Beruflich Dokumente
Kultur Dokumente
A dissertation presented
by
ANNA ANATOLIEVNA ZAYARUZNAYA
to the Department of Music
in partial fulllment of the requirements
for the degree of Doctor of Philosophy
in the subject of Music
Harvard University
Cambridge, Massachusetts
August 2010
2010, Anna Anatolievna Zayaruznaya
All rights reserved.
iii
Form and Idea in the ARS NOVA Motet
ABSTRACT
Although ars nova motets have traditionally been viewed as purely mathematical
due to their highly structured forms, recent studies by Margaret Bent, Jacques Boogaart
and others have challenged this notion, arguing that text and music are sometimes intri-
cately linked. Building upon these analyses of individual works, the present study aims at a
broader evaluation of text-music relations within the repertory.
Part One is dedicated to identifying the units and mechanisms of text-music rela-
tions. This involves exploring the reception of motet texts, on the one hand, and the variety
of their musical forms, on the other. I nd that motet reception, as revealed by citation
practices, manuscript transmission, and literary engagement favors the upper voices, which
in turn inuence the structures of motets in ways often audible to audiences. Such emphasis
seems in conict with the commonly held view that polytextuality masks texts in perfor-
mance. However, cognitive science and historical evidence can both show that the supposed
limitations of polytextuality need not hinder understanding. The idea that upper-voice
texts may generate musical forms also grates against the notion that motets are structured
from their tenors upwards, but a closer look at upper-voice rhythmic organization reveals
that a signicant number of motets in the repertory have upper-voice structures that super-
sede those of the tenor.
Part Two consists of a series of case-studies focusing on a group of motets whose
main ideas are disjunct or hybrid: the goddess Fortune, a chimera, a piecemeal statue. In
Professor Sean Gallagher, advisor Anna Zayaruznaya
iv
these works, the musical settings turn out to be as fragmented as the creatures with which
they are paired, showing segmentation on textural and isorhythmic levels. These hybrid
ideas and their far-reaching effects on musical forms inect our understanding of late-
medieval modes of musical depiction. More than this, when viewed as a group these motets
have the potential to radically alter our understanding of ars nova aesthetics, suggesting that
disjunction, rather than unity, may sometimes have been the highest aim of composition.
v
FOR MY PARENTS,
without whose courage and foresight
none of this would have been possible.
vi
CONTENTS
ACKNOWLEDGMENTS ix
ABBREVIATIONS, SIGLA AND EXAMPLES xiii
INTRODUCTION 1
PART ONE: FORM AND IDEA
1. MOTETS AS IDEAS 9
The Ontology of Motets 17
The Indexing of Motets and Manuscript Layout 17
Motet Citations in Theoretical Treatises 24
The Transmission of Text 32
Upper-Voice Texts 33
Tenor Texts 48
Intabulations 50
Motets and/as Literature 54
Reception of Ars nova Motets Outside of Ars nova Circles 63
Appendix 1A: Citations of Surviving Ars nova Motets in Treatises 68
2. HEARING VOICES 73
Medieval Listening Practices and Modern Ears 75
Timbre 84
The Cocktail Party Phenomenon 93
3. WHAT IS A TALEA? UPPER-VOICE PERIODICITY IN ARS NOVA MOTETS 106
The Tenor as Foundation? 108
The Evidence for Independent Upper-Voice Taleae 115
Terminology and Diagrams 118
Supertaleae as Grouped Tenor Taleae 119
More Intricate Upper-Voice Arrangements 127
Isorhythm and Memory 134
Shifts Between Upper- and Lower-Voice Taleae 144
Upper-voice Structures and Hermeneutics in Sil estoit/SAmours (M6) 152
Conclusions: Independent Upper-Voice Structures in the Ars nova Motet 167
vii
PART TWO: MUSICAL DISJUNCTION
4. VOICE-CROSSINGS AND FORTUNA IN MACHAUTS MOTETS 173
The Motetus Corde mesto cantando conqueror 181
Singing from Fortunes Wheel 183
Cece Fortuna and Blind Isaac 195
Amours/Faus Samblant (M15) 198
Motet 14s Lying Voices 203
The Dishonesty of Poets 207
Fausse Fortune and Amour languour 209
Other Fortuna Crossings 216
Appendix 4A: Helas/Corde mesto, Texts and Translations 222
Appendix 4B: Maugre/De ma dolour, Texts and Translations 225
Appendix 4C: Hlas/Corde Mesto/Libera me, Edition 228
5. THE MONSTER IN THE MOTET 234
The Texts of In Virtute/Decens 235
Hockets and Rhetoric 240
The Isorhythmic Scheme of In virtute/Decens 243
Hockets and Wordlessness 249
A Monster-Shaped Motet 256
Intensied Monstrosity 260
Hybriditys Ambivalence 262
Vitry and the Zytiron 264
Ut pictura motetus? 273
Appendix 5A: In virtute/Decens, Texts and Translations 278
Appendix 5B: In virtute/Decens, Edition 280
6. VITRYS CUM STATUA/HUGO/MAGISTER INVIDIE AND LATE-MEDIEVAL
INTERPRETATIONS OF NEBUCHADNEZZARS DREAM 285
The Statues Layers: A Metallurgical Summary 289
The Motets Layers: A Formal Summary 293
The Composite Tenor of Cum statua/Hugo 296
The Beginning of the Motet: Cum statua... Hugo [est] 303
Gradatim deduci ac minus: The Statues Layers in the Motet 307
The Motets Layers: Questions of Isorhythmic Form 311
Feet of Clay: Hockets and Fragmentation 314
Phi millies/O Creator/Iacacet granum/Quam sufabit 318
viii
Nebuchadnezzars Statue in Machauts Complainte 324
Nebuchadnezzars Statues 335
Gowers Divisioun and the Musical Statue 350
Epilogue: Ars nova and Disjunction 361
Appendix 6A: Cum statua/Hugo, Texts and Translations 367
Appendix 6B: Cum statua/Hugo, Edition 369
Appendix 6C: Phi millies/O Creator, Texts and Translations 372
CATALOG OF ARS NOVA MOTETS, THEIR SOURCES, AND EDITIONS 375
BIBLIOGRAPHY 386
ix
Acknowledgements
I
F, AS BOETHIUS JUDICIOUSLY WARNS, good Fortune is to be mistrusted and can be counted
on only in its inconstancy, then Im in for it. While writing this dissertation I have been
astronomically fortunate in the help and support I have received from a list of colleagues,
friends, and family so long that it can only signal my eventual demise. That is, unless we can
consider their inuence Providential rather than merely Fortunate.
Even before I had fully settled on a topic, my colleagues in the thriving eld of me-
dieval and renaissance studies were unfailingly generous with their time, energy, and exper-
tise. Margaret Bent set things in motion in 2004 by suggesting that I look at a few motets
by Machaut. And she has stood by me since then, productively challenging and encouraging
me in turn and providing access to unpublished work and other invaluable resources.
Alejandro Enrique Planchart has also been an ally for many years, and I thank
him for sharing his insight and enthusiasm on a range of topics, and for the very material
loan of the fourteenth centurys most important equine Antichrist. To Jacques Boogaart
I am grateful for his generosity with unpublished work and for saving me from several
embarrassing errors. Jane Alden, Bonnie Blackburn and Dorit Tanai were instrumental in
bringing Chapter 4 to its nal form. Lawrence Earp kindly offered advice on Chapter 1.
Michael Scott Cuthbert, my closest colleague in geographic and temporal terms, has helped
and advised me at various points in my career. On the eve of my dissertation printing I wish
I had heeded his earliest piece of advice: to buy a color laser printer. For discussing difcult
texts that would have been nonsense to me without their expertise I thank Leofranc Hol-
ford Strevens and Gabriela Currie. And for his help with a text that is actually nonsense, I
thank Michael Randall.
x
At the beginning and again at the end of my writing I was bolstered in my resolve
by the warm and knowledgeable group of colleagues that gather in the dolomites to think
about medieval music. By organizing these conferences at Novacella Karl Kgle does our
discipline an enormous service. I thank him for this as well as for his help with motet-
related things.
In Massachusetts, Jane Bernstein, Joseph Dyer, Lewis Lockwood, Virginia Newes,
and Joshua Rifkin have provided support, encouragement, and a steady stream of difcult
questions. In Californa, Anna Maria Busse Berger, Beth Levy, and William Mahrt made me
feel welcome. Jesse Rodin has been a friend, colleague, and co-conspirator on both coasts.
Members of Harvards medieval studies community have had a profound inuence
on the extent to which I have felt interested and able to deal with the ideas and images be-
hind the music. I am especially grateful to Jeffrey Hamburger, Michael McCormick, James
Simpson, Hugo van der Velden, and Jan Ziolkowski. My fellow students in medieval stud-
ies seminars, Steven Rozenski, Anna Huber, and Beatrice Kitzinger, shared with me their
enthusiasm and some very useful references.
This project would not have been possible without the tremendous aid offered by
librarians: at Harvards invaluable Isham library, Sarah Adams and Doug Freundlich tol-
erated my perpetual residency and offered continuous held and support. At Loeb Music
Library, Virginia Danielson, Kerry Masteller, and Andrew Wilson made me feel welcome
despite my propensity to leave at 9:59 PM (and sometimes, if Im to be honest, at 10:01).
I am also thankful to William Stoneman and the staff of Houghton Library for granting
me access to several treasures. Digital access to other treasures was facilitated by the Digital
Image Archive of Medieval Music, and I thank Dr. Julia Craig-McFeely for making this
xi
invaluable resource available, and especially for allowing me to consult images of the MS
FerrellVoge, and to reproduce below a few details from this fascinating source. Thanks
also to John Shepard of the Jean Gray Hargrove Music Library at UC Berkeley for making
the beautiful image of En la maison dedalus available to me.
Administrative, emotional, and chocolate support were provided by the powerhouse
staff of the Harvard music department: thank you Kaye Denny, Mary Gerbi, Jean Moncrieff,
Nancy Shafman, Karen Rynne, Charles Stillman, and Fernando Viesca. Several sources of
funding made my research and writing possible: among them a Ferdinand Gordon & Eliza-
beth Morrill Graduate Fellowship, a Richard F. French Prize Fellowship, a Presidential
Fellowship from the Graduate School of Arts and Sciences at Harvard University, and an
Alvin H. Johnson AMS 50 Dissertation Fellowship from the American Musicological So-
ciety. As a director of graduate studies and later chair of the department, Professor Anne
Shrefer has provided much sound advice and encouragement.
Friends dont let friends write alone, and I am grateful to Emily Abrams Ansari,
Andrea Bohlman, James Blasina, Carolann Buff, Daphna Davidson, Louis Epstein, Ellen
Exner, Bonnie Loshbaugh, Heather Marlowe, Drew Massey, Carrie Menke, Evan MacCar-
thy, John McKay, Rowland Moseley, Matthew Mugmon, Andrew Oplinger, Gina Rivera,
David Trippett, Emily Zazulia, and Hillary Zipper for support, encouragement, and calm-
ing words and deeds. Michelle Atwood deserves special mention for reminding me from
time to time that, after all, its only musicology. And to Ryan Baagale, Corinna Campbell,
and Katherine Leethe other members of the small but mighty (ethno)musicology G3s of
2007-8I owe warm thanks for their patience and friendship, and for making me zoom
out at key points.
xii
Where I come from, committee has a harsh ring to it, and usually implies some
body of people whose purpose is to undermine individuality and uphold the status quo. The
committee that advised this dissertation could not have been further from those commit-
tees of old. They acted as three distinct voices that together had an undeniable shaping role
to the content and form of the present work, but they also let me pursue my own instincts,
make my own mistakes, and write in my own style (sometimes ill-advised, but never by
them).
Suzannah Clark has shown me what it means to be a reader. Her careful and honest
feedback has led to much fruitful revision, and the subtlety of her thinking has pushed me
to greater care and creativity in my analytical pursuits. To Thomas Forrest Kelly I am grate-
ful for his generosity, his enthusiasm, and his scepticism. His questions, which sometimes
seemed deceptively simple, have acted as important stimuli. No part of this study has not
benetted from them, and Chapter 1 is their direct result. And to my advisor, Sean Gal-
lagher, I am indebted for asking the right questions at the right times, for thoughtful com-
ments, and for continuing to encourage me when the topic or the circumstances seemed too
daunting. He has lent this work correctness and added style and dignity to many passages.
As a course head and seminar leader he has shown me what it is to inspire and challenge
students. And as I have begun the transition from student to teacher, he has given honest
and helpful advice at every turn.
My family on both coasts (and in the middle) have been loving and supportive
throughout this long process. I thank them for their patience and understanding. And most
of all I thank my husband Yarrow, whose ways of supporting, knowing, reading, questioning,
and loving are as numerous as the leaves of the achillea millefolium.
xiii
Abbreviations, Sigla, and Examples
Abbreviations
CMM Corpus Mensurabilis Musicae. 111 volumes. American Institute of Musicol-
ogy, 1947 .
Volumes cited: 13. Charles Van den Borren, ed. Missa Tornacensis. 1957.
39. Ursula Gnther, ed. The Motets of the Manuscripts Chantilly,
Muse cond, 564 (olim 1047) and Modena, Biblioteca estense, a. M. 5,
24 (olim lat. 568). 1998.
M1M32 Machauts motets, by number. See PMFC23.
PMFC Polyphonic Music of the Fourteenth Century. Leo Schrade, Frank Ll. Harri-
son, and Kurt von Fischer, general editors. 25 volumes. Monaco: ditions
de lOiseau-Lyre, 19561991.
Volumes cited: 1. Leo Schrade, ed. The Roman de Fauvel; The works of Philippe de
Vitry; French cycles of the Ordinarium Missae. 1956.
23. . Works of Guillaume de Machaut. 1956.
5. Frank Llewellyn Harrison, ed. Motets of French Provenance. 1968.
21. Gordon K. Greene, ed. French Secular Music: Virelais. 1987.
23. . French Secular Music: Rondeaux and Miscellaneous
Pieces. 1989.
Manuscript Sigla
Apt16bis Apt, Cathdrale Sainte-Anne, Bibliothque du chapitre, Trsor MS 16bis
Apt9 Apt, Cathdrale Sainte-Anne, Bibliothque du chapitre, Trsor MS 9
Arr983 Arras, Bibliotheque Municipale, MS 983 (olim 766), yleaf
Barc853 Barcelona, Biblioteca de Catalunya (olim central), MS 853
Barc971 Barcelona, Biblioteca de Catalunya (olim central), MS 971 (olim 946)
Be421 Bern, Burgerbibliothek, A. 471 (yleaves from A. 421)
BN 1112 Paris, Bibliothque Nationale de France, MS fonds latin 1112
Br19606 Brussels, Brussels, Bibliothque Royal Albert I, MS 19606
Br5170 Brussels, Archives gnrales du Royaume, Archief Sint-Goedele 5170
(Olim758)
xiv
CaB Cambrai, Bibliothque Communale, B 1328 (olim 1176)
1
Chantilly Chantilly, Muse Cond, MS 564 (olim 1047)
Cort Cortona, Archivio Storico del Comune, 2 fragments without shelfmark
Durham Durham, Cathedral Library, MS C.I.20, yleaves
Fauvel Paris, Bibliothque Nationale de France, MS fonds franais 146
Ferrell-Vog Cambridge, Corpus Christi College, Ferrell-Vog MS. Private Collec-
tion of James E. and Elizabeth J. Ferrell, on deposit at the Parker Library,
Corpus Christi College, Cambridge
FriZ Fribourg, Bibliothque Cantonale et Universitaire, Z 260
Ivrea Ivrea, Biblioteca Capitolare, MS CXV(115)
Leiden 2515 Leiden, Bibliotheek der Rijksuniversiteit, Bpl 2515.
