally composed by the highest in the land and
(d) it is usually performed for them;
(e) it does not have a refrain (all the cantilene do);
(f) it is not associated with young people.
There is a differentiation within the cantus register, for the cantus coronatus is
clearly superior to the cantus versualis. The cantus versualis (a) lacks the
excellence of the cantus coronatus in text and concord (?music ?rhyme),”° and
is (b) performed for the diversion of the young rather than for the
199
Appendix 3
delectation of the great. We can now see that this last comment about the
proper audience for ihe cantus versualis is a suggestive one once we have
understood Grocheio’s sense of lyric registration; in his terms, to say that a
song should be performed for the young is equivalent to saying that its
registration has a tendency to sink towards the cantilena register where every
type of lyric is performed for the young. Hence Grocheio is being perfectly
consistent—pace Van der Werf—when he says that the cantus versualis is
called a cantilena by some persons.”’
What, then, are the cantus coronatus and the cantus versualis? The cantus
coronatus has been repeatedly linked with the handful of songs in trouveére
manuscripts which are marked with a crown or with the words changon
couronnée, and with the practice of the Puis whereby the composers of
winning songs were crowned and their songs copied down with an
appropriate marking.*® Yet while there is no doubt that such prize-songs
were often called changons couronnées, this term also appears in certain texts
where the meaning ‘a prize-winning song at a Pui’ seems somewhat out of
the way. When the thirteenth-century theorist Anonymous 2 says that
musica falsa is used ‘for the sake of beauty, as in cantinellis coronatis’, it seems
unlikely that he is referring to prize-songs;”’ no doubt winning songs were
often successful because they were judged to have more musical beauty than
their competitors, but can such winning songs have been so markedly
characterised by their use of semitone adjustments that a music theorist
could hold them up as a telling example of musica ficta? Surely Anonymous
2 must be referring to some broader category?
The same may be said of two further literary references to the chancon
couronnée, both of which appear to have gone unnoticed up to now:
What, then, is a cantus versualis? 1 suggest that this is Grocheio’s term for
songs which did not aim at the highest quality (or did not achieve it), yet
which belonged above the cantilena register. In other words the distinction
between cantus coronatus and cantus versualis was not so much a formal contrast
as an aesthetic one for Grocheio: a cantus versualis was simply not as excellent
as a cantus coronatus when viewed in terms of style, content, characteristic
milieu, function and ethos—in terms of register. Small wonder, then, if
modern scholars comparing Grocheio’s examples of the cantus coronatus and
the cantus versualis have been unable to discern significant differences
between them. Is our understanding of thirteenth-century taste in love-lyric
so fine that we can reconstruct the judgements of contemporaries as to what
made a poem excellent as opposed to merely fine? Jean de le Mote’s Le
Parfait du Paon gives us some idea of what was involved in judging a song’s
claim to be a changon couronnée during the lifetime of Grocheio: one had to
watch for little faults in the use of language (faus ronmant), momentary
tautologies (redicte en sens), failures to achieve the true High-Style of the best
courtly lyric (Li mot ne sont pas haut mes il sont bien plaisant) and so on.°* These
may well be some of the criteria which Grocheio has in mind when he says
that the cantus versualis lacks the excellence of the cantus coronatus in ‘text and
concord’. The songs which he cites as examples of each genre may look more
or less alike to us, but it does not follow thereby that Grocheio had no valid
reason for distinguishing them in the way that he does.
De histrionibus
Concerning minstrels
De auditu cantilenarum
202
Appendix 3
[On human skills that involve many faculties at the same time]:
Source: A. Borgnet, ed., Beati Alberti Magni Opera Omnia, 29 (Paris, 1894),
p.023:
204
Appendix 3
An extract from a chapter entitled De choro (i.e. the liturgical choir and its
arrangement). Honorius is surely describing the performance-style of
twelfth-century ring-dances in this passage: the dancers sing and move in a
circle, hold hands one moment and clap the next, and stamp their feet—all,
apparently, to the accompaniment of instruments:
Petrus de Abano
49 1310 Petrus de Abano: Expositio problematum Aristotelis
206
Appendix 3
[Aristotle]:
[Petrus de Abano]:
209
Appendix 4
String-materials in the
Middle Ages
Why wilt thou examine every little fibre of my soul,
Spreading them out before the sun like stalks of flax to dry?
I
When we speak of the vocal ‘cords’ which produce the human voice we
acknowledge the expressive power of musical strings; there is no finer
metaphor for the part of us which is musical and which best expresses our
thoughts than the tensed filaments of a musical instrument. ‘The comparison
is a suitable one, for the musical personality of achordophone resides in its
strings: to paraphrase William Blake, they are the fibres of its soul. In terms
of timbre and articulation the difference between strings of gut, metal,
horsehair and silk is considerable; a song accompanied by (let us say) a
fiddle strung with horsehair would present a very different acoustic
experience to the listener from one performed to a fiddle strung with gut or
metal. To establish the kind of string-material associated with each
medieval instrument is therefore to bring the sound-pictures of medieval
song into sharper focus.
Little has been done to determine which medieval instruments were
strung with gut, which with metal, and so on. Yet there is an abundance of
literary evidence bearing upon this topic and waiting to be sifted. The
purpose of this appendix is therefore to sketch the outlines of medieval
stringing traditions with the aid of a modest (and mostly new) corpus of
written material. Since the craft of building and maintaining musical
instruments hardly ever commanded written record during the Middle Ages
we must rely upon passing references in a wide variety of texts for our
information.' The backbone of this appendix is formed from allusions in
Latin commentaries upon Psalm 150; the injunction Laudate eum in chordis
210
Appendix 4
Gut
Latin cordas...intestinales (20)
cordas de... intestinis (23)
intestinas ...chordas (3)
chorda...de nervo (27)
corda siccatur et tenditur...et caro hominis... (13)
Goulet
(See eee aescpp :
te finecegrellhmoe <0 is Gi aca:
:|| na
Sabonbenetna:
os DyGefede
aurredesnentslord
omar
_ agincrfiraie
Fadora peg
Vad Searle
“tet"ities
212
Appendix 4
de corio (11)
corium mortui animalis (7)
vervecum...extis (8)
Metallic
Latin cordas...metallinas (20)
chorda...de metallo (27)
aeneis...chordis (11)
corde...argentee (24)
cordule...de argento (18:3)
cordas de ere (23)
cordule de auricalco (18:3)
213
Appendix 4
Salk
Latin chordas ... sericinas (20)
Horsehair
Latin seu pellibus vel pilis equorum (25)
Gut
Although Classical Latin offers certain distinct meanings for the words
intestina (‘gut’), nervus (‘sinew’), exta (‘the larger internal organs’) and viscera
(‘entrails in general’) it seems best to translate them all by the generalising
English word ‘gut’, for it is usually impossible to establish whether any more
definite meaning is involved. Accordingly, there are many references in this
appendix which will bear no more precise translation than ‘gut string(s)’ or
something similar.
Some texts offer more precise information. Sheep-gut is mentioned as
early as the fifth century in the psalm commentary of Arnobius Junior (1)
and continues to be mentioned throughout the Middle Ages more often than
any other type of gut string (4, 12 (where the animal, which I take to be a
sheep, is called pecus rather than ovis), 17, 18:1 and 2 (plus derivatives), 21).
Ram gut is mentioned in the psalm-commentary of Bruno the Carthusian
(5) and there is one reference to string made from the dyed gut ofa wether
(8).
The terms corium and pellis pose a delicate problem. Strictly speaking both
should be translated ‘skin’ or ‘hide’ or ‘leather’, yet there are reasons for
214
Appendix 4
assuming that they refer to strings of gut. The earliest use of corium appears
in Pseudo-Haymo of Halberstadt’s commentary upon Psalm 56:9: ‘[The
Psalmist] says chitara to represent mortification [of the flesh], for in the
Passion Christ was stretched on the cross like the corium of a dead animal’
(7). This version of the common allegory that the stretched strings of
David’s instruments pre-figured the bodily sufferings of Christ reveals little
by itself and it may be that ‘leather’ or ‘hide’ is the appropriate translation
here. Yet it is noteworthy that several other psalm-commentaries which
mention strings from ‘an animal’ or ‘a dead animal’ (this phrase is
something of a formula) use intestina rather than corium:?
The equation corium= gut is reinforced by the report in item 11, the
Topographia Hibernica (1185-8) of Gerald of Wales, that the Irish play upon
‘bronze’ strings as opposed to strings made of corium. Since this observation
is embedded in a passage whose purpose is to contrast the musical talents
and customs of the Irish with those prevailing in the Anglo-Norman milieu
of Gerald’s readers, it seems likely that the coritum which (by implication)
everyone except the Irish uses is the fundamental string-material of the
Middle Ages: gut.
The term pellis (‘skin’, ‘hide’) occurs only twice: once in an anonymous set
of glosses on the psalms (14) where strings are said to be made de pellibus
mortuorum animalium, and once in a tantalising gloss in the psalm-commen-
tary (25) by the English Franciscan, Henry of Cossey, where the phrase
‘Praise him upon strings’ is glossed in this way:
Horsehair
But the interest of Henry of Cossey’s psalm-commentary does not end there
for his gloss upon Psalm 150 seems to say that horsehair strings were
‘generally’ (communiter) used on the leading art-instrument of the thirteenth
PANE
Appendix 4
century, the viella. It is surprising to find the viella associated with a string-
material which has been excluded from the tradition of Western art-music
since the Renaissance, yet I suspect that Cossey’s allusion to ‘instruments
. with the hairs of horses such as fiddles generally have’ can only be
reasonably construed as a reference to strings of horsehair. It is conceivable
that he is referring to the bow of the fiddle rather than to its strings, but that
is unlikely; there seems no reason to mention bows in a gloss to a phrase
which calls the faithful to praise God upon strings. Cossey could look back
on a millennium ofChristian tradition in which that call had often been
glossed with a reference to string-materials followed by some appropriate
allegory. It is certain that horsehair strings were employed in Wales in the
fourteenth century and it is by no means impossible that they were
sometimes used on the viella.*
Silk
There are only two references to silk strings, both of c1300. The Middle
High German romance Der Busant (26) mentions them in connection with a
spectacularly lavish fiddle which is decorated with gold and ivory, whilst the
author of the Summa Musice (20) merely lists cordas .. sericinas together with
gut and metallic strings. The passage from Der Busant strongly suggests that
strings of silk were regarded as a luxury commodity at the end of the
thirteenth century, but until more references are found it will be impossible
to establish how widely they were used.
Metallic
Only two of the references to metallic strings fail to mention specific
material (they are: cordas . . metallinas in item 20 and chorda de metallo in item
27). All of the remaining sources employ terms for which dictionaries of both
classical and medieval Latin offer distinct meanings: orichalcum (‘a form of
brass or similar alloy’), electrum (‘silver-gold alloy’), argentum (‘silver’), aes
(‘copper, bronze or brass’) and the adjective aeneus (‘bronze, or other alloy of
copper’). The variable meanings of these words even in classical Latin
(where aes may denote at least three different alloys) and the likelihood that
most of the medieval writers who employed them used them loosely, suggest
that it is not safe to lean too heavily upon dictionary definitions for words
such as these. Perhaps the most satisfactory way to deal with these
references is to sort them into two categories: ‘metal strings’ and ‘precious
metal strings’. The justification for this classification is that when a
medieval author refers to aes, for example, it is usually impossible to
establish whether he means anything more precise than ‘a yellowish looking
metal’; but it seems reasonable to suppose that words denoting the precious
216
Appendix 4
metals gold or silver will have been chosen and employed with more
discrimination.
The foundation for our knowledge of metal stringing in the medieval
instrumentarium is laid by the English Franciscan friar, Bartholomaeus
Anglicus. In his encyclopedia De Proprietatibus Rerum (‘On the Properties of
Things’), completed cl250, Bartholomaeus notes that strings of the
psalterium are made de auricalco et etiam de argento: from ‘latten’ and also from
‘silver’ (18:3). I choose the translation ‘latten’ for auricalco since this is how
the term is rendered in the vernacular translations of Bartholomaeus’s work
(18c, d, e and f).° The metal (and precious metal) stringing of the psalterium
is confirmed almost two hundred years later by the French theologian Jean
de Gerson; in his Tractatus de Canticis of cl426 Gerson records that the
psalterium has chordulas vel argenteas vel ex electro, strings of ‘silver’ or of ‘silver-
gold alloy’, while in his Collectorium super Magnificat of 1428 the strings are
said to be made ex auro vel auricalco, from ‘gold’ or ‘brass’ (‘latten’?).° No
other stringed instruments of medieval France and England are associated
with metallic stringing in our texts.
Il
Several details now fall into place which allow us to sketch the outlines of
medieval stringing traditions.
Firstly, Jean de Brie’s Le Bon Berger (28) of 1379 recommends that gut-
strings are best for ‘vielles, harpes, rothes, luthz, quiternes, rebecs, choros,
almaduries, symphonies, cytholes and other instruments that one makes to give
sound by means of the fingers and of strings’. This is a very comprehensive
remark and few fourteenth-century instruments seem to be excluded from
its range. One is missing, however: the psalterion, presumably because it was
habitually strung with metallic material as attested by Bartholomaeus
Anglicus and Jean de Gerson.’
Secondly, all the medieval evidence pertaining to the string-materials of
stringed-keyboard instruments (which surely began life as mechanised
psalteries, borrowing a good deal of psaltery technology) shows that they
were strung with metallic materials. This supports the assumption that
there was a long tradition of equipping psalteries with such strings.
The picture that emerges from the texts is therefore clear (although it can
scarcely be complete): virtually all of the art-instruments of the Middle Ages
were associated with gut stringing, although horsehair seems to have been
sometimes used on the viella (and, no doubt, on some other instruments).
Psalteries seem to have stood directly outside this tradition in that they were
associated with metal stringing.
Here it may be possible to narrow down what is meant by ‘psaltery’, since
it seems that there was at least one major family of zithers, the triangular
harp-zither, or rota (Figure 11), which was associated with gut stringing;
this tradition is mentioned in two sources widely separated in date (12 and
ZAd
Appendix 4
28). When Gerson speaks of the metallic strings of the psalterium there can
be little doubt that he means a pig-snout psaltery of a familiar Gothic type
(Figure 6),° and to judge by the evidence of illustrations where the term
psalterium (or some vernacular form of the word) is accompanied by a
drawing, the same might be said for Bartholomaeus Anglicus and Jean de
Brie. .
What of the chronology of metallic strings in the West? When were they
introduced? This appendix offers some tantalising clues to how these
questions might be answered. The earliest reference to metallic stringing on
the list is contained in Gerald of Wales’s account ofIrish musicianship (11).
This report, based on Gerald’s own experience, was written during the
period 1185-8, and it implies that the Irish were exceptional for the way
they placed metallic stringing at the centre of their instrumental traditions.
The next references to metallic strings seem to maintain this ‘Celtic’
association: two Old French romances composed in the first decades of the
thirteenth century: Galeran de Bretagne (15) and the Vulgate Merlin (16).
Galeran de Bretagne mentions a harp which is strung with ‘silver’ (argent),? but
the Vulgate Merlin goes one better and presents a harp with strings of‘thin
gold’ being played by the wizard himself in disguise. Arthur is holding a
great feast, and on the second day he, Guinevere, and all the other kings and
queens (of whom there are twelve present) wear their crowns; in the midst of
the festivities a blind minstrel enters the hall; he is beautifully dressed and
all marvel at him. As he walks amongst the tables he playsj lai breton:'°
... he had a harpe around his neck which was richly worked
all over with silver and the strings were offine gold,
and there were here and there on the harp precious
stones....
218
Appendix 4
and our two romances. The evidence, fragmentary though it is, suggests
that metallic stringing was used by Irish musicians in the twelfth century at
a time when it was still regarded elsewhere as an exceptional material to
employ (the fascinating question of how long the Irish had been using such
strings awaits the attention of a specialist in Old Irish literature). It may
then have passed into the Anglo-Norman and French realms with the Celtic
musicians and story-tellers of whose activity in the twelfth century there is
so much evidence. In these lands, however, metal stringing did not become
established until the fifteenth century, for during the period 1250-1400 the
literary evidence suggests that such strings were primarily associated with
pig-snout psalteries.'!
Gut
-Silk
~ Horsehair
Metallic —_[psalteries |
Index to string-materials
REBEC NAMES
rebecs
Gut 28
GIGUE NAMES
gige
Silk 26
LUTE NAMES
luthz
Gut 28
220
Appendix 4
GITTERN NAMES
guisterne ghytaern guitarra quiterne
CITOLE NAMES
cythole
Gut 28
PSALTERY NAMES
psalterion psalterio sautry |psalterium|
HARP NAMES
harp(e) (see also cithara and lira)
ROTTA NAMES
rotta rothes
Gut 2426
SYMPHONY-NAMES
symphonies [symphonia]
Gut 28
221
Appendix 4
Latin terminology
CITHARA
‘pillar-harp’ with varying degrees of likelihood
Metallic 24
CITHARA
‘Lyre’?
Gut 4h
LIRA
‘pillar-harp’?
Gut 17
LYRA
‘lyre’?
Gut 4
PSALTERIUM
‘rotta’?
Gut +, 6,9:
PSALTERIUM
‘psaltery (probably pig-snout)
222
Appendix 4
CHOROS
‘string-drum’ (?)
Gut 28
CHELYS
‘bre’ (2)
Gut 8
‘... praise Him upon the cithara. The cithara has wood in the
cross, it has strings from the intestines of sheep, and it has a
plectrum from the Gospels. . ..’
2 9c Anonymous glossator
cordarum herzono
vel intestinis indarmum
Source: K. Strecker, ed., Vita Sancti Christophort in MGH, Poetae Latini Medii
Aevi, V, 1 (Leipzig, 1937), p.32, line 217.
220
Appendix 4
Compare:
4b (cl440) John Lydgate, Horse, Goose and Sheep (ed. M.Degenhart,
Miunchener Beitrage zur romanischen und englischen Philologie, 19 (1900), pp.66—
7), lines 379 and 383:
‘Praise Him upon strings and the organum.’ Strings are the
dried and stretched intestines of animals which give sweet
music, and they symbolise the inner thoughts of the just,
dried with vigils and fasts, and stretched with holy
meditation, giving the sweetest music of pure conscience.