Leiden 342A Leiden, University Library, MS fragment in group Ltk 342.a, from the
binding of MS Ltk 342A
Lpr 163 London, The National Archives (olim Public Record Ofce), E
163/22/1/24
Machaut A Paris, Bibliothque Nationale de France, fonds franais 1584
Machaut C Paris, Bibliothque Nationale de France, fonds franais 1586
Machaut E Paris, Bibliothque Nationale de France, fonds franais 9221
Machaut J Paris, Bibliothque de lArsenal, MS 5203
Machaut Pm New York, Pierpont Morgan Library, M.396
Machaut Vg see Ferrell-Vog
Mbs 4305 Munich, Bayerische Staatsbibliothek, Clm 4305
McVeigh London, British Library, Additional 41667(I)
ModA Modena: Biblioteca Estense e Universitaria, a.M.5.24 (Latino 568; olim
IV.D.5)
ModB Biblioteca Estense e Universitaria, a.X.1.11 (Latino 471)
Munich31 Munich, Bayerische Staatsbibliothek, Handschriften-Inkunabelabteilung,
Latinus monacensis 5362, Kasten D IV ad [31]
1
Foliation as in Lerch, Fragmente aus Cambrai.
xv
Nr9 Nuremberg, Stadtbibliothek, Fragment lat. 9 (from Centurio V, 61)
Oas 56 Oxford, All Souls College, MS 56, binding strips
Ox 213 Oxford, Bodleian Library, MS Canonici Miscellaneous 213
Ox 271 Oxford, Bodleian Library, MS. Bodl. 271, binding fragments
PadC Padua, Biblioteca Universitaria, MS 658
Paris 2444 Paris, Bibliothque Nationale, fonds nouv. acq. latines 2444
Paris 571 Paris, Bibliothque Nationale, fonds franais 571
PArs 595 Paris, Bibliothque de lArsenal, MS 595
PPic 67 Paris, Bibliothque Nationale, Collection de Picardie 67
Robertsbridge London, British Library, Add. 28550
RosL Rostock, Universittsbibliothek, phil.100/2
SL2211 Florence, Biblioteca Medicea-Laurenziana. Archivio Capitolare di San
Lorenzo, ms 2211.
2
Strasbourg Strasbourg, Bibliothque Municipale (olim Bibliothque de la Ville), MS
222.C.22 (now destroyed; facsimile of Coussemakers transcriptions of
some works in Vander Linden, Le manuscrit musical )
Tarr(1) Tarragona, Archivo Histrico Archidiocesano, ms s.s. (1)
Tarr(2) Tarragona, Archivo Histrico Archidiocesano, ms s.s. (2)
Torino 42 Torino, Biblioteca Reale, Vari 42
Trmolle Paris, Bibliothque Nationale de France, ms fonds nouvelles acquisitions
franaises 23190 (olim Angers, Chteau de Serrant, Duchesse de la Tr-
molle)
Tou 476 Tournai, Chapitre de la Cathdrale 476
Udine Udine, Biblioteca Comunale Vincenzo Joppi, ex Archivio Florio 290
Wroclaw Wroclaw (Breslau), Biblioteka Uniwersytecka Ak 1955/KN 195 (olim MS
fragment 82 from I.Q.411)
Yox Ipswich, Suffolk Record Ofce HA 30, 50/22/13.15
2
Ndass new foliation is used throughout; see Manuscript San Lorenzo 2211, 15468.
xvi
Terminology and Titles
Since it no longer seems likely that there was ever a treatise called Ars nova, I take
these words as a periodic and stylistic designation that need not be capitalized. I have used
the Latin for longa to prevent confusion with the adjective long,, but breve and min-
im as opposed to brevis and minima, to avoid excessive italicization. Foliation is indicated
with v for verso and only the folio number for the recto. Motetus rather than duplum
is used throughout to designate the middle voice in a three-voice motet, regardless of the
language of its text.
Motets are cited by short incipits in the order Triplum/Motetus. The Catalog of Ars
nova Motets, their Sources, and Editions at the end of this study provides longer incipits,
including tenor labels, as well as source and edition information for each motet.
Translations are mine unless otherwise attributed.
Examples and Figures
Unless noted otherwise, musical examples have been newly edited for this study, us-
ing the clearest or most complete source available. I have allowed note-values to remain un-
reduced and preserved original note-shapes, but it should be stressed that these are editions
using simplied ars nova notation rather than diplomatic transcriptions. Thus ligatures and
multi-bar rests are broken up to make alignment in score possible. Dots of addition are
represented, but not dots of division, since bar-lines are used. In cases where multi-measure
rests affect alteration (such as example 3.9), they are preservedotherwise not. Modern
clefs are used throughout.
xvii
In some of the examples the music has been shrunk in order to demonstrate larger
points about form. In this case it is not necessary to see the individual notes, but readers of
the higher resolution PDF version should be able to zoom in for details. In some cases color
is used to clarify analytical points. If you are reading this dissertation in a low-resolution
copy scanned by UMI, please contact the author for a PDF.
All images are either reproduced with permission, are under the authors own copy-
right, are small details printed under fair use, or are in the public domain. The latter status
for photographic copies of public domain images was reafrmed by the 1999 decision in
the case of Bridgeman Art Library v. Corel Corp.
INTRODUCTION
I
REALLY THINK THAT PROSODY AND THE SENSE OF THE WORDS have no importance in the
isorhythmic motet. It proceeds from a purely musical construction; contemporary
music, in this regard, is akin to the ars nova.
1
So opined musicologist Jacques Chailley in
response to the 1955 lecture in which Willi Apel coined the term pan-isorhythm.
2
He
was not entirely without oppositionSuzanne Clerx had earlier suggested that along-
side the mathematics which is the base of isorhythmic motets there is also the inspira-
tion, the imagination and the necessity of adapting the musicto a text which also has its
requirements.
3
But Ren Lenaerts countered with a rather categorical reply: I dont think
so, due to the fact that relations of text and music only came to life at the end of the four-
teenth and in the fteenth centuries.
4
Chailley and Lenaerts expressed a prevailing view. The relationship between text
and music, which has long been a chief concern in the study of song, is usually referred to
as word-tone relations. These unitsindividual words and small groups of notesare
1
Je pense vraiment que la prosodie et le sens des mots na aucune importance dans le motet isorythmique.
II sagit dune construction purement musicale; la musique contemporaine, a cet gart(sic), est proche aussi
de 1ars nova et la rcente cantate de Leibowitz le montre bien, published in Apel, Remarks about the
Isorhythmic Motet, 145.
2
In drawing parallels between the purely mathematical attitude of Boulez and his partisans and ars
nova motets, Chailley was not doing either repertory a favor. As a composer, he preferred more conservative
techniques, and used serialism only in the service of satire. For example, in Diafoirus pre et ls in his Suite
sans prtention pour Monsieur de Molire (1953). See Spieth-Weissenbacher and Gribenski, Chailley,
Jacques.
3
ct de la mathmatique qui est la base des motets isorythmiques, il y a aussi linspiration, la fantaisie
et les ncessites dadaptation dune musique, savamment labore, a un texte qui a aussi ses exigences, Ibid.
144.
4
Je ne le crois pas, car les rapports du texte et de la musique ne deviennent vivants qu partir de la n du
XIVe et du XVe sicle, ibid., 1445.
2
the scale on which text and music most obviously interact in later repertories. Word-tone
relations are epitomized in 16th-century text-painting: the well-known melodic ascents on
skies and stars in madrigals, the upward runs on et ascendit in caelum in renaissance
masses, and so forth. When judged by these standards, earlier repertories do indeed seem
to fail in relating words to music. Writing at the same time as Chailley, Alfred Einstein de-
scribed the addition of voices to a chant as the smothering new garb under which the text
usually disappears.
5
And we can see the same ethos operating thirty years later with Daniel
Leech-Wilkinsons assertion that in Machauts view, at least, musical form operated, to a
large extent, independently of textual association.
6
In light of the scholarship of the last twenty years such a view is no longer tenable.
A number of careful and sensitive analyses of individual worksmostly from Machauts
oeuvrehave shown that the music of ars nova motets can reect their texts through men-
sural and isorhythmic design, textural manipulation, control of diction, the symbolic use of
number, and a wide array of other techniques.
7
Attention to the musical, textual, and con-
textual content of motet tenors has widened the realm of analysis by increasing the number
5
A. Einstein and E. Sanders, trans., The Conict of Word and Tone, The Musical Quarterly 40, no. 3
(1954): 338.
6
Machauts Rose, Lis, 13.
7
The earliest analyses that argue for correspondence between form and meaning are Reichert, Das Ver-
hltnis, Eggebrechts two analyses of Machauts Fons/O livoris (M9) (Machauts Motette Nr. 9, and Mach-
auts Motette Nr. 9, Teil II) and the analyses, especially the one of Sub arturo/Plebs, in Gnther, Das Wort-
Ton-Problem. More recently Margaret Bent has argued that a motets texts are manifest in compositional
decisions which have musical, rhythmical, structural, and symbolic-numerical manifestations. In the case
of Machauts Amours/Faus Semblant (M15) the texts expressed are ideas of falsehood and deception linked
with False Seeming (Deception, Exegesis, and Sounding Number); in Vitrys Tribum/Quoniam, musical
structure and diction are linked with the Ovidian quotation at the end of the motetus and with the ideas
of sudden downfalls and reversals which it embodies (The Vitry motet Tribum que); and Machauts Fons/O
livoris (M9) shows a number of correlations between musical and textual structure; see Words and music.
Jacques Boogaart has explored text-music relations in many of Machauts motets; see O series summe rata,
and Loves Unstable Balance, Part I. See also Dillon, The Prole of Philip V.
3
of texts with which form might interact.
8
And consideration of interrelationships between
motets has further expanded the arena in which musico-poetic associations may play out,
allowing for analysis on the level of oeuvre or manuscript.
9
The perhaps inevitable side-effect of this is that motets have become difcult.
If the ars nova motet of 1955 was a purely musical and sonic object, the ars nova motet of
2010 is almost intimidatingly rich in meaning. In Alice Clarks summary,
The complexities inherent in the genreincluding bitextuality, number
symbolism, allusions to other motets, and other techniques that are inau-
dible or that cloud the surface comprehension of text and musiccan make
us wonder whether anyone listened at all, and if so, what they heard.
10
This list of complexities the result of careful and imaginative studiesis both a boon
and a weakness for our understanding of the genre. For while opening up exciting new
arenas for investigation, the existing readings present us with a challenge. So far, the most
productive approach has been to focus on individual motets, and even analyses of multiple
works may ask a different set of questions, and indeed even call upon a separate set of
methodologies to explore the semantic, cultural, and musical content of each motet. Like
a Mahler symphony, each motet is a world in itselffull of intellectual sophistication, in-
tricate compositional schemes, and deeply coded meaning. But these worlds may well be in
different galaxies.
11
8
For studies of how tenors relate to musical and poetic aspects of motets, see Clark, Concordare cum mate-
ria, Robertson, Guillaume de Macaut, and Maurey, A Courtly Lover.
9
Several studies have addressed the ordering of motets within a corpus; on Machaut, see Brown, Another
Mirror for Lovers?, and Robertson, Guillaume de Machaut. On a series of motets in Fauvel, see Bent, Fau-
vel and Marigny.
10
Clark, Listening to Machauts Motets, 487.
11
Boogaarts analyses are an exception, since he considers Machauts motets as an oeuvre and sometimes
applies the same analytical techniquefor example, the interpretation of talea rhythmsto several works at
a time. However, his most detailed analyses, such as his analysis of Motet 6 (discussed in detail in Chapter
3 below) focus on the internal poetics of one work. See his O series summe rata and Loves Unstable Bal-
4
The incommensurability of the existing analyses of ars nova text-music relations is
not necessarily a problem, but it makes it difcult to move beyond the individual work to
consider compositional aesthetics, hermeneutics, and modes of signifying within a reper-
tory. Furthermore, this approach to analysis can leave certain basic questions about genre
unanswered, since an interpretation that stresses the depth and uniqueness of a given work
is more likely to read its properties as individual rather than generic traits. But the oppo-
site approach is perhaps more unpalatable, for in deciding that a given genre is made up of
works that are similar, we will stress the similarities and miss the subtleties of individual
compositions. Indeed, how can we begin to understand the galaxy if we do not know its
worlds?
It is the aim of this study to occupy a middle ground. Though Part II concerns itself
with case-studies that follow in the methodological footsteps of existing work, these case-
studies are linked by similar analytical approaches and common units of analysis. It is my
contention that the units of word and tone, inherited from text underlay discourse and
madrigalisms, are not productive for discussions of musical-semantic relations in the rep-
ertory of ars nova motets. Rather, I will suggest that larger phrases or even entire composi-
tions depict the main semantic ideas of texts through bold textural and formal gestures.
Both form and style as possible loci of expression were addressed by Ursula Gnther
in an incisive 1984 essay. With the goal of codifying an array of text-music relations in
the middle ages, she built a ve-step ladder whose rungs progress from the most obvious
relations (mimetic) through more complex arrangements, such as pictorial and emphatic
uses of music. On the nal rung of the ladder are located those works which relate text and
ance, Part I.
5
music in the formal makeup of a composition, as in motets, canonic works or retrograde
rondeaux.
12
Here Gnther includes two works notated on circular staves and three compo-
sitions by Machaut: the self-descriptive rondeau Ma n est mon commencement, the trinitar-
ian three-voiced Lai de le fonteine, and the tritextual canonic ballade Sans cuer/Amis dolens/
Dame par vous, which stages a conversation among its voices.
13
Motets are rather underrep-
resented on all rungs of Gnthers ladder, but it is on this fth level that the issue comes to
the fore. Though she concedes that there are even some motets in which form and text are
connected so as to produce meaning, in the end she cites only one: Sub Arturo/Fons, whose
motetus text summarizes the diminutions which the tenor undergoes.
14
In other motets,
cautions Gnther, it seems at least less certain, or even questionable whether we can nd
intentional connections between the isorhythmic construction and the numbers mentioned
in the text.
15
In conclusion, she tentatively suggests two possible motets in which a tenor
talea repeated seven times might signify the seven liberal arts.
It is interesting to see how ideas and the forms which depict them change for Gn-
ther depending on the genre in question. Theme or subject is interpreted loosely when it
comes to songs: one of the circular rondeaux depicts a labyrinth, Machauts ballade evokes
a conversation, the lai enacts the Trinity. And form, too, is a broad enough concept there
to include number of voices, canonic techniques, and page-layout. But in motets the only
12
im formalen Aufbau einer Kompositon(sic), etwa bei Motetten, kanonischen Werken oder retrograd
aufzulsenden Rondeaux, Ibid., 236.
13
Ballade 17, edited PMFC 3:889. On this work, see Newes, Dialogue and Dispute, 715.
14
Schlielich gibt es sogar einige Motetten, bei denen Text und Form eine Sinnbeziehung aufweisen,
Gnther, Sinnbezge zwischen Text und Musik, 267. Sub Arturo/Fons is edited in Bent, Two 14th-century
Motets.