Anonymous: Jn Psalmos
227
Appendix 4
The cithara gives its sound from below, and it has strings
made from the intestines of dead animals. .
230
Appendix 4
far. For this reason people look upon her now as the
fountain of the art.
This section forms art of the treatment for a wounded man prior to the
compounding of a medicine.
251
Appendix 4
232
Appendix 4
CITHARA
‘tuning-key’
‘the upper ,
[wooden] part peera
[which] is sls]
solid’
A
parts’
‘the lower
[wooden]
part [which]
is hollow’
Source: L. Foulet, ed., Galeran de Bretagne (Paris, 1925), lines 2320f. See
above, p.96.
236
Appendix 4
In the printed edition of 1495 lute is substituted for fedele in reference 1 (see
OED sv String sb. 3a).
18f Anonymous, Van den Proprieteyten der dinghen (quoted from the edition of
Haarlem, 1485).
The story of the wolf-gut strings appears to have been proverbial. Compare,
for example, the following three texts:
An extract from questions put by the Abbot to the priors of the cells which
2o9
Appendix 4
reveals the semi-proverbial character of the notion that wolf-gut strings can
destroy strings made from the guts of sheep, or will not accord with them.
... whether anyone has been the wolf-string which does not
accord with the others in the harmonious cithara of the
monks.
The cythare are probably pillar-harps, while vielle and phiale are presumably
various kinds of fiddle (or they may be synonyms). Psalteria would then
cover psalteries, monocordium the monochord (the text is probably too early
for any more exotic meaning such as ‘clavichord’), while simphonia seu
organistrum, possibly synonyms in this author’s vocabulary, cover hurdy-
gurdies. The chorus may be a string-drum.
Source: G.Ehrissman, ed., Der Renner von Hugo von Trimburg, 4 vols (Tiib-
ingen, 1908-11), 1, lines 12441—50.
238
Appendix 4
Von tanzen
Concerning dancing
For St Gregory, the holy man, said, as I have read, that it
is better to do manual agricultural labour than to go
dancing; that kind ofplay is injurious to virtue. Whenever
a man bows with a horse’s tail over four sheep-guts let his
fingers and arms be as tired as if he had been weeding all
day.
240
Appendix 4
This MS was not used in the edition of F.H. von der Hagen, Gesammtaben-
teuer (Stuttgart and Tiibingen, 1850), 1, p.348, lines 397-411.
Da hif er ym bereyden
Mit syden seiten
Ein fedele erzuget wol
Als sie ein furst foren sal.
Der korper gezieret,
Das lijt gebrieuieret
Mit golde vnd mit gesteine
Von edelm hellfen beyne.
Hinder dem swebet ein palmat siden
Borte; sie waz an allen orten
Mit gulden borten vber leit.
Alsus die gige wart bereit.
Die nagel woren guldin.
Der gygen sag von syden fin
Gewircket wol mit bylden clar.
I include the following, which lies beyond the period covered by this book,
since it offers such full information:
262
Notes
Preface
] I have assumed throughout this book that the High Style songs of the
troubadours and trouvéres were not generally performed in ‘modal’ or any
directly comparable system of fixed rhythm, but were delivered in
something approaching the ‘declamatory’ rhythm proposed by Van der
Werf (Troubadours and Trouvéeres). Van der Werf’s proposals are beginning
to find favour amongst specialists in courtly monody (see, most recently,
Switten, Cansos). There are also strong arguments in favour of the
isosyllabic style of delivery recently defended by John Stevens (‘Grande
Chanson Courtoise’).
Introduction
] On the change from Epic to Romance and its significance see Southern,
The Making of the Middle Ages, pp.209ff; Vinaver, Rise of Romance;
Hanning, The Individual in Twelfth Century Romance. See also Morris,
Discovery of the Individual and Benton, ‘Consciousness of Self’. For an
encyclopedic guide to the presentation of manhood in Old French epic
up to 1250 see Combarieu du Gres, L idéal humain, and for the place of
music in the change see Page, ‘Music and Chivalric Fiction’.
Pope, Horn, gives the text, and on the ethos of this poem (especially its
relation to epic tradition) see Burnley, ‘Roman de Horn’. While I agree
with Burnley that Thomas takes a positive view of Horn, his claim that
Thomas ‘aimed ... at heroic biography, at presenting ... an example of
traditional feudal and epic virtue’ (p.395) is quite out of keeping with
Horn’s expert musicianship. Epic heroes never play instruments before
the later thirteenth century, and even then only in a few isolated cases.
On this question see Page, ‘Music and Chivalric Fiction’.
Pope, Horn, lines 2830-44. This passage has often been quoted,
translated and interpreted. See, for example, Handschin, ‘Estampie
und Sequenz’, 1, pp.6ff; Reaney, ‘Lais of Guillaume de Machaut’,
pp.20ff Bliss, Sir Orfeo, pp.xxix—xxx; Dobson and Harrison, Medieval
English Songs, pp.87—8 (by far the most useful and reliable discussion).
For the interest in songs and their composers see Pope, Horn, lines
2788ff.
See Klaeber, Beowulf, 2105ff (where the musician may be the Danish
king, Hrothgar), and below, Chapter 8, on the story of Apollonius of
Tyre. The earliest appearance of a harper-hero in anything which
might plausibly be called romance occurs in the Ruodlieb, a Latin verse-
tale, probably of south-German origin and dating from cl050. For the
text see Zeydel, Ruodlieb, p.110.
244
Notes
Text from Lecoy, Guillaume de Dole, lines 4120—40; music from the
Chansonnier de |’Arsenal, p.179.
] Text and music from Van der Werf, Extant Troubadour Melodies, p.\17*.
Ibid.
Ibid.
.
For the text and music see Van der Werf, Extant Troubadour Melodies,
R227.
Ibid., p.220*.
For some of the evidence see, for example, Faral, Les Jongleurs,
Appendix 3, item 91, and the testimonies of Albertus Magnus
(Borgnet, Beati Alberti Magni Opera Omnia, 8, p.748), Thomas de
Chobham (Broomfield, Summa Confessorum, p.292), Gerbert de Mon-
treuil (Buffum, Roman de la Violette, lines 1400-29; see Appendix 3:33 of
this book), and an anonymous thirteenth-century preacher quoted in
Techener, Description Raisonnée, 1, p.273.
I am assuming here that the text dates from the middle decades of the
twelfth century, as argued by Kimmel (Daurel et Beton), pp.34ff,
although it may date from significantly later (cl200, perhaps?). For
further material from this epic see Appendix 3:1.
Text and music from Van der Werf, Extant Troubadour Melodies,
pp.96*—7* (MS R).
For the relative chronology of these terms see Marshall, Linguistic and
Literary Theory, pp.864ff.
Ibid., p.889.
See, for example, the usage of Marcabrun in his famous Pax in nomine
Domini (text and music in Van der Werf, Extant Troubadour Melodies,
p.227* (line 2)). See also Marshall, Linguistic and Literary Theory,
pp.669ff.
246
Notes
20 For the surviving descort tunes see De la Cuesta, Cancons dels Trobadors,
pp.402ff (Qui la vi en ditz, by Aimeric de Peguillan), and pp.531ff (Ses
alegratge, by Guillem Augier Novella). There is some doubt about the
structure of Qui la vi en ditz; Maillard (‘Lai lyrique’, p.125, n.28a)
interprets it as two poems, the second beginning Cilh qu’es caps e guids.
De la Cuesta, op. cit., interprets it (correctly, in my view), as one poem.
24 For an edition of the full text see Riquer, Guillem de Bergueda, 2, pp.64f.
For further material on refrains in troubadour poetry (including
refrains of this type) see Gorton, ‘Arabic Words and Refrains’, and
Newcombe, ‘The Refrain in Troubadour Lyric Poetry’.
26 Field, Raimon Vidal, lines 1ff. and p.61. In what follows I paraphrase or
247
Notes
32 Ibid., p.33.
] On the northern imitation of the southern canso and other genres see
Bec, La Lyrique Frangaise, pp.44-8.
The text is from Lecoy, Guillaume de Dole, lines 4653—9, and the music
from Van der Werf, Extant Troubadour Melodies, p.78*. 1 have added
248
Notes
For a catalogue of refrains and an attempt to define the corpus see Van
den Boogaard, Rondeaux et Refrains. For further bibliography see
Chapter 2, note 2.
Ibid., line 2234. For the typology of the chanson de toile see Bec, La Lyrique
Francaise, pp.107—19, and for some examples with music, Rosenberg
and Tischler, Chanter m’estuet, items 9-11 and 108. On the use of the
refrain in this genre see Jonin, ‘Chansons de toile’.
12 Ibid., lines 3419-30. Perhaps the only point of contact between this
song and trouvere lyric is via the Tournoiement des dames of Huon d’Oisi
aa9
Notes
Ibid., p.98
Music from the Manuscrit du Roi, f.3v. For convenience I take the text
from De la Cuesta, Cangons dels Trobadors, pp.741—2. There are several
errors in De la Cuesta’s representation of the original (mensural)
notation.
The various versions and revisions of the Leys d’Amors are edited in
Gatien-Arnoult, Gay Saber; Anglade, Gay Saber, and idem, Leys d’Amors.
The earliest version is generally supposed to be the treatise published
by Gatien-Arnoult, customarily labelled the ‘A-text’, but Gatien-
Arnoult’s text does not represent the earliest recoverable form of the
Consistori’s doctrines. When his edition is compared with the manu-
script upon which it is based (Figure 3) it becomes clear that he arrived
at his text by running together the work of the main hand with the
250
Notes
Ibid.
On the changing sense of vers in the twelfth and thirteenth centuries see
Marshall, Linguistic and Literary Theory, 2, pp.865ff and 951-2.
Text and music from Van der Werf, Extant Troubadour Melodies,
pp.292*—293*. This song has been much discussed; see, for example,
McPeek, ‘Kalenda Maia’.
Chapter 5: Paris
252
Notes
Material like the following, from the unpublished Flores Psalterii (?11c)
attributed to Lietbertus of Lille, shows the way in which countless
Bible commentaries deal with instruments:
Here the literal sense has shrivelled up, but not so far that we cannot
see what it once was: the tympanum relates to ‘mortification ofthe flesh’
because the skin of a drum is like the mortified flesh of a believer; the
psalterium is equated with the ‘pleasure of praise’ because its name is
connected with psallendo, ‘praising’, and so on. For surveys of references
to musical instruments in the writings of the Church fathers see Gérold,
Péres de VEglise, especially pp.123ff, 175ff, and 180ff, McKinnon, Church
Fathers and idem, ‘Musical Instruments’.
Vernacular instrument-terminology is very rare in Bible-commen-
taries and glosses compiled before the thirteenth century (I except here
the vernacular psalm-commentary of Notker Labeo and associated
glosses). In a search of published and unpublished commentaries upon
the Psalter the earliest instance I have found is in an eleventh-century
205
Notes
On the use of Hebrew sources see Smalley, Study of the Bible, pp.241ff.,
and Dahan, ‘Interprétations juives’. The study of Hebrew did not make
an immediate impact upon the glossing of instrument-names in
Scripture, as may be judged from the psalm-commentary of one of the
earliest Hebraists, Herbert of Bosham (London, St Paul’s Cathedral,
MS 2). Herbert’s material on musical instruments is conventional and
of no special interest.
Among the friar-commentators who employed Hebrew material
mention must be made of Nicholas de Lyra, whose monumental
commentary upon the Bible contains several snippets of information
about fourteenth-century instruments (mostly in the glosses on Daniel
3:5). I have consulted the text in Oxford, Bodleian Library, MS Laud
misc. 152; see f. 71v (on metallic strings; see Appendix 4:23); ff.236r-v,
on the fistula (a shepherd’s pipe), ctthara (harp) and psalterium (psal-
tery).
Sed quia Ysidorus, ethy[mologie] libro] 3, dicit quod psalterium habet lignum
concavum unde redditur sonus a part superiori, et quod eius forma est secundum
figuram littere grece que deltha dicitur, videtur quod psalterium antiquitus non
Juenit illius forme cuius est nunc. Unde advertendum quod in instrumentis modernis
omnia genera viellarum, quarum alique plures alique pauciores habent cordas,
lignum concavum habent unde sonus redditur superius applicatum ad humerum
sinistrum. Cuius corde dum sonant tractu virgule cordate, modulacio illius soni
[MS: sonus] fit inferius tangendo cordas digitis manu [MS: manus] sinistre ut
acucius vel gravius sonent. Convenit ergo psalterium cum viella in condicionibus
duabus; una quod habuit lignum concavum unde redditur sonus superius; alia quod
modulacio eius fiebat tactu digitorum inferius. Sed in tribus erat differencia. Primo
in figura et forma, quia secundum Ysidorum formabatur ad modum littere delthe
quam esse constat huius formeA ,sed nullum inter instrumenta moderna memini me
vidisse talis forme vel similis. Secunda differencia est in numero cordarum, quia
nullum genus viellarum modernarum septenarium numerum cordarum transcendit,
set psalterium vel .x. cordarum erat vel octo. Forte erat et tercia differentia, quia
non tractu virgule sicut moderne vielle sed cum pulsu digitorum sonabat.
207
Notes
... St ipse [expositor], animo in aeterna suspenso, non possit his intrumentis uti,
vocet saltem cum Elizaeo propheta, ut divinas revelationes devotius suscipiat, et
totus in Deum efferatur....Nam licet non oportet propter scientiam scripturae
quod habeat theologus usum cantus et~instrumentorum et aliarum rerum
musicalium, tamen debet scire rationem omnium istorum....
(...if the commentator, his mind set upon eternal things, cannot play
these instruments, he may at least pray with the prophet Elijah that he
may devoutly receive divine revelations and that all may be brought
forth by God. .. . Even if it is not actually necessary that the theologian,
pursuing his studies of Scripture, should have any ability with song,
instruments and other musical matters, he must nonetheless know the
ratio of all these things... .)
10 Borgnet, Beati Alberti Magni Opera Omnia, 7, p.4. For other references to
the viella see p.30 (upon the necessity of fiddling and other arts to the
proper running of the state—a view which Johannes de Grocheio
elaborates to produce his notably original De Musica), p.165a, and
Volume 8, p.748. See also Stadler, Alberti Magni de Animalibus, 2,
p.1293. Crombie (Robert Grosseteste, p.189) says that Albertus Magnus
came under the influence of Grosseteste, first lecturer to the Francis-
cans at Oxford from 1229/30, later bishop of Lincoln, and a key figure
in the history of experimental science in the West. There are striking
references to contemporary instruments in the writings of several other
friars connected with Grosseteste’s ‘circle’, suggesting that, as in the
case of Albertus, the friars’ alertness to the properties of instruments
often went hand in hand with an interest in Aristotelian science. One of
these friars is the Franciscan Bartholomaeus Anglicus, whose cele-
brated encyclopaedia, De Proprietatibus Rerum (c1250), records that the
strings of the psalterium ‘are best made from “‘latten” and also from
“silver”? (Appendix 4:18). Another of these friars is the Oxford
Franciscan Roger Bacon, mentioned in the previous note, who seems to
have become a member of Grosseteste’s ‘circle’ by 1249.
Summa Musice may not reveal anything about instruments and notation.
The most confusing and tantalising passage, however, is undoubt-
edly the one by Anonymous 4 (Reckow, Anonymous 4, 1, p.40) which
records that simplicia puncta (principally @ and @) are used ‘in various
kinds of music-books [and] in all and every kind of musical instrument’
(in quolibet genere omnium instrumentorum). 1 can find no secure or
convincing interpretation of this very important passage.
20 Ibid.
ai Ibid., pp.90-1.
32 Ibid.
4] Tbid., p.cii.
258
Notes
For the most part the abundant references to coreae in the Vices and
Virtues literature refer to dances ‘performed through the streets and
roadways’ (per vicos et plateas) rather than to the courtly occasions so
often described in romances (see Sahlin, La carole médiévale, for a very
generous selection of romance references).
British Library, MS Arundel 395, f.58v: quando plures sunt in chorea vel
quando plures spectantes tanto est letior....
260
Notes
Peyraut, f.27r.
Loc. cit.
Most references are ambiguous in some way. See, for example, Lecoy,
Guillaume de Dole, lines 503ff; are the vieleors in line 503 accompanying
the caroles or do they provide a separate entertainment? See also Sahlin,
La carole médiévale, pp.18—21 (where almost all of the texts cited with
reference to instrumental participation date from the fourteenth
century).
261
Notes
Loc. cit.: a magistris et studentibus circa sonos coronatur. It is far from clear
what is meant by circa sonos coronatur (literally ‘[the cantus coronatus] is
garlanded around its notes [by Masters and students]’). These words
have been interpreted to mean ‘accompanied (coronatur) with other
sounds (i.e. instrumentally)’; see Seay, Johannes de Grocheo [sic], p.16. I
find this a somewhat forced interpretation but I have none better to
offer.
Text and music from Marrocco and Sandon, Medieval Music, p.62.
Ibid., p.XLVI.
Ibid., p. XVI.
] Pope, Horn, lines 2830—44. This text has already been fully quoted and
translated above, pp.4—5; it is repeated here for convenience since it is
the centrepiece of this chapter.
262
Notes
The passage has often been quoted, translated and interpreted. See, for
example, Handschin, ‘Estampie und Sequenz’, 1, pp.6ff; Reaney, ‘Lais
of Guillaume de Machaut’, pp.20ff Bliss, Sir Orfeo, pp.xxix—xxx;
Dobson and Harrison, Medieval English Songs, pp.87-8.
Text and music from Jeanroy et al., Lais et Descorts, pp.62—3 and 142.
Ibid., £.222v.
For an account of the form of these Jais see Fotitch and Steiner, Tristan
en prose, pp.137ff.
Text from Goldschmidt, Sone von Nausay, pp.412ff, lines 15963—76 and
15979-91.
Tolkien, Tree and Leaf, passim, where the image is developed with
characteristic insight and charm.
For the history of the Apollonius story see Kortekaas, Historia A pollonii,
263
Notes
pp.9f, and for an important account of the story’s influence upon the
emergence of romance in twelfth-century France see Delbouille,
‘Apollonius de Tyr’. The Old French prose-version of the Historia is
available in Lewis, ‘Apollonius Romans’.