15
Bei anderen motetten scheint es allerdings weniger sicher oder sogar fraglich, ob man zwischen den
im Text erwhnten Zahlen und der isorhythmischen Konstruktion eine bewut angestrebte Versinnlichung
sehen sollte. Gnther, Sinnbezge zwischen Text und Musik, 267.
6
valid form is isorhythmic, and thus the only text which can be represented by this form is
one that evokes number. To be sure, numbers were important to the study of late-medieval
motets when Gnther wrote , and they remain so today despite challenges to isorhythm
as the paradigm for motet construction.
16
Gnthers evaluation is instructive here in that
it reminds us that both form and subject are slippery notions whose denitions can be
broad or narrow depending on ones view of the genre.
Part One of the present study is concerned with rening these vague terms as they
might apply to ars nova motets. In Chapter One, I bringing together various strands of re-
ception, from the scribal to the poetic, to explore the ontology of motets for their medieval
listeners. Specically I focus on the different roles played by upper-voice and tenor texts in
the naming, transmission, and citation of motets. The genre emerges as rather top-heavy:
though tenor melodies undoubtedly have a role in the construction of motets, reception
within ars nova circles repeatedly stresses upper-voice texts, which are more carefully trans-
mitted in both musical and poetic sources. Nor is the situation very different for more
peripheral audiences, though interesting variations in emphasis are evident.
But the idea that motets would be encapsulated for medieval listeners primarily by
their upper-voice texts raises a set of questions about performance. For in combining mul-
tiple texts in their upper voices, motets are often charged with rendering those texts inau-
dible. How, then, can upper-voice texts be a key to reception? The question of intelligibility
is the focus of Chapter Two, where I argue that we may be underestimating the extent to
which texts can be audible in live polytextual performance.
16
See the accounts of the elevation of isorhythmic construction and objections to the term isorhythm
in Bent, Isorhythm and What is Isorhythm?
7
But if the ideas of motets come from their upper-voice texts, their forms are usually
thought to come from the tenor. If this is necessarily the case, then the relationship between
form and idea could only be tenuous or retrospective at best. But does all isorhythmic form
come from the tenor? In Chapter Three, I analyze a number of works that use superta-
leaeupper-voice structures which supersede those of the tenor. The variety of form that
results is in line with Bents recent assertion that isorhythmic structures are varied rather
than unied.
17
Upper-voice taleae also have interesting implications for the thesis recently
put forward by Anna Maria Busse Berger that isorhythmic composition had a strong mne-
monic component.
18
Part Two of this dissertation is devoted to the analysis of a number of works whose
forms are closely aligned with the content of their texts. Or perhaps I should say with the
denizens of their texts. For it turns out that form-idea relations play out most dramatically
in works whose hybrid ideas lead to hybrid forms. Thus Chapter Four looks at a group of
works by Machaut that embody the corporeal discontinuities of the goddess Fortune, who
is often depicted as split down the middle. Chapter Five concerns itself with an even more
disjunct creaturethe chimera from the beginning of Horaces Ars poetica. This monster,
who has the head of a woman, a horses neck, feathers and a shtail, turns out to be the main
idea of the Ivrea motet In virtute/Decens. Despite a message that seems to disparage hybrid-
ity, this motet highlights the presence of its unusual subject, controlling text declamation
in such a way as to maximally separate the creature into its disjunct parts. Chapter 6 is
concerned with a biblical monsterthe statue made of gold, silver, copper, iron, and clay
17
What is Isorhythm?, 1389.
18
Busse Berger, Medieval Music, 21252.
8
that King Nebuchadnezzar saw in an apocalyptic dream (Daniel 2). This allegorical im-
age of decay through time is present in several works by Vitry and Machaut. But the statue
also has broader cultural signicance, and can allow us to situate ars nova thought within
contemporary intellectual currents as manifest in the writings of Boccaccio, Deguileville,
Dante, Gower, and others.
When taken together, the analyses in Chapters 4, 5, and 6 show that, in some cases,
semantic contents have been the generating concept for a piece of music. Vitry emerg-
es as a composer particularly interested in disjunct forms and ideas: In virtute/Decens has
been attributed to him by several scholars, and two other motets dealing with hybridsPhi
millies/O Creator and Cum statua/Hugo are among his most securely attributed works. But a
focus on hybridity and disjunction also allows us to compare the compositional approaches
of Machaut and Vitry, since both dwell on monsters in their works, but they do so using
different devices. Most broadly, the hybrid ideas in ars nova motetsand their far-reaching
effects on musical formshave the potential to inect our understanding of late-medieval
musical aesthetics. Viewed as a group, these works suggest that disjunction, rather than
unity, may sometimes have been the highest aim of composition.
9
CHAPTER ONE
MOTETS AS IDEAS
O
N FOLIO 129 of a well-used copy of Guillaume de Nangis Chronicon (a history of
the world from the creation until 1300)
1
, Philippe de Vitry found himself read-
ing about the defeat of the Parthian Army. Their downfall reminded him of another. And
so he picked up his pen and wrote in the margin: Nota: Post zephiros plus ledit hiems, post
gaudia luctus, etc.
2
With this rst line of a couplet from Joseph of Exeters account of the
Trojan war, Vitry linked two parallel scenes of downfall and griefNangiss description of
Orodes I grieving for his son, and Exeters account of the death of King Priam, survived by
Hecuba. For both mourners,
Winter harms more after gentle west winds, griefs [harm more] after joys;
whence nothing is better than to have had nothing for the second time.
3
For us the annotation is of interest because the same couplet appears at the end of the
triplum voice in Vitrys motet Tribum/Quoniam.
4
Andrew Wathey, who rst discovered the
marginal note, has pointed to a number of thematic parallels between the chronicle and the
motet.
5
If, he argues, the motet predates the annotation, then the later use of the couplet
may well have been intended to recall not only its generalized moral proposition but also
1
Vatican, MS Regin. Lat. 544.
2
Vitry was a frequent annotator who engaged with his books in both personal and erudite waysthe for-
mer as when he annotated the year of his birth, 1291, on fol. 361r of this same book. See Wathey, Philippe
de Vitrys Books, 1458, and Myth and Mythography, 95.
3
Post zephiros plus ledit hyems, post gaudia luctus;/Unde nichil melius, quam nil habuisse secundum,
trans. Howlett in Bent, Polyphony of Texts and Music, 86. See Watheys discussion of this quotation in
Auctoritas and the Motets, 689.
4
For more information on motets cited in the text, see the Catalog of Ars nova Motets, their Sources, and
Editions at the end of this study.
5
Both deal with reprobate tribes, with the excessive ambition and cruelty of their leaders and with their
nal reduction by fortune to misery and death, Wathey, Myth and Mythography, 97.
10
to signal the parallels with the topoi of the motet.
6
But what if the annotation predates the
motet? It is possible that it did.
7
If so, I would like to imagine that here Tribum/Quoniam
was born.
And the motet reects its priorities. Writing about an Ovidian couplet (also about
sudden downfalls) that concludes the motetus text, Margaret Bent has argued that the work
is constructed backward from these lines, which are as much starting points and building
materials for both the texts and music as the motets tenor.
8
Thus whether Vitrys annota-
tion in the Chronicon represents an actual genesis for Tribum/Quoniam or simply a refer-
ence to the motet, it does in a sense encapsulate that work, causing us to think of it as a
weaving-together of disparate texts whose juxtaposition is interesting, fundamental, even
germinative: Nota...!
That is one way to think about Tribum/Quoniam. Here are two others:
tenor V+12(6i); Tr: 9L+2(12+ 12L) + 12L+9L
(upper voices 3x24 i)
9
Mo: (3+ 12L) +2( I I + 13L )+ 15L
T: 6L+3[4(6L)]
10
6
Ibid.
7
Wathey is hesitant to put the annotation before the motet, writing that seems likely [that] Vitrys motet
predates his annotation of the Chronicon but notes also that it remains unclear when Vitry acquired his
copy of the Chronicon, and that the dateable annotations indicate only that it was almost certainly in his
hands by 1342 and probably by the late 1320s (Myth and Mythography, 97n41 ). Elsewhere, however,
Wathey points out that the name of Louis de Bourbon has been erased from a list of deserters given in the
manuscript under the description of the Battle of Courtrai in 1302, and Louis de Bourbon was Vitrys
employer from the early 1320s (or earlier), so that Vitry may well have been the one to make this early
change. There is no evidence, in other words, that the book could not have been with Vitry before he wrote
Tribum/Quoniam.
8
Bent was writing before Watheys ndings had been published, and unaware of the quotation in the triplum,
The Vitry motet Tribum que, 87, 89.
9
Adapted from Besseler, Studien zur Musik des Mittelalters II, 223 and 223n12.
10
Sanders, The Early Motets, 27.
11
These are formulae Heinrich Besseler (left) and Ernest Sanders (right) used to describe as-
pects of Tribum/Quoniam which they found important. Besseler identies the overall struc-
ture of the motet as consisting of a texted introduction followed by 12 periods.
11
Sanders
focuses on the length of phrases in each of the works three voices, arguing that Tribum/
Quoniam represents an imaginative ordering of modal tradition to produce a novel, large-
scale structure in which the rst four multiples of the number 3 are all represented.
12
And here is Tribum/Quoniam in yet another guise. In explaining the mechanics of
rhythmic organization to his readers, the author of the Compendium totius artis motetorum
cites an example of each type of meter:
An example of [perfect time with] minor [prolation] is the motet Playn sui
de ameer. An example of imperfect time with major [prolation] is the motet
Adesto sancta trinitas. An example of [imperfect time with] minor [prola-
tion] is the motet [Tribum/]Quoniam secta latronum and many other motets,
rondeaux, and ballades.
13
Here the same work that we have already seen characterized as a venue for the juxtaposition
of comments on the idea of sudden downfalls and a mathematical expression of multiples
of three is being invoked as an example of imperfect tempus with minor prolation.
If in that case the upper voices of the composition are being evoked (only they dis-
play differences in prolation), other readings of motets evoke their tenors. To take another
piece as an example, Anne Robertson has argued that the key to understanding Machauts
11
V is for Vokaleinleitung. The bottom line species that the upper voices are arranged in three periods
of twenty-four imperfect longae. The observation is signicant and I will return to the question of indepen-
dent upper-voice structures in Chapter Three.
12
Sanders, The Early Motets, 267.
13
Exemplum de minori in uno moteto Playn sui de ameer. Exemplum de tempore imperfecto majori in
moteto Adesto sancta trinitas, exemplum de minori in moteto Quoniam secta latronum et in multis aliis
motetis, rondellis et baladis, Wold, ed., Ein anonymer Musiktraktat, 37. On the relationship between
the Compendium totius Artis Motetorum and the treatises which present the ars nova teachings, see Fuller, A
Phantom Treatise, 4550, and Balensuela, The Borrower is Servant to the Lender, 124.
12
Amours/Faus Samblant (M15) is through its sacred tenor, Vidi dominum (I have seen the
Lord), which illustrates the inherent importance of sight of God at this point in the [al-
legorical journey].
14
Although the motets courtly upper voicesat rst seem to resist the
sacred implications of the tenor, their content does not undermine a sacred reading of
[Amours/Faus Samblant], since even the Christian pilgrim must sometimes keep bad com-
pany.
15
Here a motet is rendered a complex object whose main ideasacred and allegori-
calows from its tenor, which is primary, into its upper voices, which are subordinate in
meaning.
And, nally, here is the same motet in different clothes: In a dream, the narrator
of Jean Froissarts Joli buisson de jonece (1373) witnesses a competition in which a group
of young allegorical people write and perform wish poems.
16
After they have recited their
wishes, the question of who will judge the competition arises, and Desire suggests that the
company go to the God of Love, who happens to be nearby. The narrators heart leaps at the
opportunity, and his joy causes him to sing:
And when I heard them say this,
My spirit rejoiced
That I would be going on this trip,
For I greatly desired
To see and also to know
The god of Love, who is so esteemed,
What kind of man he was and of what age.
As I traveled along on this excursion,
In peace, joy, and gaiety,
Singing a new motet
That had been sent to me from Reims,
I was neither in the front nor in the back,
14
Guillaume de Machaut, 164.
15
Ibid.
16
Figg & Palmer, Jean Froissart: An Anthology, 44461.
13
But very comfortably in the middle,
Dressed in new lace-up shoes,
The way lovers go for a late night out.
17
Sylvia Huot has argued that the unnamed motetclearly by Machaut since it came to the
singer from Reimsmay well have been Amours/Faus Semblant.
18
Here placed into the lips
of a happy lover who sings while walking, the motet (or one of it voices) acts as a sound-
track to an idyllic courtly scene in which the Vidi Dominum of the tenor refers to the
God of Love, upon whom the singer expects imminently to gaze.
19
But it is not to be. This
musical euphoria is all the more striking for the denouement which follows it. Someone
shoves me and then I wake up, the next line reads, and then the dream that takes up most
of the dit is over, and the lover is, in Sylvia Huots evocative summary, plunged back into
the realities of winter, encroaching age, the waning of desire and penitential concerns.
20
Thus the motet, whichever motet it is, here represents an extreme of emotion encapsulated
by joyful sound that is all the more loud for being suddenly interrupted.
In bringing together these various contexts and readings of Vitrys Tribum/Quoniam
and Machauts Amour/Faus Semblant, I have purposely mingled medieval and modern views
of motets. All of these interpretations t comfortably under the broad category of recep-
tion. Between them, motets emerge as intersections of texts and quotations, as mathemati-
cal structures and explorations of number, as exemplary representatives of a new notational
17
My emphasis; trans. Figg & Palmer, Jean Froissart: An Anthology, 4613.
18
Huot, Reading Across Genres, 2, 89.
19
Huot has suggested that given the poems broader narrative, the tenor contrasts the deceptive appear-
ances of the courtly lady with the unmediated and always salvic vision of God, ibid., 9. This interpretation
gives the motet broader signicance within the narrative, but it is also useful to read this as a moment of
diagetic, plot-driven music making in which the narrator sings out of joy, especially given the denouement
that follows, which in fact interrupts the song.
20
Ibid., 1.
14
system, as elaborations of their tenor texts, and even as spontaneous, joyful sound in a
dream. Nor is such a list exhaustive. That all of these different interpretations are possible
is a function of the complexity of motets. For they combine multiple voices, multiple texts,
and sometimes involved structural schemes with textual and melodic quotations in the up-
per voices and the tenor that lead necessarily outside of the work and into the historical
and intellectual context of its composition. Needless to say, all of these are valid critical
approaches (for even Froissarts decision to cite a motet at this particular moment is critical
and reveals a particular view of the genre), and motets have at some time been all of these
things, and continue to be all of them in the pluralistic climate of early twenty-rst-century
academia.
But in practice these various views of motets have some trouble coexisting. For ex-
ample, numerical approaches tend to minimize the role of text, owing to an attitudeoften
implicitthat mathematical construction results in absolute music which is unable or
unwilling to signify.
21
Similarly, approaches that begin with the tenors text and liturgical
context tend to interact less with upper-voice texts and quotations, except when these sup-
port the sacred message of the tenor.
22
Studies focused on texts and the roles of quotations
often do engage with structure, and especially with numerical symbolism, which is able to
render structure semantic by imbuing number with meaning.
23
But this approach depends
on an imaginary listener who can appreciate these numerical subtleties because he is in-
21
See the comments of Chailley and others quoted at the beginning of the introduction and published in
Apel, Remarks about the Isorhythmic Motet, 145. Other studies that focus on structure include Besseler,
Studien II, Sanders, The Medieval Motet, and Leech-Wilkinson, Related Motets.
22
E.g. Clark, Concordare cum materia, Robertson, Guillaume de Machaut.
23
For example, Bent, Deception, Exegesis, and Sounding Number, 15-27, and Roesner, Labouring in
the Midst of Wolves, 21245.
15
timately familiar with the work, and presupposing this kind of familiarity in an analysis
in turn gives a less active role to sonority and audible events. None of this is necessarily a
problem: any analysis must decide which factors are to be signicant and central, and which
peripheral. The alternativea reading in which text, number, liturgy, sonority, quotations,
culture, and contrapuntal progressions are all expected to signify threatens to render a
work innitely complicated.
Thus decisions must be made about what aspects of motets will be more signicant,
and these decisions will have ramications at every step of the scholarly and performing
process. How we edit motets, whether we translate their texts, how we group them in edi-
tions, what other genres we pair them with in concertsall of these aspects of our interac-
tions are predicated on attitudes about the nature of the genre and the relative importance
of its competing elements. Even names can be instructive. For example, the same motet
has variously been referred to as O canenda/Rex/Rex regum, Rex quem metrorum, O
canenda/Rex/T:Rex regum/CT, O canenda vulgo per compita-Rex quem metrorum de-
pingit prima gura-Rex regum, and V14. This is not just a matter of academic fashions
or of the disconnect between modern and medieval notions of what a title isas we shall
see, the fourteenth centurys naming conventions for motets were very consistent. But O
canenda/Rex/Rex regum encourages us to think of a composite entity the different ele-
ments of which are kept neatly apart by slashes. O canenda/Rex/T:Rex regum/CT with
its complete top-to-bottom listing of voices seems most concerned with indicating that
this is a four-voice motet, and with putting each listed voice on equal terms with the oth-
ers. O canenda vulgo per compita-Rex quem metrorum depingit prima gura-Rex regum
seems rather to be three musical works than one, and V14 evokes an easy-to-encapsulate
16
element of an oeuvre (here, Vitrys) ready to be compared to others like it (presumably
V1V13) and having little to do with any text that may or may not be present.
This variety of naming betrays a difference of opinion about what these works actu-
ally are. Are they networks of related compositions or individual creations? Is their mean-
ing contained in the long upper-voice texts or the tenors pithy content? To what extent do
the separate voices represent independent musico-poetic statements, and to what degree are
they linked? How dening a characteristic, after all, is text to this genre, and if text, then
which text? Where should untexted voices (such as contratenors) or silently texted voices
(most tenors) stand in our analytical priorities? What is a motet? It may be that not so much
is in a name, but these are the questions that arise from the difference between O canenda/
Rex/Rex regum and V14.
If our ways of editing, naming, and citing motets can reveal something about com-
peting modern notions of the essence of these works, certainly analogous medieval prac-
tices can do the same. Though no texts have survived that deal specically with the ques-
tions posed above, a broad range of sources comment upon the cultural and intellectual
presence of motets, hinting at the ways in which they were received. We can begin by consid-
ering naming conventions as they are revealed in manuscript indexes and treatises. These
give some compelling hints about where medieval composers or theorists may have located
the work within the complex of relationships that constitutes ars nova motets. I will then
look to the manuscript transmission of motet textsthose of the upper voices and of the
tenorfor what they can tell us about the relative importance of different texts. Next I will
turn to the poetic presence of motets. Citations, allusions, and even irreverent re-workings
of motet passages in dits and short poems attest to a role for upper-voice motet texts in a
17
broader literary culture.
Throughout these different arenas, we will see that the reception of motets in ars
nova circles emerges as relatively uniedespecially so in contrast with more peripheral
practices. For the priorities or scribes, theorists, and even poets are surprisingly dependent
on geography. Already in the fourteenth century, it seems, the ars nova motet was different
things to different people.
* * *
THE ONTOLOGY OF MOTETS
It may well be argued that the idea of naming a piece of music is a later invention,
and that we should not expect motets to be called anything. But insofar as the medieval
evidence can answer these questions, there were clearly conventions involved in referring to
motets. In almost all contexts, ars nova motets are referred to by the rst few words of their
motetus text. This is the case in both indexes and treatises for the rst three quarters of the
fourteenth century at which point, as we shall see, focus shifted to the triplum. I will discuss
the evidence provided by indexes and theoretical treatises in turn.
The Indexing and Manuscript Layout of Motets
The index to the interpolated Roman de Fauvel (c. 1317) gives its motets pride of
place among the musical examples and divides them into two categories. The three-voice
motets are labeled motez a treblez et a tenur[es]; the two-voice works, motez a tenures
sanz trebles. This designation may not seem intuitive. We can take it for granted that a mo-
tet has a tenor, and in keeping with the construction a modern editor would perhaps re-write
these headings to read tenors with motetus and triplum and tenors with motetus, without
18
triplum. But this would miss the fact that motet means both the genre of motet and the
voice motetus. The Fauvel headings seem to preserve this distinction. What makes a work
a motet, in other words, is not having a tenor with repeating rhythmic patterns or different
texts in different voices (though these are certainly properties of the Fauvel motets a treblez
et a tenur), but having a motetus. That all motets should be called by their middle voice is
perhaps most surprising in a piece like Tribum/Quoniam, where the triplum comes in rst
and declares its opening text clearly, but the motetus enters only later, echoing the triplums
opening notes and singing under it (see Example 1.1). Here the experience of hearing the
piece and of calling it Quoniam secta latronum seem rather different, and we may wonder
whether someone seeking to nd this motet would not be looking under Tribum que non
abhorriut. Nevertheless, even this work is indexed by its motetus.
Example 1.1: Tribum/Quoniam, mm. 16
24
The only exception in Fauvel serves rather to prove the rule than otherwise. The
motet Zelus familie/Ihesu tu dator is listed in the index as Zelus familiethat is, by the
incipit of its triplum voice. But upon examination of the music we nd that the voices are
almost identical in their rhythmic activity, rate of textual declamation, and range.
25
In-
24
Reproduced from PMFC 1:6.
25
Both voices declaim on the level of longa and breve, in a Mode 1 rhythm. Individual semibreves do not
carry text.
19
deed the motetus begins on top, delivering its Ihesu above the triplum in mm. 23 (see
Example 1.2). Granted, the triplum here is still the one with the smallest possible note-
values, which is a requirement according to theorist Johannes Boen: nor should a triplum
of some motet be called by that name (triplum) unless it has in it some notes having a triple
proportion to the notes of some, be it motetus or tenor.
26
And indeed the triplums notated
semibreves (as in m. 5) are on a rhythmic order higher than both tenor and the motetus for
most of the motet.
Example 1.2: Zelus/Ihesu mm. 16
27
However, towards the end of the piece the motetus suddenly begins to speak more
quickly, while the triplums rate of text-declamation crawls at times to a halt for three
to four longae.
28
By the nal measures the motetus Ihesu tu dator seems really to have
changed roles: it sings semibreves and carries the highest note in the works nal sonority
(see Example 1.3).
29
The scribes confusion in this case is thus justied. The listing of the
motet in the index suggests that he knew the opening texts of both voices and may have
26
Nec debet triplum cuiuscumque moteti vocari hoc nomine nisi sit in ipso aliqua notula triplam habens
proporcionem ad notam aliquam puta moteti vel tenoris, Frobenius, Johannes Boens Musica, 42.
27
Reproduced from PMFC 1:65.
28
See the passages beginning at mm. 56, 87, and 92 in PMFC 1:667.
29
This fact alone would not be sufcient justication for indexing the motet with a triplum incipit: Quant
je le voi/Bon vin doit the triplum ends lower than the motetus, but the motet is still listed in the index as Bon
vin doit.
20
brought his understanding of their musical functions to bear on his decision in making the
index.
Example 1.3: Zelus/Ihesu mm. 84end
30
Dated 1376 and thus standing near the end of our period of interest, the index of
the Trmolle manuscript also separates motets from other genres and, collecting them un-
der the rubric Motez ordenez et escriz ci aprs, lists the works by their motetus incipits.
31
Within the Machaut corpus, the same approach is taken in the famous index that heads
30
Reproduced from PMFC 1:67.
31
See Bent, A Note on the Dating, 2223. Bent has determined that the date of 1376 only acts as a
terminus ante quem for the rst part of the index and for the contents of folios 132. Even though most of
the manuscript has been lost, it can be ascertained from the many concordances which exist that all works
are indeed listed by their motetus. Droz and Thibaut assumed that the motet indexed as O Philippe was O
philippe/O bone dux, but is probably the Fauvel motet with the motetus O Philippe; see Droz & Thibaut, Un
chansonnier, 5. See also comments in Earp, Scribal Practices, 66. The Strasbourg codex treats Portio
nature as the triplum of the motet elsewhere presented as Ida/Portio, and Trmolles citation of this work
as Yda capillorum may rest on this conception; see Gnther, The Motets of the Manuscripts Chantilly, p.
lix. Here the distinction between voices is difcult to make, as it was in the case of Zelus/Jesu: the Ida capil-
lorum voice sings a long Ida at the opening of the motet, while the Portio naturae voice declares more
quickly. That there was some question about what this motet should be called can be gleaned also from the
Ars cantus mensurabilis, whose author normally refers to motets by their triplum voice but in one case calls
this work Portio nature vel Ida capillorum; Balensuela, ed., Ars cantus mensurabilis, 256. See also Appendix
1A.
21
Manuscript A. Here a list headed Les motets uses a motetus incipit for all the works but
two.
32
The rst exception is Quant/Amour (M1), which is entered into the index as Quant
en moy vint premierement, with a big decorated Q. Earp has suggested that the triplum is
cited by virtue of its reference to coming rst, but the decision to list it thus could also
be a nding aid, since this motet has a miniature above the triplum voices decorated Q, and
the eye would naturally be drawn to the left column on fol. 414v, and with it the triplum
Quant en moy. The motetus here has only a one-line initial. In all other cases, however,
the motetus has a two-line initial, while the triplum, tenor, and contratenor (where there
is one) have a one-line decorated letter (as in the beginning of He mors/Fine amour, repro-
duced in Figure 1.1.
33
Thus the manuscripts decoration is keyed to the index.
34
It is a strange aspect of motet reception that sometime during the last quarter of the
fourteenth century the naming convention seems to have changed suddenly and decidedly.
In the Machaut corpus, we may compare the priorities underlying the decoration of He
mors/Fine amour (M3) in Manuscript A (Figure 1.1 above) with the same motets layout in
the slightly later Manuscript Ferrell-Vog (Figure 1.2). Here and elsewhere in the motet
section of this manuscript, it is the triplum voice that begins with a 2-line capital, while
motetus, tenor, and contratenor parts have one-line capitals.
35
32
In contrast, Machauts three polytextual ballades are indexed by all of their texts: Sans cuer/Amis/Dame
(B17) and De triste cuer/Quant/Certes (B29) are each listed by three incipits, with brackets added on the left
to show the three texts go together. Quant Theseus/Ne quier (B34) was originally indexed as Quant Theseus,
but the hand providing pagination later added Ne quier veoir to the right of the rst incipit.
33
Observed in Earp, Scribal Practice, 65n146.
34
The other exception is M18, in which all voices begin with Bone pastor but the incipit given in the
index, Bone pastor Guillerme, refers to the triplum. Earp has suggested that the triplum is cited because
it names the works dedicatee; ibid., 67. It is also possible that the scribe had a momentary lapse given the
similarity of the incipits.
35
The only exception is Christe/Veni (M21), where 2-line-high initials in both upper voices allow a clever
trick in texting whereby both the introitus section and the beginning of the motet proper use the same deco-
22
Figure 1.1: The Start of He mors/Fine amour (M3) in Machaut A (fol. 416v)
Figure 1.2: He mors/Fine amour in Machaut Vg-Ferrel (fols. 262v263)
rated inital (fols. 280v281).
23
The change of priority demonstrated by these decorations is also reected in manu-
script indexes after Trmolle. Within the Machaut corpus, Manuscript E (copied in the
early 1390s) lists motets by their tripla. Chantillys index lists triplum incipits under the
category Motes (fol. 10), and the destroyed Strasbourg manuscript (early 15th c.) en-
ters motets by their triplum under alphabetical headings.
36
The same holds true for mid-
fteenth-century sources. In Ox 213 (c. 142636) the index is organized by alphabetic
headings and could in theory have accommodated multiple entries for the same work, listing
its different voices under different letters if it were likely that a user should have look for a
work by several of its texts. However this is not done, and each of the motets and polytextual
rondeaux in the chansonnier is listed in the index only once, by its top voice.
37
Similarly,
the section headed Hic Incipiunt Motteti in the index to ModB (c. 144050) gives a
textual and musical incipit for the highest voice of each motet in the manuscriptusually
the triplum, though sometimes a quadruplum.
38
These later collections largely transmit a
separate repertory, but there are enough ars nova motets present for us to be sure that the
naming conventions have indeed changed, and not only with respect to newer works but
also retrospectively.
39
36
Copied in Vander Linden, ed., Le manuscrit musical, 1625. Both texts of Ida/Portio are listed in the
indexperhaps the indexer thought it was two compositions.
37
Polytextual motets whose top voices are found in the remaining part of the index are catalog nos. 51,
68, 267, 275, 277, 279, 308, and 321. Polytextual rondeaux listed by their top voice are nos. 202, 208,
219, 254, and 284. No. 77 may be a polytextual ballade or notit is also listed by its top voice/rst verse;
see Fallows, ed., Oxford, Bodleian Library MS. Canon. Misc. 213, 36.
38
On the ModB index and its category of motet see Cumming, The Motet, 48, 514. Q15 also groups
motets together but the indexer does not include them in his list. See Bent, Bologna Q15, 1:8990.
39
Degentis/Cum vix, Ida/Portio, Lardure/Tres dous, Tant/Bien, and Apta/Flos are concordances between Tr-
molle or Ivrea and Chantilly, and there are several more concordances between the earlier repertoire and
Strasbourg.
24
Motet Citations in Theoretical Treatises
There is a large number of references to specic ars nova motets in music treatises
from the fourteenth and fteenth centuries. Though these have been mined for what they
can tell us about motet authorship and music theory, there is good reason to explore the
rhetorical habits employed by theorists in citing pieces: in referring to particular works of
music, theorists stand to provide valuable hints about the ontology and anatomy of motets
as they were represented in the minds of their early audiences.
References to ars nova motets drawn from 16 treatises are collected in Appendix
1A. The rst 55 or so of these belong to the period before c. 1375. As is the case with
manuscript indexes from this time, theorists identify motets by their motetus text almost
without exception.
40
Thus the author of De musica antiqua et nova, when he wishes to refer
to Cum statua/Hugo, writes in moteto qui vocatur Hugo, and the Compendium totius artis
motetorum talks of uno moteto Praesidentes in tronis seculi, by which we are to understand
Super cathedram/Presidentes.