22 For Celtic analogues of the Tristan legend see Schoepperle, Tristan, and
most recently, Padel, ‘Cornish background’ and de Mandach, “Tristan
et Iseut’. Padel and de Mandach present very detailed arguments in
favour of aCornish origin for many elements of the Tristan story. As far
as Tristan’s musical accomplishments are concerned, most research
has focused upon the ‘Harp and Rote’ episode. See Newstead, “The
Harp and the Rote’ and Cluzel, ‘Harpeur d’Irlande’. For a startlingly
different view of the history of the legend see Gallais, Tristan et Iseut,
where it is argued that many elements of the story are Persian in origin.
1 Ruini, Ameri Practica Artis Musice, p.79. For further references to the
tuning of open-string instruments in the theorists of this period see
Reaney, MS Oxford Bodley 842, p.21 (a diagram of a cythera per modum
boycit, that is, a diagram of Boethius’s fifteen-note gamut imposed upon
a drawing of a medieval pillar-harp); Anon, Summa Musice in Gerbert,
Scriptores, 3, pp.214—15.
For the text see Hardy and Martin, Lestorie des Engleis, p.351. For a text
of the Gesta with facing English translation see Miller and Sweeting, De
Gestis Herewardi Saxonis.
Ibid., f.416v.
Ibid., f.485r.
Loe. cit.
The /ais of the Vienna manuscript are edited in Fotitch and Steiner,
Tristan en prose.
Ibid., pp.150-1.
Ibid., £.113v.
26 The exact meaning of essais in this context is unknown. See Stone and
Rothwell, Anglo-Norman Dictionary, sv assai, ‘attempt’. The basic sense is
presumably something like ‘the testers’, perhaps referring to certain
strings which occupied a central position in any tuning procedure or in
some such tuning prelude as is described in the Roman de Horn.
27 For the relevant section of the text in a modern edition see Bossuat,
Anticlaudianus, IV, lines 345ff.
266
Notes
Particula 19, problem 3, and on the archaic character of the rota strung
on both sides, problem 39.
For the diagram and a translation of the text see Chapman, Marin
Mersenne, pp.224—6.
See, for example, Panum, Stringed Instruments, figures 119, 123 and 133.
See Page, ‘False Voices’, for the kind of semantic and terminological
difficulties which stand between us and a satisfactory solution of the
question.
Ibid., plate 2.
Ibid., p.81.
For Old Provencal evidence see Wright, ‘Gargilesse’, p.67 and n.3, on
the illustrations in manuscripts of the Vidas which show the troubadour
Perdigo said to be skilled with the viola. For Old Spanish evidence see
Appendix 3:4.
For examples see Bachmann, Origins, plates 55 and 60, and Panum,
Stringed Instruments, figures 124 and 125.
Ibid.
Ibid.
270
Notes
] See further Chapter 3 and the evidence from Jean Renart’s romance of
Guillaume de Dole.
Most readers will no doubt find that items which they would have
expected to be included in this list are not in fact present. Choosing the
material for inclusion has been a difficult—and no doubt somewhat
arbitrary—process. As far as the music-theorists and writers on
vernacular poetry are concerned I have aimed at comprehensive
coverage. The material from theological treatises towards the end of the
appendix, however, is only a minute sampling of a vast quantity of
material still waiting to be gathered and assessed. In the case of
narrative poems and prose-texts I have not hesitated to include
references which, even if they reveal very little by themselves, none the
less illustrate some of the formulaic patterns which underlie the
presentation of musical life in medieval fiction. I have however omitted
references to the performance of /ais to the harp; most of these are quoted
and translated in Chapter 6.
Since the remains of narrative and didactic literature in Old
Provengal are so sparse it is worth extending the above remarks to
record here the musical references which I encountered in Occitan
sources but did not include in the appendix.
Arnaut Guilhem de Marsan Ensenhamen au Cavayer, 131ff.
Blandin de Cornualha (ed. Horst), pp.115 and 990ff.
Canso de la Crosada (ed. Martin-Chabot): II, pp.52, 118, 200, 284 and
298; III, pp.182 and 300.
Garin lo Brun Ensenhamen a la Domna, 525ff, 539ff.
Guilhem Anelier Histoire de la Guerre de Navarre, 27\ff.
Jaufre 3073ff 4485ff and 10786ff.
Matfre Ermengaud Le Breviari D’Amor (ed. Azais), 2, pp.97ff (also
Ricketts, ed., Le Breviari d’Amor, 27924ff and 33994ff).
Raimon Vidal So fo e.l temps, 1097ff.
La Vida del Glorios Sant Frances (ed. Arthur), p.168
Vie de S.Auzias (ed. Campbell), p.96.
I have also omitted: (1) the celebrated Supplication of Guiraut Riquier.
Despite its many references to troubadours and instrumentalists I do not
find that it reveals anything about accompaniment. (2) The satirical
poems against joglars. (3) The Cantigas d’Escarho e de Mal Dizer in
Galician- Portuguese and (4) the writings of the Catalan chroniclers such
as Muntaner. These sources all contain many interesting references, but
they are either somewhat late (4), peripheral (3), or do not reveal
anything about instrumental practice upon closer inspection (2).
2 Arnold, Brut, lines 10545—8. Further on the technique of these lists see
aN
Notes
Ibid.
See, for example, Criado de Val and Naylor, Libro de Buen Amor, p.378
(all MSS) at 1228b.
The historical elements in the Vidas have been much discussed. See
Egan, ‘Vida’; Bec, Anthologie de la Prose Occitane, 1, pp.20ff; Wilson,
‘Literary commentary’, and De Ley, ‘Provengal biographical tra-
dition’.
For the text of the Dit des Taboureurs see Jubinal, Jongleurs et Trouvéres,
pp.164—9.
35 The part of Peter’s text dealing with minstrels is printed at the foot of
Chobham’s discussion of the issue in Broomfield’s edition.
36 £.37v.
273
Notes
Page, ‘Early fifteenth century instruments’, pp.341 and 346 and idem,
the Direction of the Beginning’, pp.121—2.
See Appendix 1.
brow’ or at least informal ethos. The line of descent here seems to run
from the medieval gittern—an instrument habitually associated with
foppish young men and tavern brawlers during the fourteenth cen-
tury—to take in the Renaissance cittern and, eventually, the man-
doline. On the medieval evidence for the ethos of the gittern, see
Wright, ‘Medieval Gittern and Citole’.
219
Bibliography
Abbreviations
AIM American Institute of Musicology
AM Acta Musicologica
EM Early Music
FM Forum Musicologicum
MD Musica Disciplina
MQ Musical Quarterly
PL Patrologia Latina
Psalm-commentaries
MS Bodley 251
Nicholas de Lyra
MS Bodley 318
Lietbertus of Lille
ie
Bibliography
MS Bodley 727
Johannes Halgrinus de Abbatisvilla
MS Bodley 737
Pseudo-Haymo of Halberstadt
MS Bodley 738
Nicholas Trevet (with coloured drawings)
MS Bodley 860
Anonymous glosses
MS e Mus. 15
Grosseteste (with sketches, f. 43r)
MS Hatton 37
Anonymous glosses
278
Bibliography
Sermons
Many sermons by major medieval authors are available in early printed
editions. See bibliography, passim.
MS Royal 11 B.iu
Anonymous
MS Lyell 12
Guillaume Peyraut
y, Summa de Vitiis et Virtutibus
MS Bodley 251
Guillaume d’Auvergne De Viciis et Peccatis
MS Bodley 864
Anonymous Nota de choreis (ff.146v—147v)
MS Harley 4903
Peliarmenus
Kanor
f.fr. 1446
Kanor
f.fr. 22549—50
Peliarmenus
Kanor
f.fr. 1374
Le Roman de la Violette
fir 1553
Le Roman de la Violette
Miscellaneous
London, British Library MS Royal 8 G.iv
William of Auxerre, Summa Aurea
Additional MS 18322
Humbert of Priallaco, commentary upon Peter Lombard, Sentences,
Book IV
MS Bodley 749
Bartholomaeus Anglicus De Proprietatibus Rerum
280
Bibliography
Anderson, G.A. Notre Dame and Related Conductus: Opera Omnia. Institute of
Medieval Music, Collected Works X (Henryville, 1978), including volume
8, Ipt Conductus—the Latin Rondeau Répertoire.
Arnold, I. (ed.) Le Roman de Brut de Wace, 2 vols, SATF (Paris, 1938 and
1940).
Artiles, J. El ‘Libro de Apolonio’, poema espanol del siglo XIII (Madrid, 1976).
Barassi, E.F. Strumenti musicali e testimonianze teoriche nel Medio evo (Cremona,
1979).
282
Bibliography
—‘Les plus anciennes danses francaises’, Revue des Deux Mondes, 31:1
(1906), pp.393—424.
Benson, R.L. and Constable, G. (eds) Renaissance and Renewal in the Twelfth
Century (Oxford, 1982).
Borgnet, A. (ed.) Beati Alberti Magni... Opera Omnia, 38 vols (Paris, 1890-9).
Boutiére, J. and Schutz, A.H. (eds) Biographies des Troubadours, 2nd ed.