41
This consistent use of the motetus incipit is all the more tell-
ing in the presence of formulations such as in moteto qui vocatur..., which give the sense
that these incipits functioned as titles. The motetus incipit/title could be employed when a
writer wished to refer to the entire piece (such as in the common cases where a motet serves
as an example of a certain mensuration). But it was also used to reference a specic voice or
40
This tendency has been noted by Besseler, Studien zur Musik des Mittelalters II, 2356, Earp, Scribal
Practice, 656, and Bent, A Note on the Dating, 2234. The three exceptions are the Ars (musicae)
citation of Rex/Leticie as Rex Karole, the citation of Orbis/Vos as Orbis orbatus in the treatise which
begins Sex minimae possunt poni, and the reference to Qui es/Ha Fortune (M8) as Qui des promesses de
fortune se e in the treatise Cum de signis temporibus... (see Appendix 1A). The latter two treatises are
in fact later (possibly early fteenth-century) copies of ars nova-era texts, and the scribes may well have been
supplementing their exemplars with new citations, using the naming conventions that were current when
they were copying. The rst exception, Rex Karole, appears only in a reworked version of Boens treatise;
see Boen, Ars, 14.
41
Coussemaker, Scriptorum, 3:347; Wolf, Ein anonymer Musiktraktat, 37.
25
passage in either of the upper voices. Thus in his Musica Johannes Boen uses the formula-
tion in motheto Florens vigor super verbo Mardocheo to identify a particular dissonance
that occurs on the word Mardocheo in the motetus text.
42
But the motetus incipit could
also be used to reference the triplum. For as Boen is explaining the rules of alteration and
imperfection in his Ars, he offers Impudenter/Virtutibus as an example of the similis ante
similem... rule and of the inability of breves to alter minims:
No note before a like note takes on imperfection. And this we can observe
clearly in the rst four little notes of that most excellent motet Virtutibus.
From this we can infer that a minim is never altered before a breve, nor a
semibreve before a long.
43
Although the motet is cited by its motetus like Boens other examples, the primis
quatuor notulis must belong to the triplum, which is the only voice singing at the begin-
ning of the motet. Its rhythm there (---a) illustrates both of Boens points, since the rst of
the two semibreves is perfect and the minim is not altered by the breve.
44
Toward the end of the fourteenth century (and earlier in England and Italy) the
fashion for naming switched to the triplum. Since the word motetus still referred to the
genre, this led to some seemingly contradictory statements, such as the reference in the
Tractatus Figurarum in motetis ipsorum magistrorum videlicet Tribum que non abhorruit
et in aliis (to the motet(use)s of those old masters, such as Tribum que non abhorruit, and
others). The old masters would of course have referred to this work as Quoniam secta la-
42
Frobenius, ed., Johannes Boens Musica, 68.
43
Similis ante similem nullam capit imperfectionem. Et hoc in primis quatuor notulis illius excellen-
tissimi moteti Virtutibus clare possumus contemplare. Ex hoc subinferri potest, quod numquam minima
alteratur ante brevem, item nec semibrevis ante longam, Boen, Ars (musicae), 26.
44
When the motetus enters it is with the rhythm - --a, which does not illustrate either point. Neither
does the tenor whose rst four notesa a in imperfect moduspresent no problems of alteration or
imperfection.
26
tronum. Now the term motetus paired with the triplum incipit could refer to the triplum
or motetus voice, as well as to the motet as a whole. For example, in the fteenth-century
literary miscellany F-Pn lat. 3343 the triplum text of Petre/Lugentium is transmitted with
the annotation Hunc motetus fecit Philippus de Vitriaco pro papa Clemente.
45
Elsewhere
in the same miscellany, the texts of all four voices of the motet Phi millies/O creator are la-
beled as Triplum, Motetus, Tenor, and Contratenor.
46
Across this change from motetus to the triplum as the voice of reference, one as-
pect of naming stays constant: when referring specically to the tenor voice, theorists use
an upper-voice incipit but specify the tenor. Thus the author of the Compendium totius artis
motetorum writes in tenore de In arboris to cite the tenor of the motet Tuba/In arboris.When
the tenor of the same motet is cited in the early fteenth century by the anonymous Bre-
slau author, the citation reads in tenore Tube sacre dei.
47
After what we have said about
indexes and theorists in general, this should not be surprising. But it is worth noting that
neither author wrote in tenore Virgo sum he did not give the text of this tenor, though
it is unique in the repertory of motet tenors and would thus have been able to work as a suf-
cient identier of this particular voice.
Despite the uniqueness of many motet tenors, the tenor text was never used by ars
nova theorists to refer to any part of a motet. This simple fact is potentially meaningful,
though it is not clear exactly what it means. Possibly the theorists themselves did not know
the texts of the tenors. Or more likely they did not count on their readers to know them,
45
Earp has noted this as a contradiction, but it need only be one if we read motetus as the name of the
voice part rather than the genre; Scribal Practice, 66. See also ibid. for another example of a motet cited
by its triplum in the Illustre Lieve Vrouwe Broederschap (c. 1423).
46
Wathey, The Motets of Philippe de Vitry, 126n16.
47
See Appendix 1A. Insofar as I know this is the only case in which a motet title is declined.
27
which would make citing them in a treatise inexpedient. Or even if tenor sources were com-
monly known, it is possible that theorists did not consider them texts in the same sense
as upper-voice texts, since they never sounded in performance. Indeed, a passage from
the seventh book of the Speculum musicae conrms that tenors were not thought to carry
text. Explaining that some kinds polyphony have one text and some, several, Jacobus cites
cantilenae as an example of the former. And as for a polytextual work, his example is not
simply a motet, but a motet with a triplum:
Some of the discanting [voices] are with the same words, and some with dif-
ferent: the same [words] in cantilenae and other diverse ecclesiastical songs;
different, in motets with a triplum.
48
There is even some evidence that the tenor voice, since it belonged to plainchant,
was less fully a part of the motet than the upper voices. For example, Johannes de Muris, in
discussing the terms color and talea, makes a distinction between the tenors of motets and
motets themselves:
The placing of one series of similar rhythms repeated several times in the
same voice is called color. But note that some singers make a distinction be-
tween color and talea, for they call it color when the same notes are repeated,
48
Discantuum aliqui sunt cum littera eadem vel diversa: eadem, ut in cantilenis et aliquo cantu ecclesias-
tico; diversa, ut in motetis triplum habentibus, Bragard, ed., Jacobi Leodiensis Speculum musicae, 25. This
distinction is preserved and expanded by the author of De musica antiqua et nova, who however seems to
qualify the comment with the addition of the phrase in quibus tenor equipollet littere, which may mean
that the tenor is of equal value to a text. Nevertheless, his example of a polytextual work is still a motet with
triplum: Modus operandi in discantu talis est: aut enim discantus cum littera aut sine; si cum littera, hoc
est dualiter: aut cum eadem littera discantus t ut in cantilenis, rondellis et in quodam cantu ecclesiastico;
aut cum diversis litteris t discantus, ut in motetis qui habent triplum cum tenore in quibus tenor equipollet
littere. [The manner of executing polyphony is as follows: discant is either with words, or without; if with
words, these are of two kinds: either the discantus is made with the same words [in all parts], as occurs in
cantilenae, rondeaux, and in some liturgical songs; or the discantus is made with diverse texts, such as in
motets which have a triplum with a tenor, in quibus tenor equipollet littere.] Coussemaker, Scriptorum, 3:361.
See also Aluas, Quatuor principalia, 521 and 747, where the nal line is translated in which the tenor is the
equipollent of the texts.
28
but talea when the same rhythms are repeated and thus make diverse notes.
This difference, although it may be observed in a great many tenors of motets, is not
observed in the motets themselves.
49
What Muris seems to be saying is that the distinction between color and talea disap-
pears in the upper voices, probably because here only rhythms but not pitches are repeated.
Thus the two different kinds of repetition can only be observed in tenors and not in motets
as a whole. For this reason the author prefers to call both kinds color. Muriss distinction is
preserved in the third treatise of the Berkeley manuscript.
50
Keeping in mind that there may be an ontological distinction between motetis ipsis
and motetorum tenores can help nuance our reading of the oft-quoted Tractatus cantus mensu-
rabilis of Egidius de Murino. This text has been interpreted as detailed instructions for the
composition of motets and a precept on motet composition.
51
In fact, the treatise is titled
De modo componendi tenores motettorum (On the manner of composing motet tenors).
52
Thus when the instructions begin rst take the tenor from some antiphon or responsory...,
we can by no means assume that the process of composing a motet begins with selecting a
49
Emphasis mine. Color in musica vocatur similium gurarum unius processus pluries repetita positio
in eodem cantu. Pro quo nota, quod nonnulli cantores ponunt differentiam inter colorem et tallam: nam
vocant colorem, quando repetuntur eedem voces, tallam vero, quando repetuntur similes gure et sic unt
diversarum vocum. Que differentia, licet servetur in quampluribus tenoribus motetorum, non tamen serva-
tur in ipsis motetis, Berktold, ed., Ars practica mensurabilis cantus, 78.
50
Que differncia licet in quampluribus motetorum tenoribum observetur, non tamen observatur in ipsis
motetis, ut in eis liquidem est videre, ed. and trans. Ellsworth, The Berkeley Manuscript, 1803; Ellsworth
translates the line in question This differentiation may be observed in a great many tenors of motets, but
not in the motetti themselves. This would imply that the author has shifted from motet as genre to motet
as voice over the course of one sentence, which is unlikely.
51
Leech-Wilkinson, Compositional Techniques, 18; Robertson, Guillaume de Machaut, 146.
52
It is unclear why it is sometimes referred to as De motettis componendis, for example in Reany, Egidius
[Aegidius] de Murino [Morino], and consequently in a number of general reference works such as Randel,
ed., The Harvard Biographical Dictionary of Music, 241, and Kibler, ed., Medieval France: An Encyclopedia,
316. Coussemakers title is Tractatus Cantus Mensurabilis and his subtitle is De modo componendi te-
nores motettorum; Scriptorum, 3:124. The latter seems to be the correct title of treatise when on its own.
In several sources it is appended to the Tractatus gurarum; see Lefferts, ed., Regule, 72n175, and Schreur,
ed., Tractatus gurarum, 67.
29
section from plainchantthough certainly the process of composing a motet tenor must
often have begun there. Murino explicitly states that the tenor is chosen in accordance with
some unspecied but pre-existing materia, with which its words should concord (et debent
verba concordare cum materia de qua vis facere motetum), but this feels like an aside.
53
In Clarks estimation, the theorist seems uninterested in the issue of materia: it is easy
to overlook this statement concordare cum materia, and perhaps even Egidius is more
interested in talea and color formation.
54
Indeed he is, since this is both the stated goal of
his treatise and a personal interest: we know from Apollinis/Zodicaum that Murino had a low
voice, and thus probably sang tenor or contratenor in motets.
55
But reading between the lines, we can lean that some materia is already decided
upon when the tenor is chosen, and that indeed the texts of the upper voices may already
be written. (Like reading a recipe from the middle, after the ingredients have been already
measured out, we dont know where these texts are to be gotten, or what has been done to
them before thisthe cook is simply instructed to take the words which are to be in the
motet and divide them into four parts.)
56
Since the texts of motets are often arranged ac-
cording to their isorhythmic structures, the possibility that they exist before the tenor has
many implications for the role of the upper voices in determining forma point which will
be addressed in Chapter Three. For now, it will be enough to note that Murisa distinction
53
Edited in Leech-Wilkinson, Compositional Techniques, 18.
54
Clark, Concordare cum materia, 12.
55
Egidius de Morino baritonans cum Garino, triplum ll. 223. Gallo reads this as Egidius de Murino
singing tenorista with Garino, Music of the Middle Ages II, 127. Henricus Helene is identied as one who
noscittonorum tenorem bene, but this is translated as he knows well the tenor of the tones (Bent, Two
14th-century Motets, 10) and knows well the characteristics of the ecclesiastical modes (Gallo, Music of the
Middle Ages II, 127). Other singers are indicated as having higher voices, such as Arnold of Martin who is
compared to a nighingale.
56
Ibid., 22.
30
between motetis ipsis and motetorum tenores is preserved both in the organization of Murinos
treatise and more broadly in the habits that theorists exhibit when citing motets: while
either of the upper voices can be referenced by the incipit of one of them, tenors seem to
stand rhetorically apart.
This is not to say that the tenor was considered unimportant. Johannes de Gro-
cheios turn-of-the-century comparison between the tenor of a motet and, on the one hand,
the foundation of a house, and on the other, the skeleton of the body, was echoed by numer-
ous later writers.
57
For example, the Quatuor Principalia indicates that singers should sing
the tenor articulately so as to avoid dissonant mis-alignment with the upper voices, then
appends the familiar architectural comparison:
It should be known according toall musical singers that the tenor which
holds the discant should be sung well and strongly to the beat, lest those
discanting above should meet with a dissonance, and thus [musical] sense
should be expelled. For a stable building cannot be constructed upon an
unstable foundation, nor can a discant be performed over an unstable tenor
without dissonance.
58
Jacobus de Montibus invokes the same idea in the seventh book of the Speculum miscae.
In the course of providing a possible etymology for discantus, he suggests that the word
comes from de (from) and cantus (song), since polyphony indeed arises from a song
namely, the tenor:
59
57
Tenor autem est illa pars, supra quam omnes aliae fundantur, quemadmodum partes domus vel aedicii
super suum fundamentum. Et eas regulat et eis dat quantitatem, quemadmodum ossa partibus aliis, ed.
Rholoff, Die Quellenhandschriften, 1467. The implications of this foundational metaphor for our percep-
tion of upper-voice structures is discussed in Chapter Two.
58
Sciendumomnes musicales cantores quod Tenor qui discantum tenet integre et solide pronunciari
debet in mensura ne supra discantantes dissonantiam incurrant, et hoc ratio exigit. Nam super instabile fun-
damentum stabile aedicium construi non potest, sic nec instabilem tenorem vix sine dissonantia discantus
pronunciari potest, De musica antiqua et nova in Coussemaker, Scriptorum, 3:362.
59
Vel potest dici discantus a dy quod est de et cantu, quasi de cantu sumptus, idest de tenore, Bragard,
ed., Jacobi Leodiensis Speculum musicae, 7:9.
31
The discant is established upon the tenor, somewhat like a building upon its
foundation; and hence that voice is called tenor, since it holds (tenet) and
upholds the discant. For what man sings polyphony without a tenor; what
man builds without a foundation?The discant is derived from the tenor,
by which it should be should be ruled, and with which it should be in har-
mony rather than in discord.
60
But directly following this clear statement about the tenors harmonic dominance and foun-
dational nature, Jacobus turns things on their heads, explaining that the upper voices are as
much a song as the tenor, and hence the cantus in discantus may well refer to them:
Not only is the tenor set under the discant, but also the opposite; for the
discanting voices can either be compared to the tenor with which they are
obliged to concord (and hence this voice is called discantus), or they can be
considered by themselves: not with respect to the notes of the tenor and sung
at the same time as them, but separately and in turn, one after the other, as
when someone by themselves sings some motetus or triplum or quadruplum
without a tenor, and then the whole work (et tunc absolute). Such notes have
the grammatical sense of an independent song; and in the same way it is said
of the notes of the tenor, that they can be compared to [those of] discantus
and sung together with it, or sung separately on their own.
61
There is much to wonder at in this passage. Directly following his traditional statement
about the tenor as a foundation and source of origin for the upper voices, Jacobus now ex-
plicitly states that the upper voices are indeed independent entities, as much a song as the
tenor, and hence not really dependent on the tenor at all. He ends by comparing the tenor to
the upper voices, and using their independence as paradigmatic of the tenors independence.