(Panis; 1973).
Brewer, J.S. (ed.) Fr. Rogeri Bacon Opera... Inedita, Rolls Series 15 (London,
1859).
Bridges, J.H. (ed.) The Opus Majus of Roger Bacon, 3 vols (Oxford, 1897—
1900).
Brown, H.M. Sixteenth Century Instrumentation: The Music for the Florentine
Intermedii, MSD 30 (AIM, 1973).
284
Bibliography
Burney, C. A General History of Music, 4 vols (London, 1776, 1782 and 1789).
Burnley, J.D. ““The ‘Roman de Horn” and its Ethos’, French Studies, 32
(1978), pp.385—97.
Carey, R.J. (ed.) Jean de Le Mote: Le parfait du paon (Chapel Hill, 1972).
Casimiri, A. “Un trattatello per organisti di anonimo del sec. XIV’, Note
d’Archivio per la storia musicale, 19 (1942), pp.99-101.
Chiarini, G. (ed.) Juan Ruiz... Libro de Buen Amor (Milan and Naples, 1964).
Combarieu du Gres, M. de. L’idéal humain et l’expérience morale chez les héros des
chansons de geste des origines 4 1250, 2 vols (Aix-en-Provence, 1979).
Cowper, F.A.G. (ed.) Jile et Galeron par Gautier d’Arras (Paris, 1956).
Crombie, A.C. Robert Grosseteste and the origins of experimental science, 1100-1700
(Oxford, 1953).
Curtis, R.L. (ed.) Le Roman de Tristan en Prose, 2 vols (Munich, 1963 and
Leiden, 1976), in progress.
d’Avray, D.L. The Preaching of the Friars: Sermons Diffused from Paris before 1300
286
Bibliography
(Oxford, 1985).
Dimock, J.F. and Brewer J.S. (eds) Giraldi Cambrensis Opera, Rolls Series 21
(London, 1861-81), including: Jtinerarium Kambriae (21:6) and Topographia
Hibernica (21:5).
Dobson, E.J. and Harrison, F.L1 (eds) Medieval English Songs (London,
1979)
Doss-Quinby, E. Les refrains chez les trouvéres du XII’ siécle au début du XIV’
(New York, etc., 1984).
Egan, M. ‘Commentary, vita poetae, and vida. Latin and Old Provengal
“Lives of Poets” ’, Romance Philology, 37 (1983), pp.36—48.
Faral, E. Les Jongleurs en France au Moyen Age, 2nd ed. (Paris, 1971).
Fay, P.B. and Grigsby, J.L. (eds) Joufroi de Poitiers (Geneva and Paris,
1972).
288
Bibliography
Field, W.H.W. (ed.) Raimon Vidal: Poetry and Prose. 2, Abril Issia (Chapel Hill,
1971).
Flutre, L.-F. Table des noms propres...figurant dans les romans du moyen age
(Poitiers, 1962).
Fotitch, T. and Steiner, R. (eds) Les lais du roman de Tristan en prose d’apres le
manuscrit de Vienne 2542 (Munich, 1974).
Frank, I., Répertoire Métrique de la poésie des troubadours, 2 vols (Paris, 1953).
Gallais, P. Genése du Roman Occidental: Essais sur Tristan et Iseut et son Modele
Persan (Paris, 1974).
Gallo, E. The Poetria Nova and its Sources in Early Rhetorical Doctrine (Paris,
1971).
Gallo, F.A. (ed.) Johannis Boen Ars Musicae, CSM 19 (Rome: AIM 1972).
289
Bibliography
Galpin, F.W. Old English Instruments of Music: their History and Character, 4th
ed., revised T. Dart (London, 1965).
Gatien-Arnoult, A-F. (ed.) Las Flors del Gay Saber, estiers dichas Las Leys
d’Amors. Monuments de la Littérature Romane, 1-3 (Toulouse, 1841-3).
Gibson, M. (ed.) Boethius: His Life, Thought and Influence (Oxford, 1981).
Gonfroy, G. “Le reflet de la canso dans le De Vulgari Eloquentia et dans les Leys
d’Amors’, CCM, 25 (1982), pp.187—96.
Grier, S. (ed.) The Otia Imperialia of Gervase of Tilbury: a critical edition of Book
III, with introduction, translation and commentary (unpublished doctoral thesis,
University of York, 1981).
Griscom,A. and Jones, R.E. (eds) The Historia Regum Britanniae of Geoffrey of
Monmouth (London, etc., 1929).
— ‘Two Central Places: Paris and the French Court in the Early
Fourteenth Century’, in Kuhn, H. and Nitsche, P. (eds) Bericht iiber den
Internationalen Musikwissenschaftlichen Kongress Berlin 1974 (Kassel, etc., 1980),
pp. 135-57.
Hanning, R.W. The individual in twelfth century romance (New Haven and
London, 1977).
Hardy, T.D. and Martin, C.T. (eds) Lestorie des Engleis, 2 vols, Rolls Series
91 (London, 1888-9). Contains Gesta Herewardi.
Haynes, S. ‘Music and genre in the alba’, in Proceedings of the Third British
Conference of Medieval Occitan Language and Literature (University of Warwick,
1-3 April, 1985). Forthcoming.
291
Bibliography
292
Bibliography
Hume, K. ‘Why Chaucer calls the Franklin’s Tale a “Breton lai””’, Philological
Quarterly, 51 (1972), pp.365—79.
Hunt, R.W. (ed. and revised M. Gibson) The Schools and the Cloister: The Life
and Writings of Alexander Nequam (1157-1217) (Oxford, 1984).
(ed.) Das Cantuagium des Heinrich Eger von Kalkar 1528-1408 (Cologne,
1937).
Jaeger, C.S. The Origins of Courtliness: Civilizing Trends and the Formation of
Courtly Ideals 939-1210 (Philadelphia, 1985).
Jeanroy, A. La poésie lyrique des troubadours, 2 vols (Toulouse and Paris, 1934).
Jeanroy, A., Brandin, L. and Aubry, P. (eds) Lais et Descorts Francais du XHT’
siecle (Paris, 1901).
Jones, E. Musical and Poetical Relicks of the Welsh Bards, 4th ed. (London,
1825).
(ed.) Nouveau Recueil des Contes, Dits, Fabliaux, 2 vols (Paris, 1839-42).
293
Bibliography
Kastner, H. Harfe und Schwert: Der hifische Spielmann bei Gottfried von Strassburg
(Tubingen, 1981). ;
Kimmel, A.S. (ed.) A Critical Edition of the Old Provencal Epic Daurel et Beton
(Chapel Hill, 1971).
Koenig, V.F. (ed.) Les Miracles de Nostre Dame de Gautier de Coinci, 4 vols
(Geneva and Lille, 1955-70).
294
Bibliography
Langlois, Ch. Table des noms propres ... dans les chansons de geste (Paris, 1904).
Leo, F. (ed.) Venanti Honori Clementiani Fortunati Presbyteri Italici Opera Poetica,
MGH, Auctorum Antiquissimorum, 2, Tomi IV Pars Prior (Berlin,
1881).
295
Bibliography
Little, Lester K. Religious Poverty and the Profit Economy in Medieval Europe
(London, 1978).
Loomis, R.S. Arthurian Literature in the Middle Ages, corrected edition (Oxford,
1961).
296
Bibliography
—— The Razos de Trobar of Raimon Vidal and associated texts (London, 1972).
Mayr-Harting, H.M.R.E. The Venerable Bede, the Rule of St Benedict and Social
Class (Jarrow Lecture, 1976).
Miller, S.H. (ed.) and Sweeting, W.D. (trans.) ‘De Gestis Herewardi
Saxonis’, with Fenland Notes and Queries, 3 (1895-7).
Morton, A.L. “The Matter of Britain: The Arthurian Cycle and the
Development of Feudal Society’, Zeitschrift fur Anglistik und Amerikanistik, 8
(1960), pp.1—24.
Mundy, J.H. ‘Urban Society and Culture: Toulouse and its Region’, in
Benson and Constable, pp.229—47.
—— ‘The Harp and the Rote: An Episode in the Tristan Legend and its
Literary History’, Romance Philology, 22 (1969), pp.463—70.
298
Bibliography
Noiriel, G. ‘La chevalerie dans la geste des Lorrains’, Annales de VEst, 5th
Series (1976), 3, pp.167—96.
— ‘The fifteenth century lute: new and neglected sources’, EM, 9 (1981),
pp. 11-21.
Pattison, W.T. The Life and Works of the Troubadour Raimbaut d’Orange
(Minneapolis, 1952).
Perrot,J. (trans. N. Deane) The Organ: from its invention in Hellenistic Times to
the end of the thirteenth century (Oxford, 1970).
300
Bibliography
Pertz, G.H. (ed.) Ekkehard: Casus Sancti Galli, MGH, Scriptorum, 2 (Han-
nover, 1829).
Piper, P. (ed.) Die Schriften Notkers und seiner Schule, 3 vols (Freiburg and
Tubingen, 1882-3).
Pope, Mildred (ed.) The Romance of Horn by Thomas, 2 vols (Oxford, 1955 and
1964).
Randel, D.M. ‘Al-Farabi and the Role of Arabic Music Theory in the Latin
Middle Ages’, JAMS, 29 (1976), pp.173—88.
Ritson, J., Ancient Songs from the time of King Henry the Third to the Revolution
(London, 1790).
Robinson, F. (ed.) The Works of Geoffrey Chaucer, 2nd ed. (London, 1966).
Rosenberg, S.N. and Tischler, H. (eds) Chanter m’estuet: Songs of the Trouvéres
(London, 1981).
Rouse, R.H. and Rouse, M.A. Preachers, Florilegia and Sermons: Studies on the
‘Manipulus Florum’ of Thomas of Ireland (Toronto, 1979).
Secreta Secretorum. Cited from the edition published at Anvers, Mattheus van
der Goes, c1488.
Schoepperle, G. Tristan and Isolt: A Study of the Sources of the Romance, 2 vols
(Frankfurt and London, 1913). New ed., 1960.
Smalley, B. The Study of the Bible in the Middle Ages, 2nd ed. (Oxford, 1952).
Sommer, H.O. (ed.) The Vulgate Version of the Arthurian Romances, 8 vols
(Washington, 1909-16).
Switten, M.L. The Cansos of Raimon de Miraval: a Study of Poems and Melodies
(Cambridge, Mass., 1985).
Val, Manuel Criado de and Naylor, E.W. (eds) Arcipreste de Hita: Libro de
Buen Amor (Madrid, 1965).
Van den Boogaard, N., Rondeaux et refrains du XII’ siecle au début du XIV’
(Paris, 1969).
Van der Veen, J. ‘Les aspects musicaux des chansons de geste’, Neophilolo-
gus, 41 (1957), pp.82—100.
Van der Werf, H. The Chansons of the Troubadours and Trouvéres (Utrecht,
1972).
—— Organistrum in HMT.
Walsh, K. and Wood, D. The Bible in the Medieval World (Oxford, 1985).
306
Bibliography
White, J.E. Jr (ed.) La Conquéte de Constantinople (New York, 1968).
Wright, R. Late Latin and Early Romance in Spain and Carolingian France
(Liverpool, 1982).
307
Index
Ab joi mou lo vers e.l comens (Bernart de Arnaut de Maroill (Maruelh), 12—16,
Ventadorn), x 26—28
Abril issia (Raimon Vidal), 12, 25-28, Les grans beutatz e.ls fis ensenhamens, 12—14
247-8 Arnaut Guilhem de Marsan Ensenhamen
Adémar II of Poitiers, 22, 23 au Cavayer, 271
Ademar, see Guillem Ademar Arnaut Vidal de Castelnaudary Guilhem
Aegidius of Zamora Ars Musica, 236 de la Barra, 164, 173, 184
Aimeric de Peguillan Qui la vi en ditz, 247 Arnobius Junior Jn Psalmos, 214, 219,
Alain de Lille Anticlaudianus, 119-120, 223,218
266, 281 Ars Cantus Mensurabilis (Franco of
De Planctu Naturae, 143 Cologne), 57
Alba, 41 Ars Musica (Aegidius of Zamora), 236
Albertus Magnus, 57—58, 73, 246, 255, Ars Musicae (Johannis Boen), 260
258 Arthurian narratives, 3, 19, 22, 96-107,
Commentary upon the Sentences of Peter 121-2
Lombard, 82-83, 165, 204—5, 259-60 Au repairer que je fis Prouvence, 198
De Animalibus, 237 Au tens nouvel (Perrin d’ Agincourt), 33
Albumasar, 143 Avicenna (on scientia of fiddling), 57-8
Alfonso el Sabio, 177, 184 Aye d’Avignon, 159
Alfonso of Aragon, 117-18
Amarcius: see Sextus Amarcius Bacon, Roger, 254—5
‘Amateur’ string-playing, 4—8, 20, 92— Bal, 43—48, 173, 179, 184, 252
107, 158-9, 166-71, 189 Bamberg Manuscript, 75—6, 112—3, 258
Amerus Practica Artis Musice 62, 111-13, Bartholomaeus Anglicus De
117, 123, 144-5, 256, 264 Proprietatibus Rerum, 211, 217—9, 235—
Ami et Amile, 165, 184 6, 2559280
Amors, m’ard com fuoc amb flama, 247 Bate, Henri, see Henri de Malines
Amundesham (John) Annales Monasterii Bele Aelis, 245, 249
Sancti Albani, 237-8 Bele Aiglentine, 35-7, 250
Annales Monasterii Sancti Albani (John Bele m’est la voiz altane (Daude de
Amundesham), 237 Pradas), 31—2
Anonymous glossator, 219, 223 Benoit de Sainte Maure Roman de Troie,
Anonymous, 2, 200 29
Anonymous, 4, 256—7, 269 Ben volgra, s’esser pogués, 247
Anticlaudianus, see Alain de Lille Ben volgra que.m vengues mer [cles
Apollonius of Tyre, see Historia Apollonii (Blacasset), 17
regis Tyre & Libro de Apolonio Beowulf, 5, 244
Aristotle, 57-61, 73, 123, 133, 159, 165, Bernart de Ventadorn, 32—33, 179, 184
206-9, 258 Ab joi mou lo vers e.l comens, x
Armand de Belvézer Sermones, 237 Bertran II, Count of Forcalquier, 7
308
Index
Blacasset, Ben volgra que.m vengues mer(cles Chanter m’estuet, 198, 249
17 Chastelain de Couci, 34, 200
Blake, William, 210 Chelys, 121, 223, 226
Blandin de Cornualha, 271 Chi encor querez amoretes, 203
Boethius De Institutione Musica, 149 Choros, 59, 60, 78, 217, 223, 242
Boeve de Haumtone, 157 Chorus, 229, 230, 238
Boncompagno da Signa Rhetorica Chrestien de Troyes Erec et Enide, 154,
Antiqua, 164, 179 187
Bourdon/Bordunus/Bordone/Burdun, Cilh qu’es caps e guids, 247
64, 69-71, 118-120, 125, 127-131, Cithara/cythara/cythera, 53-5, 112,
158, 266 119-20, 123, 144, 149, 204-6, 215,
Bruno the Carthusian Jn Psalmos, 210— 222-3, 228-38, 253-4
12, 214, 219, 224-5, 278 Citoles/citola/chistolles/cistolla/cythole,
Brut: see Wace 6, 54, 56, 59, 132-3, 146, 149, 166,
Burney, Charles A General History of 217, 221, 242, 274-5
Music, x Claris et Laris, 84, 137, 157, 165, 185, 261
Collatio de coreis, 77, 259
Canso, 6, 15, 22—4, 32, 166-7, 172 Collectorium super Magnificat (Jean de
Canso de la Crosada, 164—7, 271 Gerson), 217
Cantigas d’Escarhno e de Mal Dizer, 271 Compendi de la coneixenga del vicis en els
Cantigas de Santa Maria, 126, 177, 184 dictats del Gai Saber (Joan de
Cantilena, 67-8, 81, 196-201, 261 Castellnou), 164, 180, 251
Cantus coronatus, 67—8, 196-201 Complainte Douteuse, 200
Cantus versualis, 67—8, 196—201 Conflictus Ovis et Lini (Wenrico), 219, 224
Cantus gestualis, see ‘Chanson de Conductus/conduis, 50, 68, 73, 77, 85-91,
Geste’ 135, 261-2
Carole/corea/chorea, 53, 60, 77-84, 89— Conon de Béthune, 8
91, 135, 155-6, 203-6, 259-62 La Conqueste de Constantinople, 245
Ce fut en mai (Moniot d’Arras), 135 Corbechon (Jean) La proprietaire des
Cele d’Oisseri, 37-8 Choses, 236
Cerveri de Girona, 247 Corea/chorea: see Carole
C'est la gieus, en mi les prés, 14—15, 80 Counterpoint, 72—4
C’est la jus la praele, 34—5, 83—4, 88 Courtliness
C'est tot la gieus enmi les prez, 80 and musical skills, 5-8
Chanconete/chanconnette, 35, 86 and sagesse, 8-9
Chanson/chancon, xi, 26, 34, 44—6, 84—
5, 135, 188, 194 Dame a vos icestui lay mant, 99
Chanson de Geste/Cansonz de Gesta/ Dance, 11, 14, 24, 38, 48, 60-1, 77-91,
Cantus gestualis, 22, 68, 129, 162, 135, 154-6, 163, 169, 173, 181, 187,
189, 194 190, 2393255
Chanson de toile, 35—8, 249 Dansa/danga, 17, 22—25, 38, 41-8, 134—
Chan(son) poitevin(e), 29-34, 184—5, 6, 247
194-5 Daude de Pradas Bele m’est la voiz altane,
Chanson de Guillaume, 159 31-2
Chanson des Quatre Fils Aymon, 158, 165, Daurel et Beton, 6, 20, 22, 28, 164, 166,
187-8 246
Chansonnier de |’Arsenal, 51, 245 De Animalibus (Albertus Magnus), 237
Chantan dissera si pogues (Guillem De bonne amour et de leaul amie, 74