Things seem to have gone full circle.
60
Idest de tenore supra quem discantus fundatur, sicut aedicium aliquod supra suum fundamentum; unde
ille cantus tenor dicitur, quia discantum tenet et fundat. Quis enim sine tenore discantat, quis sine funda-
mento aedicat?Discantus igitur a tenore dependet, ab eo regulari debet, cum ipso concordare habet, non
discordare, Ibid.
61
Non tenor de discantu sumitur, sed e converso. Possunt autem voces discantus ad voces comparari te-
noris cum quibus debent concordare, et tunc talis cantus discantus dicitur. Vel possunt per se considerari,
non per respectum ad voces tenoris et ut simul dicuntur cum illis, sed divisim successive una post aliam cum
aliquis per se cantat motetum aliquem, triplum vel quadruplum, sine tenore et tunc absolute. Tales voces
rationem cantus habent. Et eodem modo dicendum est de vocibus tenoris ut ad voces comparantur discantus
et simul cum illis dicuntur, vel ut per se et divisim decantantur, Ibid.
32
And then there is Jacobuss example of this independencea practice that seems to
have been known to his readers, thus requiring no more than a passing reference: as when
someone by themselves sings some motetus or triplum or quadruplum without a tenor, et
tunc absolute. This glimpse of a sort-of dissection of motets constituted in the performance
of the upper voices on their own rings true with the other aspects of motet reception so far
discussed and gives some hint about one way of overcoming the difculties of a polytextual
texture (though more about this in Chapter Two). Signicantly, Jacobus does not indicate
that a motet tenor would be sung on its ownhis assurance that the notes of a tenor can be
sung separately on their own is back in the range of polyphony more broadly dened, or
indeed of discant specically. Certainly the tenors of discant had their own lives as plain-
chant. But I believe that it is no coincidence that when he wants to give the upper voices
their full due Jacobus switches from talking about discant in general to motets in particular.
The upper voices of motets, with their separate texts, did indeed represent a new level of
independence for discanting voices.
THE TRANSMISSION OF TEXT
Just because a theorist cites a motetby whatever voicedoes not mean that he
knows its text. Indeed, a few of the incipits listed in Appendix 1A are already corrupt.
Christopher Page has drawn attention to this, arguing that the decay rate for even the most
obvious verbal subtleties in these pieces was very rapid once the piece moved away from
the composer and his immediate circle.
62
As an example, he notes that the words In nova
fert animus at the beginning of Vitrys Garrit/In nova are distorted in one manuscript of a
treatise containing Vitrys teachings, becoming In nova sit animus. Although such a slip may
62
Page, A Reply, 130.
33
reveal only a commonplace ignorance quite beneath the small circle of connoisseurs for
motets, we could also regard it as signicant evidence for an indifference among medieval
listeners to some of the kinds of subtlety that modern critics advocate, and therefore as a
cautionary sign that those advocacies may be in some degree misplaced.
63
The treatise in question was copied around 1400 by an Italian scribe who also mis-
spelled Vitrys name as Vetri.
64
Nor can it be said to contain Vitrys teachings, although
it shares a source with a fragment of a treatise that does.
65
But Pages larger point is well
taken. Certainly the relative importance of various aspects of composition may be gleaned
from transmission insofar as scribes themselves were members of the circles in which mo-
tets were written or performed.
66
Furthermore, differences between central and more pe-
ripheral sources relative to a works original context may help to reveal, if not indifference,
then shifts in priorities.
Upper-Voice Texts
To see what can be gleaned from this sort of evidence let us consider the variants of
textual transmission and underlay for Apta/Flos. This motet is an especially good candidate
for such an analysis because survives in four legible sources that range widely in provenance
and date. Its Durham source is from the mid-fourteenth century and thus the earliest, but
its English provenance renders it the most peripheral. The other three sources are French
63
Ibid., 1301.
64
Fuller, A Phantom Treatise, 33.
65
Fuller has characterized it as a potpourri of conventional topics copied form a defective exemplum,
Ibid., 24.
66
Though it should be stressed that the care or negligence displayed by scribes in the copying of motets does
not indicate anything about authorial intentions, or indeed about performance: scribal errors, especially
minor ones, do not guarantee errors in performance. See the discussion of scribal intention and the status
of variants as evidence in Bent, Text Setting, 2923.
34
or Italian, but later: Ivrea copied c. 1390; Chantilly from the following decade; Modena
A from c. 1410.
67
These four sources will thus allow us to ask questions about textual cor-
ruption as a function of time and of place. Furthermore, Apta/Flos has a notoriously dif-
cult text: A. G. Rigg has written that the exact translation of the [triplum] is doubtful at
all parts, and the syntax is obscure.
68
This motet thus represents an extreme test-case for
the transmission of verbal subtleties. Finally, Apta/Flos is conducive to exploring different
kinds of textures, since it includes syllabic texting as well as hockets and melismatic writing.
Figure 1.4 compares readings for the rst 37 breves of the triplum voice, and Figure
1.5 for the entire motetus.
69
Not surprisingly, differences in spelling between the sources
are common, especially as regards doubled letters (barum vs. barrum), c for t, (segnities/
segnicies), and so forth. There is also one instance of metathesis: cabarsis for carbasis in
Ivreas triplum. Textual corruptions that could threaten sense are few. The most serious one
is in Ivreas version of the motetus, where multa byssus seems to have become non multa
abissu or perhaps abissusit is difcult to tell. This is the only case in which a word has
been added in any of the sources. No words are missing. Ivrea has indiga (needy) where the
three other sources have indigna (undeserving), though Rigg suggests that the Ivrea read-
ing makes more sense.
70
Regarding the nal word in the segment of the triplum reproduced
67
Two more sourcesCaB and SL2211are illegible, or nearly so. The motet is also listed in the index
to Trmolle.
68
PMFC 5 supplement, 5.
69
In order to represent the alignment of text and music as faithfully as possible, the relative spacing and
size of text and notes have been adjusted. Thus when a word is written with spaces (c o n s u m i t o, etc.) it
is not because the letters are spaced out in the source, but because a large number of notes are squeezed in
over this word in one source, but not in others. Scribal abbreviations and ligatures are indicated with italics,
and boxed areas represent passages which will be analyzed in more detail below, and for which facsimiles are
provided.
70
PMFC 5 supplement, 5.
3
5
a-----------a-a-----..-- - -..- - - ---..----- - - --a. - -- - - - - - --a..-..
AP ta caro plumis ingenii desidie barum et studii Laboris que foco moli ci es |
A apta caro plumis ingenii desidie barrum & studii La borisque foco mollici|es
[ ]P ta caro plumis ingenii desidie barum se studii La borisque foco molici es
AP ta ca ro plu mis ingenii desidie bar rum & studii L a b o r i s que foco mollicies
Iv
Mod
Ch
Dur
-- - - - -..-..- - - - - -.. - - - - - - --- --------- - - - - - - --. - - - -
et c o n i u g a centro s egni ties quo piscressit plumbm consumi to nablum tangevocis plus so lito duc pulmonis |
coniuga centro sengnities quo pigrescit plumbum consumito nablum tange vocis plus solito |duc pulmonis
et coniu|ga centro segnicies quo pigrescit plumbum consumito nablum tange vocis plus solitus duo pulmonis
& coniunga centro segnicies quo pigrescit plumbum c o n s u m i t o nablum tange |vocisplus solito duc pulmonis
Iv
Mod
Ch
Dur
Figure 1.4: Apta/Flos, mm. 137; a comparison of triplum text underlay in four sources
- - - - - - .. - - --..- - - - - - ..-..----..-..-..- - - - - -.. - - - -..-- - - - --..- -
cabarsis cardinem lingue lini gutris arundinem tua d ? i ? a l i t u r a m littere quam conetur in preces fundere iusto
carbasis cardinem lingue lini guctris arundinem tua clio licturam lictere quam| conetur in preces fundere iusto
carbasis|cardinem lingue lini gutris arumdinem tuam dio licturam lictere quam conetur in pre ces fundere visco
carbasis cardinem lingue lini guttris arundinem tua c l i o l i t t u<> ram| l e? quam c onet ur in perces fundere visco
Iv
Mod
Ch
Dur
Iv Ivrea, fol. 5v6
Mod ModA, fol. 18v19
Ch Chantilly, fol. 60v61
Dur Durham, 338v339
facsimiles in Figure 1.10
| line-break
? unclear text
<> missing notes
3
6
aa-------------a---------a.a.a.-a - - - - - - - - a. a - a- - a - a -- -------------------
Flos virginum decus et species adultere lucis con<>nubio |non in
F Los virginum decus et speties Adultere| lucis connubio non In di
fLOS Virginum decus & species a dul tere lucis co nu bis non in
FLos v i r g i n u m d e c u s & species adultere lucis connu bi o non in di ga
Iv
Mod
Ch
Dur
------- -- --a - - - ---- - - - -a.- - a - . aaaa- - -a - - - - -a. - - - a --a - ----a -
dig na us auri radio si ge marum vesperescat dies te surgente de cor ipse quo nichil gra|tius & agrada gratie l i n e a que
ga vel auri radio Si gemmarum vespera|scit dies te surgente decor ipse quo nichil gratius Et agrada gratie linea que|
diga |vel auri radio Si ge ma rum verperescit dies te surgente de cor ipse quo nichil gracius Et agrada gra|ci e li nea qua
vel au ri |radio si gemmarum vesperescit dies te s u r g e n t e d e c o r ip se q u o nichil gratius &a g r a d a gratie l i n e a q u a |
Iv
Mod
Ch
Dur
-- ----- - - - - ------- - - - a - - - --a. - - a - . a aaa- - - a - - - - - a
st?ix atrum sol scalor t r a b e a limus ostrum quo tinxit ti ri us t candente non| m u l t a a bis su ? virtutam calcu lo
stix a t r u m sol scolor trabea limbus ostrum quo tinxit tyrius t ca den te |multa byssus virtutum calcu lo
stix a t r u m sol scalor tra be a Lumbus ostrum quo tinxit tirius sit candentem.Mul | ta bissus virtutum calcu lo
stix a t r u msolscalor tra be a l i m b u s ostrum quod tinxit tirius sit candente m u l t a a bis s u s virtutum cal|culo nu-
Iv
Mod
Ch
Dur
. - - - a - - a - ---a-- - - - - - - - ---- ------- - - - - a - - -----a. - - a - --.
nu- mer orum non certa li a nati mater patris p l u e pia la crimarum|nimbos pro po p u l o te p o s c e n t e.
nu me- rorum non certa li a Nati mater patris plue |pia Lacrimarum nimbos pro popu lo te poscente.
nu- merorum non certa li a nati mater patris p l u e pia Lacrimarum|nimbos pro popu lo te p o s c e n t e :
m m o r u m non certa li a n a t i mater p r i ? s p l u e pia l a c r i m a r u m nimbos pro p o p u l o te p o s c e n t e.
Iv
Mod
Ch
Dur
Figure 1.5: Apta/Flos, comparison of motetus text underlay in four sources (abbreviations and symbols as in Figure 1.4)
facsimiles in Figure 1.9
facsimiles in Figure 1.8
37
in Figure 1.6, the sources are evenly divided between iusto and visco, as they are about
whether the motetus should read t or sit before ca(n)dente(m) (Figure 1.5, middle
of third line). Finally, two scribes seem to have had problems with the name of the muse
Clio in the triplum (Figure 1.4, middle of line 3). Perhaps not expecting the reference, the
Chantilly scribe has read clio as dio, while Ivrea has something illegible at this point.
71
These are all of the signicant variants that I have noticed, and some ambiguities
in Ivrea may be the fault of the microlm rather than the scribe. Certainly these kinds of
differences should give us some pausewe might well object, for example, to an argument
that the whole motet is focused around the presence of Clio, which had special signicance
for listeners in 1380s Ivrea.
72
But overall the readings are remarkably similar, especially
given the difculty of the text. Broadly speaking, I do not see here too many examples of
the kind of corruption that would preclude consideration of texts as analytically signicant.
This in itself is interesting when we remember the temporal and geographic scope covered
by the sources, which indeed take us far beyond the composer and his immediate circle.
A more interesting set of differences is to be found in comparing these four scribes
approaches to text underlayor really, I should say with regard to note overlay, since in
all four cases words were clearly copied before notes.
73
This order of operations makes the
most sense where text-setting is syllabic or mostly syllabic. But an ars nova motet such as
Apta/Flos also includes passages of melismatic writing and of hocket, and it is here that
alignment becomes a more complicated matter. For if in syllabic textures words take up
71
On the problem of illegible proper nouns (and proper nouns born of illegibility) in Chantilly, see Dulong
& Sultan, Nouvelles lectures, 1057.
72
On the circumstances surrounding the copying of Ivrea, see Kgle, The Manuscript Ivrea, 4679.
73
This seems to have been true of ars nova motets more generally. On the order of copying in the Machaut
manuscripts, see Earp, Scribal Practice, 17894.
38
more space than notes, the reverse is often true in hockets and melismas, so that space must
be left between some words and syllables, but not others. A look at the beginning of the mo-
tetus in Chantilly is instructive (see Figure 1.6). The scribe has left a space for the opening
melisma on osthough almost too small a space, as the crowded notation showsand
also later on non and in- at the end of the line. The nal syllable of the line is also fol-
lowed by a melisma, and this too is crowded in. Clearly some calculation and experience was
required to achieve the correct text underlay in this repertory, and even this expertise would
not guarantee even note spacing.
74
Figure 1.6: The opening of the motetus of Apta/Flos (Chantilly, fol. 61)
A different approach to the problem is taken by the Durham scribe, whose rendi-
tion of the same passage is reproduced in Figure 1.7. In writing the text he left no spaces
between words, with the exception of the introitus on Flosand even here, not enough.
Thus Virginum begins six breves earlier in the Durham source, where it is underlaid to
breves separated by rests.
Figure 1.7: The opening of the motetus of Apta/Flos (Durham, fol. 339)
74
Elizabeth Randell Upton has critiqued editorial reliance on scribal text underlay as based on the as-
sumption that a medieval manuscript is a score in the modern sense, arguing that the Chantilly scribe in
particular was often more interested in the symmetry of page-layout resulting from text-placement than he
was in music overlay, Aligning Words and Music, 115, 1213. However this evaluation of scribal practice
does not take the motets into account, for example when Upton cites Armes, amours as exhibiting some of
the clearest and most unambiguous word/tone alignment in the entire manuscript (118).
39
A look back at Figures 1.4 and 1.5 shows that this example is characteristic of the
Durham scribes less precise approach to underlay. While Ivrea, Modena, and Chantilly
agree in essentially all details of text underlay for the syllabic parts of the motet, the notes
in Durham phase in and out of synchrony with their texts. As a result, words or parts of
words are sometimes underlaid to rests, and the alignment of many passages is unclear.
And yet the Durham scribe does not bungle things completely. Especially in the more syl-
labic triplum, his underlay is actually quite close to the other three sources. Furthermore,
it is clear that he copied from an exemplar that had good underlay, since the line-breaks in
Durham leave nothing to be desired and align perfectly with the text set in other sources
for those passages. Such would not have been the case if Durham were copied from a source
like Durham. In any case, we do arguably see a different set of priorities at play here, and
it would be telling if other insular sources of ars nova motets were equally cavalier (well
discuss one more below).
Differences in text-setting among the other three sources are limited to melismatic
passages and hocket. In melismas, syllable placement varries slightly from source to source.
See, for example, the word lia, in the passage from the motetus represented in Figure
1.8. Here Modena gives a semibreve to -, two ligated semibreves to -li-, and the rest
(a--) to -a, while Chantilly gives two semibreves in ligature to , one to li, and, again,
three notes to -a. In both cases the text-setting is unambiguous. Ivrea seems to have the
same text-setting as Chantilly, also with a ligature on -, though it is less clear because
the spacing is more cramped. Only Durham does not allow three notes to -a, and this
seems less a deliberate choice than a function of the more inexact approach already men-
tioned. (Strangely, PMFC follows the Durham texting.) The other two melismatic words
40
(a) Ivrea:
(b) Modena:
(c) Chantilly:
(d) Durham:
Figure 1.8: Apta/Flos, motetus mm. 7482; text placement in four sources
(a) Ivrea:
(b) Modena:
(c) Chantilly:
(d) Durham:
Figure 1.9: Apta/Flos, motetus mm. 3245; text placement in four sources
41
in the passage, plue and pia, are also clearly texted in Modena and Chantilly. Ivrea does
not indicate the switch from one syllable to another, but does clearly group the appropriate
notes over each word. Overall, then, the differences in underlay in the passage in question
are tiny, and this in itself is signicant. Slightly more divergence can be found in the melis-
matic section between connubio and indiga in the motetus, but here too the divergences
are fairly small, since short note-values are involved (Figure 1.5, end of rst line).
The four sources of Apta/Flos also differ in their approaches to texting hockets. The
motetus passage represented by gure 1.9 is indicative. Only Durham makes no allowance
for the hocketed notes, writing them over the second half of surgente and decor. The
other three sources accommodate the hocketed notes, but demonstrate two approaches:
Ivrea and Chantilly (Figure 1.9a and c) both leave space by splitting up the word decor,
thus setting the hocketed notes to de-. Modena, on the other hand, leaves space between
surgente and decor, so that no word is split by the rests in the hocket (Figure 1.9b).
Similar discrepancies can be seen when we compare readings for a triplum passage that
mixes syllabic declamation and hocket (Figure 1.10). Here Modena makes it clear that a
four-syllable phrase (tua[m] clio) is surrounded by untexted hockets on either side, which
separate these words from others in the text (Figure 1.10b). Durham has a similar read-
ing, though less precisely aligned (Fig 1.10d). The Chantilly scribe has also left space after
tuam and clio, but seems to set the former to the two hocketed minims (Figure 1.10c).
Ihe Ivrea reading is similar to Chantilly (Figure 1.10a).
42
(a) Ivrea:
(b) Modena:
(c) Chantilly:
(d) Durham:
Figure 1.10: Apta/Flos, triplum mm. 324; text placement in four sources
Some of these text-underlay solutions are prescriptive, and others pragmatic. The
Modena scribe has made an effort in both examples of hocket discussed so far (Figures 1.9
and 1.10) to set text unbroken to groups of notes which can accommodate it. On the other
hand, the solutions offered by the Ivrea and Modena scribes seem to be a means to an end:
words are split or spaced out as necessary when text is written to ensure that there will be
enough space for the hocket. Then, notes are added in an arrangement as close as possible
to the desired texting, but this still often means placing rests over syllables. The latter ap-
proach to texting would degrade quickly as copies became exemplars. We may perhaps agree
with Page that this kind of subtlety is less robust, and may become lost more easily when
works left the circles of performers and audiences who knew them well.
And yet, the texting of hockets was an important enough issue to form the subject of
the motet Musicalis/Sciencie. Here the motetus text is a letter from Rhetoric to Music asking
that she inuence her disciples not to split words with hockets, and the triplum is a letter
from Music to these very disciples (a long list of whom is given) asking them to comply with
43
Rhetorics request.
75
This of course tells us that some people were splitting text and others
were notwhich may seem unhelpful. But a closer look at scribal practice and composi-
tional design shows that far fewer words are so split than the available editions would have
us believe.
76
In fact, the intended texting is often claried by the notes which follow the hocket
section. In the example of Apta/Flos discussed above, the solution offered by Modena is
most convincing not only because words stay intact, but because the melodic line seems to
be built to accommodate the text following the hocket. In Figure 1.10, litturam clearly
belongs with the minim-semibreve-minim phrase after the hocket, and In Figure 1.9, the
rst syllable of Decor also belongs after the hocket, and in that case the entire phrase
Decor ipse quo nichil gratius is texted syllabically, like the corresponding phrase in the
rst taleavirginum decus et species.
77
It is important to note that in both of these cases Ivrea does not support the reading
which is indicated by compositional design. Since it is our only source for many ars nova
motets, we have trusted it perhaps too readily to provide texting solutions for these passages.
But there are a number of cases in the manuscript where a word that seems split by a hocket
is in fact perfectly accommodated by notes after the hocket which have been transcribed as
untexted. The most striking example is Vitrys Petre/Lugentium, where all hockets seem to
split words according to Schrades edition (which mostly follows Ivrea). In fact it is not clear
to me that any words are split. The Ivrea readings and PMFC text underlay for the motets
75
Musicalis/Sciencie is discussed in more detail in Chapter Five.
76
As Bent has observed regarding a slightly later repertory, all too often, modern editors observe the actual
physical alignment without adequate consideration of the practical constraints which produced it, Text
Setting, 294.
77
Mm. 146.
44
rst hocket are given in Figure 1.11 and Example 1.4, and these represent the hockets as
sparsely texted. But there is space for this text in mm. 2732. Moving it there leaves us
with an alternation of texted and untexted musical space that is much more convincing (see
Example 1.5, which also uses an improved reading of the text).
78
The same kind of change
applies in six more passages in the motet: in each case the text which has been loosely un-
derlaid to hockets by the scribe or editor can be snugly added after them, where there is
often exactly enough space for themto the note. Making this series of changes renders
Petre/Lugentium an essay on alternating textures, since the motet in fact cycles regularly be-
tween sections of hocket with no text in either voice, sections in which the upper voices al-
ternate in delivering passages of text two breves in length (as in mm. 316 in Example 1.5),
and truly polytextual sections. More broadly, it is my impression that there are few loosely
texted hockets in this repertory. Very often, and especially in the work of Vitry, hockets
either have words, or they dont , and this has consequences for the editor and performer as
well asin some casesserious implications for the analyst.
79
Scribal corrections and erasures serve as further evidence both of the difculties
involved, and of the importance of text-music alignment in the copying of motets.
80
These
usually involve either the erasure and re-notation of notes to bring them into better align-
ment with syllables, or the drawing of lines between syllables and notes that belong together
but are not aligned. Lawrence Earp has shown that in the Machaut manuscripts a large
78
The edition in Example 1.5 uses textual readings drawn from a literary source for the motet from 1340,
which has tenuit instead of temnunt and suscepisti in place of superesboth make more sense given
the meter. See David Howletts edition and discussion of the source in Wathey, The Motets of Philippe de
Vitry, 1336. Note however that owing to an error the motetus and triplum text switch places partway
through Howletts edition, without notice.
79
See the analyses of In virtute/Decens and Cum statua/Hugo in Chapters Five and Six.
80
See comments in Bent, Text Setting, 294.
45
Triplum: Motetus:
Figure 1.11: Petre/Lugentium mm. 1930 (Ivrea, fols. 37v38)
Example 1.4: Petre/Lugentium mm. 1930 (reproduced from PMFC1, 97)
Example 1.5: Petre/Lugentium mm. 1938, alternate text-setting
46
portion of the erasures and corrections in the texts and musicwere made specically to
clarify the correlation between words and notes.
81
And while all genres were subject to such
emendations, motets seem to be the most frequently affected.
82
Not surprisingly, correc-
tions often clarify texting during melismas and hocket the places where, as we have seen,
normal procedures of copying may not be precise enough. Figure 1.12 shows two examples
of guide-lines used by the scribe of Machaut Manuscript A: the rst of these connects the
last syllable of Desir with the longa that ends the phrase, and the second claries the
placement of -na in the melisma over superna. Guide-lines are also used to help align
text and hockets: the lines barely visible in Figure 1.13a and simulated in Figure 1.13b
(but clearly present in the original) correct text underlay that would have split the word
dolereus. The reulting reading for this passage, which is not followed in any of the avail-
able editions of Motet 10, is given as Figure 1.13c.
83
Figure 1.12: Clarication of text underlay during melismas in Machaut Manuscript A
(a) motetus of De bon/Puis (M4, fol. 417v)
(b) motetus of Fons/O livoris (M9, fol. 422v)
81
Scribal Practice, 1956. See also Bent, Text Setting, 294.
82
See ibid., 2668 for a list of corrections by genre within each manuscript.
83
Machaut, Musikalische Werke, 3:39; PMFC 2:143; Boogaart, O series summe rata, 607. Manuscript
C has an extra word at this point (fol. 214v), so that the line reads et pour plus croistre mes dolereus me-
schiesthis can work within the meter if croistre functions as a monosyllable, but the music scribe as-
signed it two notes, and dolereus is again split. Ferrell-Vog follows MS C, including plus in its text but
giving two notes to croistre. Perhaps the discrepancies in later MSS arise from this initial error in MS C,
or the uncertainty that produced it. In any case, the correction in MS A is clear in its intention.
47
Figure 1.13: Hareu/Helas (M10), triplum mm. 7582 in Manuscript A (fol. 424)
(a) Faint lines clarify text underlay before a hocket
(b) transcription in PMFC (2:143), with lines from source replicated
(c) text underlay adjusted to match corrections in source
It is not surprising that such care should be taken with Machauts motets in sources
that were in all likelihood copied under his watchful eye.
84
But similar types of emendations
were also made in other manuscripts transmitting ars nova motets. As Kgle has noted,
Ivreas numerous corrections and erasures in text and music suggest [that] most entries
were routinely proofread, and some of these are clearly motivated by text-underlay.
85
In
Chantilly, we nd a correction of a different sort near the beginning of the motetus of Rex/
Leticie. Here the music scribe noticed the words ac salutis missing from the text ac salutis
humano generi, and supplied them with a small hand, carefully placing the syllables under
the appropriate notes. He then repositioned the rst syllable of humano to accord with the
new text underlay (Figure 1.14).
Figure 1.14: Missing text added to the motetus of Rex/Leticie in Chantilly (fol. 66)
84
See Earp, Machauts Role.
85
For example, several minims have been erased and re-notated to clarify underlay in the triplum of A vous/
Ad te on fol. 19v.
48
Tenor Texts
So far the discussion of textual transmission and underlay has been limited to the
upper voices of motets. Tenor text underlay is not an issue in most cases, but we can look at
the transmission of tenor texts and compare the variations here with those observed in the
upper voices. Tenors are as subject to small variants as the upper voices ante thronum
trinitatis of Ida/Portio is labeled Ante thorum trinitatis in Chantilly, with disastrous re-
sults for the translation.
86
But more common are variants in how much, if any, of the ten-
ors text is given. In the case of Apta/Flos, all the sources give a tenor incipit. Ivrea, Modena,
and Durham have Alma redemptoris mater, while Chantilly has Alma redemptoris. On
the other hand, the tenor of O canenda/Rex is variously labeled as Tenor o canenda (Ivrea
fol. 55), Rex regum tenor (Durham, fol. 337v and FriZ, fol. 86v), and Tenor primus
(Paris 2444, fol. 48v). None of these four souces transmits the full text of the chant frag-
ment used, which is Rex regum regi lio.
87
Clark has listed 17 other motets whose tenors
use more melodic material than their manuscript text indicates, and this number is in
fact much larger when we take different sources into accountfor example, Merito hec
patimur, the tenor of Tribum/Quoniam in Fauvel, is labeled Merito in Brussels.
88
Clark
86
In the presence of the throne of the trinity becomes in the presence of the trinitys protuberance (or
the trinitys mussel); Chantilly, fol. 61v. The variant is transcribed without emendation in PMFC5, 24.
87
Clark, Concordare cum materia, 217.
88
See Ibid., pp. 467. The list does not include O canenda/Rex. Clark also gives a list of four Motets that
use less melodic material than their manuscript tag indicates (p. 48, Table II), but it is not clear that the
four should be grouped together. In once case (Apollinis/Zodiacum) only a nal melisma is omitted, so the
label is correct, as Clark notes (48n47). Firmissime/Adesto uses an Alleluia chant as tenor and the Fauvel
label, Alleluya Benedictus et cetera, simply indicates which Alleluia is being used (Alleluia Benedictus es
from Trinity Sunday), and as such is acting as a label rather than an incipit or a text (also noted by Clark,
p. 48). The same tenor in Brussels is labeled [A]lleluya alleluya alleluya, which is closer to an incipit,
depending on the number of times Alleluia is repeatedthe tenor is repeated twice. Kgle argues that the
change was deliberate; Two Abbots and a Rotulus, 162n35. On the source of the chant, see Robertson,
Which Vitry, 5769. The nal motet in Clarks Table II does not in fact have a labeled tenor in either of
its surviving copies, and the tenor melody was identied in Gunther, The Motets of the Manuscripts, xlviii.
49
has suggested that because their tenor labels diverge from the amount of chant used, these
works make up a group of motets where the composer apparently asserts a bit more control
over the preexistent melody. But it is not clear whether there is a stable enough pattern of
transmission to speak of authorial control, or whether in fact the decision of how much of
a tenor text to write was made by scribes. Only Machauts tenors are consistently labeled,
and that only within the Machaut corpus. Beyond it, they too are subject to emendation and
omission.
89
The levels of divergence between tenor labels differ markedly from the kinds of
differences we have seen in the upper voices. In the upper-voice texts it is rare indeed to see
a word missing, and when missing, this text is sometimes supplied, as in the example from
Chantilly. In tenors, on the other hand, there is clearly not the same idea of a text that
need to be completetenors have labels, not texts, and these vary greatly in accuracy and
completeness from source to source.
And, of course, many tenors are not labeled at all in some or all of their sources.
With few exceptions, these missing labels seem not to have been subject to correction. In
Ivreas redaction of Les lormel/Meyn the tenors textoriginally omittedwas indeed sup-
plied later in a different ink color (see Figure 1.15). However, the tenor of this late thir-
teenth-century motet is itself a rondeau, whose text can and probably was intended to be
sung. Thus the omitted text is in this case treated as missing lyrics rather than as a missing
label.
90
89
Amours/Faus semblant has the tenor Vidi Dominum in Trmolle (fol. 8)this matches the Machaut
sources. In Ivrea (fol. 21), this tenor is written as Vidi dm &c., which is not strictly incorrect since the
chant used includes the words vidi dominum facie ad faciem. Qui es promesses/Ha Fortune (M8) has the
tenor Et non est qui adiuvet in Trmolle (fol. 8), Non est qui adiuvet in Ivrea (fol. 24v), and no label
other than Tenor in CaB (fol. 16v). Martyrum/Diligenter, which has the tenor A christo honoratus in the
Machaut manuscripts, is labeled Tenor x
po honorate & c. in Ivrea (fol. 11); in this case, the & c. is actu-
ally incorrect, since the chant used ends with honoratus.