Ademar), 164, 174—5 De Institutione Musica (Boethius), 149
309
Index
Gerson: see Jeande Gerson 112-22, 136-7, 148-9, 155, 158-9, 162,
Gervase of Tilbury Otia Imperialia, 7, 166, 172, 186—8, 191, 194, 217-8, 221,
164, 177-8 229, 233-7, 240, 242, 264-5, 270
Gesta Herewardi (Richard of Ely), 112, He! aloete, 82
265 Heinrich Eger von Kalkar, 269
Geste de Blancheflour e de Florence, 200 Henri de Malines/Henri Bate Nativitas
Giga/gige/gigue/zigga, 54—5, 84, 123, magistri Henricit Mechliniensis, 59-61,
145-7, 162, 170, 172; 187,191,220, 81, 85, 90, 256, 260
241 Henry of Cossey In Psalmos 146, 215-6,
Gille de Chyn (Gautier de Tournai), 30— 220, 240, 277
4, 157, 165, 186 Herbert of Bosham In Psalmos, 254, 277
Gilles Li Muisis, 58 Hervis de Metz, 31, 32, 34, 155, 162-3,
Girart de Roussillon, 129, 268 165, 187, 248
Gitterns/Quiterne/Guisterne, 132-3, High Style Manner, 10-17, 22-7
P75 829) 2173 2205236; 242) 275 (instruments and subversion of), 31—
Grand chant courtois, 12, 14, 15, 200 34, 38, 44, 48, 67-77, 85, 86, 102, 134,
Gregory Dialogues, 203 137
Grocheio: see Johannes de Grocheio Histoire de la Guerre de Navarre (Guilhem
Grosseteste (Robert), 255, 278 Anelier), 271
Guido of Evreux, 259, 279 Historia Apollonii Regis Tyri, 103-5, 167—
Guillem Anelier Histoire de la Guerre de 171, 244, 263—4, 272
Navarre, 271 Historia Regum Britannie (Geoffrey of
Guilhem Augier Novella Ses alegratge, 247 Monmouth), 121, 122
Guilhem de la Barra (Arnaut Vidal de Hiyya ben Abba, 54
Castelnaudary), 164, 173, 184 Honorius Augustodunensis Gemma
Guillaume d’Auvergne De Universo, 81, Anime and In Psalmos 81, 165, 205—6,
83, 165, 203—4, 259 219, 226-7, 278
De Viciis et Peccatis, 165, 204, 219, 238, Hortus Deliciarum, 144, 150
pH Hugo von Trimburg Der Renner, 220,
Guillaume de Deguilleville La pélerinage 238-9
de la vie humaine, 144 Humbert of Priallaco, 280
Guillaume de Dole, see Jean Renart Huon de Meri Le Torneiment Anticrist,
Guillaume Peyraut Summa de Vitiis et 165, 194-5
Virtutibus, 89, 165, 202—3, 259-260, Huon d’Oisy Tournoiement des dames, 249—
279 250
Guillem Ademar Chantan dissera si pogues, Hurdy-gurdy, 144, 147, 150
164, 174-5
Guillem de Bergueda, 24, 247 Imbert d’Aiguiéres, 177
Guillem Molinier Leys d’Amors, 42, 45, In Psalmos (Arnobius Junior), 214, 219,
164, 173, 179, 184, 251 223, 278
Guiraut de Bornelh, x, 26—28 In Psalmos (Bodleian Library, Oxford,
Guiraut de Cabreira (Giraldus de MS Bodley 860), 219, 227-8
Cabreriis), 7—9, 18—20, 38, 178-9 In Psalmos (Bodleian Library, Oxford,
Guiraut Riquier, 271 MS Hatton 37), 219, 233-4
Guiron le Courtois, 97-102, 279 In Psalmos (Bruno the Carthusian), 210-—
12, 214, 219, 224—5, 278
Harmonie Universelle (Marin Mersenne), In Psalmos (Henry of Cossey), 146, 215—
125,-137 16, 220, 240, 277
Harp/Arpa/Harpe, x, 4-6, 86, 92-107, In Psalmos (Henry of Bosham), 254, 277
SAL
Index
Libre de Contemplacié en Deu (Ramon Coinci), 61, 86—7, 165, 177, 191-4,
Llull), 164, 180-3 256, 262
Libre d’Evast e d’Aloma e de Blanquerna Mireour du Monde, 259
(Ramon Llull), 164, 183-4 Modus, 137
Libre de Meravelles (Ramon Lull), 164, Molinier: see Guillem Molinier
183 Moniot d’Arras Ce fut en mai, 135
Libro de Alexandre, 169, 170, 272 Montaner, 271
Libro de Apolonio, 163, 167—71 Monocordium, 238
Libro de Buen Amor (Juan Ruiz), 169, 272 Mors vite propitia, 87
Libro de las Propriedades de las Cosas Motet, 50, 68, 73, 75-9, 259
(Vingente de Burgos), 236 Musica falsa, 195-6
Lietbertus of Lille Flores Psalterii, 253, Musica ficta, 62, 63, 112, 117-8, 170, 200,
277 265
Literacy, verbal and musical, 53—76 Musica recta, 62, 63, 117, 130, 170
Llull: see Ramon Llull Musica speculativa (Jehan des Murs), 52—
Lower Styles in French and Occitan 3, 143, 253
lyric poetry; their affinity with
instrumental accompaniment, Narrative songs in minstrel repertory,
chapters 2 and 3 passim 19
Lumiere as Lais (Pierre of Peckham), 115, Nativitas magistri Henrici Mechliniensis
119 (Henri de Malines), 59-61, 81, 85,
Lute/luthz/liuto, 54, 132—3, 138, 147, 90, 256, 260
217, 220, 237, 242, 274 Nicholas de Gorran In psalmos, 220, 239,
Lydgate (John), 224 ae)
Lyre/lira, 144—50, 171, 179, 192-3, Nicholas de Lyre Postilla, 220, 239, 254,
206-8, 234—5 277-8
Lyre/lire, 121, 124, 148, 192-3, 222—4 Nicholas Trevet, 143, 277-8
Nicholaus Inclitus, 88
Macaire, 165, 195 Nota/note, 4, 25, 30, 92-3, 162, 172,
Manuscripts, 277-81 176, 179, 191, 262
Manuscrit du Roi, 17, 23, 42, 250, 252 Notker Labeo, 253
Marcabrun, 18, 19, 29, 173, 246
Marie de France, 106, 107 Odon Rigaud, 90
Martianus Capella, 137 Organ, 60, 117-21, 144, 266
Matfre Ermengaud Le Breviari d’Amor, Organer, 92-3, 121-2, 161
271 Organistrum, 144, 150, 238, 269
Medius canon, 122—4 Organon, 187
Merlin, 218-19, 234 Organum, 118, 125, 136, 227, 230
Mersenne (Marin) Harmonie Universelle, Or voi je bien qu’il mi, 75
'25,137 Otia Imperialia (Gervase ofTilbury), 7,
Mervelles de Rigomer, 158 164, 177-8
Messire Thibaut Le Roman de la Poire, Otto IV, 178
165, 188 Ovid, 53
Michael of Meaux In Psalmos, 219, 231—
3, 278 Palsgrave Lesclarcissement de la Langue
Minstrel(s), x, 9-12, 19-22, 26-8, 30-1, Francoyse, 140
33-5, 83—4, 122, 136-7, 154-8, 162- Paris, 49-75, 190-201
3, 170-95, 201-2, 218, 226, 243, 253 Par un matinet lV’autrier, 76
Miracles de Nostre Dame (Gautier de Partimen, 22—3, 44-6
o13
Index
Trouvéres, 8—12, 19, 29-38, 51-2, 60, 135-9, 145, 155, 158, 162—3, 166—96,
64, 66-9, 72—75, 85, 87, 93, 106, 111, 204, 216—17, 220, 235-8, 241-2, 255-
134-8, 200, 249 .8, 270
Tuning, 58, 62—5, 70, 93, 111ff Vincente de Burgos Libro de las
Tympanum, 229, 230, 253 propriedades de las cosas, 236
Vinsauf: see Geoffrey de Vinsauf
Van den Proprieteyten der Dinghen, 237 Virelai, 24, 68, 81—4, 247, 261
vers, xi, 11, 18, 26, 40, 44-6, 135, 161, Vita Sancti Christophori, see Walther of
189, 194 Speyer
Vidame de Chartres Quant la sesons del Voyage de Charlemagne, 155, 163
douz tens s’asseure, 9-11
Vidas, x, 7, 27, 164, 175—6, 243, 270, 272 Wace Brut, 121-2, 154, 161-2, 187, 271
Vida del Glorios Sant Frances, 271 Waleys (Thomas), 133
Vie de S. Auzias, 271 Walther of Speyer Vita Sancti Christophori,
Viella/vielle/viele/viola/viula/vihuella/ 219, 223—4
viuela/fedele/fiddle, 6—7, 10, 18-23, Wenrico ConflictusOvis et Lini, 219, 224
30, 37, 53-74, 84, 88, 111, 126-133, William of Auxerre Summa Aurea, 280
316
66 00 0538 dl
Page, Chris “Awe i.
S an
;
Daun Blasberp "3 "
hiwt hoals be 10.a ae
"Chetan 3/4 4 l
The Music of William Byrd, Volume IIT: Consort and Keyboard Music
.
by Oliver Neighbour —
ISBN 0-520-05932-8