90
Though in Ivrea the text is not underlaid exactly to the tenor notes (it could not t under the tenor as
50
Figure 1.15: Tenor text added to Les lormel/Meyn (Ivrea, fol. 22)
The text added to the rondeau tenor of Les lormel/Meyn and the missing syllables
supplied in Chantilly for the motetus of Rex/Leticie serve as reminders that when text is
missing it can be supplied, if someone deems it necessary. The large number of works in the
repertory whose tenors remain unlabeled in one or all of their sources strongly implies that
a motet was considered complete in its redaction whether or not its tenor had a text.
Intabulations
The last type of material to consider from the point of view of scribal reception for
motet texts is intabulations. On the face of it, the very existence of instrumental intabula-
tions of motets would argue against the importance of words, since the fully instrumental
(probably keyboard) performance that tablatures make possible precludes the delivery of
text and seems to silence the poetic aspects of composition.
91
But the transmission of sev-
eral ars nova motets in the so-called Robertsbridge Codex tells a surprising story.
92
written), and is allowed to ow into the margin. The text given is Je ny saindrai plus graile saintureite, mon
ami est marie. Il a mis mon cuer en si grant distresse, which provides enough text for all but the internal ab
of the rondeau. CaB does not give a tenor here. Torino 42 has only Je ne chaindrai mais; Udine includes
the tenor text as part of the original copying effort, though as in Ivrea it is not underlaid, but simply written.
The source is fragmentary, thus Je ne chaindray plus haingre chainturete mes amis est is all that survives
of the tenor text (fol. 1).
91
Though Jane Flynn has suggested that in the case of Machauts De toutes ous in the codex Faenza the
intabulator emphasizes certain aspects of the poetic content and musico-poetic structure: The Intabula-
tion of De toutes ours, 178.
92
This source, BL Add. 28550, is probably from the 1350s; see Roesners introduction in Vitry, Complete
51
At the end of this manuscript, which contains a chronicle of the abbey of Rob-
ertsbridge between the years 1308 and 1333, are found two folios which are the earliest
surviving examples of instrumental notation. In addition to three estampies, three motets
are intabulated: Firmissime/Adesto and Tribum/Quoniam from Fauvel and an unicum with
a triplum which begins Flos vernalis stirps regalis. The reason that we can know the text of
the latter, and easily identify the two former works, is that, unlike the estampies, the motet
intabulations are in fact texted.
The rst three lines of Firmissime/Adesto, reproduced in Figure 1.16 along with a
diplomatic transcription of the text underlay, are instructive. Here the motetus text Ades
is underlaid to the beginning of the motet, this being as much of the word Adesto as
sounds before the triplum text begins, overlapping with -to. After this the triplum text
only is given. It may come as a surprise that in all three motets the triplum text is given in
full. Though the writer has used a large number of abbreviations, not a single word is miss-
ing from either Firmissime dem or Tribum que non abhorruit.
93
If an initial explanation for the presence of this text is that it served as an aide-
mmoire for the intabulator, a closer look dispels this theory: the text was entered after the
notes and letters of the intabulation, which it studiously avoids. At two points reproduced
in Figure 1.16 the scribe has run out of space and added a word above the text, and in
once case, the addition of morti near the end of line 2, he has drawn a faint line showing
where this interlinear word ts into the text. A more detailed comparison of the motet with
its intabulation shows that the underlay of the triplum is surprisingly faithful. Despite the
Works, vvi.
93
Flos vernalis is an unicum, and preserved incompletely, but the portion of the text that survives seems to
be metrically complete.
52
trinita
|| Fir missime dem teneamus
morti
2. tis patrem dili g[a]mus qui nos tanto amore dilexit datis ad vitam e[re]
3. xit
Figure 1.16: Intabulation of Firmissime/Adesto, mm. 134 (Robertsbridge, fols. 43v
44) with text underlay transcribed
Example 1.6: Firmissime/Adesto, mm. 718 (reproduced from PMFC 1:60)
53
transformative arranging that makes the intabulation much more orid than the original,
the basic points of congruence between the triplum text and its melodic contour remain.
Thus the intabulation preserves the A on -mis[sime] in m. 8, the F on qui in m. 16,
and the B-atFB-at sonority on -xit in m. 23 (see the congruent points marked b,
d, and g in Figure 1.16 and Example 1.6). Certainly not all the underlay is the same:
the parallel fourth movement between the upper voices in m. 22 is set to ad vitam in the
intabulation and [e]rexit in the motet (f in Figure 1.16 and Example 1.6). However, the
F-C leap in m. 21 appears in both versions with the words [mor]ti da[tos]a congruence
achieved in Robertsbridge by the addition of morti above text already entered (see label
e). The other interlinear placement of textthat of trinita- at the end of the rst line
in Figure 1.16, is also necessitated by text underlay: as written, this text accompanies the
notes B-atAG, as it does in the original (see label c). All around, then, it seems that
the scribe of the codex was surprisingly faithful to the original motet in his text-underlay,
and that in entering the text, after the intabulaiton had already been written, both com-
pleteness and accuracy seem to have been priorities.
What does all of this mean? For our understanding of the Robertsbridge Codex,
this messy-looking but actually quite careful texting opens up the possibility of vocal per-
formanceperhaps by a self-accompanying musician. But even if we do not co-opt this
the earliest source of purely instrumental music for sung performance, the care taken
about text in all three motets implies that the presence of this text was an important part
of the composition on some level, whether musical or intellectual. Of course, the bitextual
framework is not preservedthe motetus is represented only by two syllables in Firmissime/
Adesto, and not at all in the other two works. And no sign of the tenor text is present. But
54
the complete presence of one voicehere the triplumemphasizes the importance of up-
per-voice text to motets while also showing how these texts can sometimes take on lives of
their own.
94
This independence is further conrmed by the literary transmission of motet
texts and by their citations in a wide range of poetic contexts.
MOTETS AND/AS LITERATURE
In a frequently cited passage of the Rgles de la seconde rhtorique (c. 141132), the
anonymous author places Vitry chronologically between Jean de Meun and Machaut, com-
menting on his achievements and innovations:
Aprz vint Philippe de Vitry, qui trouva la manire des mots, et des balades,
et des lais, et des simples rondeaux, et en la musique trouva les .iiij. prola-
cions, et les notes rouges, et la novelet des proporcions.
95
[After [Jean] came Philippe de Vitry, who discovered the manner of motets,
and of ballades, and of simple rondeaux, and in music he discovered the four
prolations, and red notes, and the novelty of proportions.]
In light of our scant knowledge of Vitrys biography and reputation, these few lines
have been mined for anything they can tell us about the authors lost output and legacy. The
evidence of his having found a manner of motets is often read as synonymous with the
ars nova.
96
As for the ballades and rondeaux, commentators have lamented that no musical
settings of the formes xes by Vitry survive.
97
These readings, though by no means impossible, seem to ignore the larger structure
94
Cf. Roesner, who notes the presence of the triplum voice in the arrangements, but cites their embellish-
ment and thinning of texture as indicative of the complete transformation in idiom occurring when a motet
was stripped of its most distinguishing feature, its texts, and performed as an idiomatic instrumental work,
Vitry, Complete Works, v.
95
Langlois, Recueil darts de seconde rhtorique, 12.
96
For example in Leech-Wilkinson, The Emergence of Ars nova, 285.
97
As in Brothers, Chromatic Beauty, 91.
55
of the quotation, which is clearly divided into two halves. The rst part, in keeping with the
focus of the treatise (namely, the art of versication) lists Vitrys poetic achievements. Then
the author shifts focus to his musical nds. Signaling the change of subject unambigu-
ously with et en la musique, he identies Vitrys musical innovations as the four prola-
tions, red notes, and the novelty of proportions. The prose format of the treatise places
few constraints upon the writer, and thus we should take him at his word when he includes
two separate lists. Thus motets would seem to be listed among other poetic formsthough
forms that are often (ballades, rondeaux) or sometimes (lais) set to music.
98
That ars nova motet texts could in themselves have been considered poetry should
not be a surprise, given that the two most famous fourteenth-century motet composers
were also two of Frances most illustrious poets. Machauts inuence on later writers such
as Geoffrey Chaucer and Chrisine de Pisan is well known, as is Petrarchs praise of Vitry as
the only true poet among the French (poeta nunc unicus Gallicarum . . .).
99
And yet the
idea that Vitrys newly-discovered manire des mots is a poetic achievement will perhaps
sit uncomfortably with received views of the motets evolution. While inquiries into Mach-
auts literary and musical output have recently joined forces to some extent, Vitrys poetic
output has been traditionally down-played, or assessed separately from him musico-poetic
works.
100
And as for truly anonymous ars nova motet texts (as opposed to the texts of works
98
Indeed, Vitrys only surviving ballade, De terre en Grec Gaule appellee, survives without music and was
probably never set. It was part of an exchange to which poets Jean de le Mote and Jean Campion also contrib-
uted ballades, all of which survive without music. Their theme is the poetic craft, and it is possible that none
of them was setwe have no evidence that le Mote, the target of the exchange, was a musician. This series of
ballades is discussed in Chapter Six. For an edition and commentary, see Diekstra, The Poetic Exchange.
99
This praise is transmitted in the longer (a) version of Petrarchs letter to Vitry; see Paris, Les manuscrits
franois, 3:181. On Vitrys literary reputation, see Wathey, The Motets of Philippe de Vitry, 11921.
100
Ibid. 1212.
56
surviving anonymously but probably by Vitry), these have received no attention whatsoever
as poetic texts.
And yet motet texts clearly functioned in literary ways. Andrew Wathey has recent-
ly drawn attention to the survival of seventeen motet textsmost by Vitryin fourteen
sources from the 14th and 15th centuries. These manuscripts, which range from a collec-
tion of sermons from 1340s Avignon to humanist miscellanies copied by German scholars
working in Italy in the mid- and even late- fteenth century, transmit motet texts in ver-
sions that are in some cases more accurate than the versions preserved in sources such as
Ivrea.
101
Thus motet texts were interesting enough to be transmitted on their own as poetry,
long after the music with which they were associated would have gone out of style. But at the
same time, this poetry seems to have gained something from being motet texts. Wathey has
stressed the initial dependence of the tradition on copies including music, demonstrated
most clearly by the inclusion of voice-labels in the earliest of the text-only sources.
102
That
texts would be copied into manuscripts grouped by the motet to which they belong, and
with the indications of which motet voice sings which text, clearly shows that the scribes
and readers of these manuscripts took pleasure from the idea that separate texts combine
to make one work. In fact, the motet Phi millies/O creator survives only in a text manuscript,
but owing to the care of the scribe, we know not only which words belong to the motetus and
which to the triplum, but also that the tenor and contratenor sang a parodic rhyming cou-
plet between them.
103
Clearly the co-existence of texts in motets was an important aspect
101
Wathey, The Motets of Philippe de Vitry, 1224.
102
Furthermore, the literary transmission of Vitrys motet texts is an anomaly for the 14th century. As
Wathey shows, the transmission of motet texts outside of music manuscripts was not unusual in the 13th
and 15th centuries, but for the 14th century it is conned to Vitry. Idem, Words and Music, 1550.
103
Phi millies/O creator is discussed in detail in Chapter Six.
57
of their overall aesthetic value.
The poetic transmission of texts not by Vitry is rarerso much so that Wathey has
suggested that the fact of literary circulation may even emerge as a pointer towards attri-
bution, pushing, for example, the anonymously transmitted text Quid scire proderit closer
to the Vitry corpus.
104
This may well prove to be the case, or it may be subject to revision if
more motet texts from anonymous works are discovered in literary collections. From the
Machaut corpus, the two texts of Qui/Ha Fortune (M8) survive in a collection of French
poetry from the 1480s.
105
Aside from the transmission of entire texts, a number of literary allusions, cita-
tions, and elaborations attest to broader late-medieval interest in motet texts. These in-
volve works by Vitry and Machaut as well as pieces transmitted anonymously. Sometimes
the motet text is quoted explicitly, and carries the force of a proverb. Such is the case in
Gace de la Buignes hunting treatise in verse, the Roman de Deduis (135977). Here, in the
course of discussing the behavior of falcons the speaker quotes most of the rst three lines
of the motetus in Vitrys Douce/Garison. Vitrys original is as follows, with the text quoted
by la Buigne italicized:
106
Garison selon nature
Desiree de sa dolour
Toute humaine creature
Mais je qui ai dun ardour
[Every human creature desires a natural cure for its sorrow.
But I, who am seized by an ardor...]
104
Wathey, The Motets of Philippe de Vitry, 128.
105
Earp, Machauts Role, 4945.
106
On the place of this quotation within the larger narrative of la Buignes poem, see Leach, Sung Birds,
22932.
58
In the Roman de Deduis the word order has been shifted slightly, creating a rhyming couplet
where Vitrys rhyme-scheme was ABAB. More importantly, the word humaine has been
left out of Vitrys original because la Buigne is applying the maxim to a bird rather than a
person. In a sense he has rendered Vitrys message more universal, and also perhaps more
humorous:
107
Et, se loysel se va baignier,
Queer il a de leaue mestier,
Len ne le doit mie blasmer
De ce de quoy fait a loer
Quant garison selon nature
Desire toute creature
De sa douleur, si comme dit
Un aucteur, qui le nous escript
En un motet, quil st nouveaux,
Et puis fu evesque de Meaux;
Phelippe de Vitry ot nom,
Qui mieux st mots que nulz hom.
108
[And if the bird goes to bathe itself, it seeks from the water that which is
necessary. One cannot blame it at all for this but rather must praise it, be-
cause every creature desires a natural cure from its sorrow, as an author
said who wrote it for us in a motet, which he newly made. Afterwards he was
Bishop of Meaux. Philippe de Vitry was his name, and he knew how to make
motets better than any other man.]
109
A similar (proverbial but transformative) use of the rst several lines of a motetus
text occurs in a fatras impossible. In this form of nonsense poetry a two-line refrain AB of-
ten taken from pre-existing material is allowed to dictate the rhythmic and metrical scheme
for a poem of eleven lines with structure AabaabbabaB.
110
These poems were not usually
107
Or perhaps not: Elizabeth Eva Leach has read in this passage an evocation of burning love, especially
Didos love for Aeneas, to which the garisonis sexual consummation, Sung Birds, 230.
108
ll. 634556
109
Trans. Leach, Sung Birds, 312.
110
The couplet is sometimes considered part of the poem, which is then summarized as AB AabaabbabaB,
but as Uhl has noted, the manuscript layout and placement of large capitals does not support this, and the
59
set to music, although there are two fatras in the interpolated Roman de Fauvel whose rst
and last lines are notated.
111
And conversely a bit of the Roman de Fauvel made its way into
a fatras. Specically, Watriquet de Couvin (. 131929) used the opening two motetus
lines of Super cathedram/Presidentes to structure a fatras. The borrowed material is a com-
plaint about corrupt rulers that sits comfortably with the broader world of Fauvel:
Presidentes in tronis seculi presiding on the thrones of [our] age
Sunt hodie dolus et rapina. today are treachery and plunder
In the twisted world of the fatras, the two lines get split, as we would expect, and their sense
becomes twisted. The rst line of the motetus still refers to this world, but the second is
linked hypothetically with the next: when upon the death of a sinner the Our Father is
said in order to absolve his wrongdoings, treachery and plunder are now in the sky with
the angels. An early transition into Latin in the middle of line 10 helps to cement this re-
reading:
Presidentes in thronis seculi,
Ce dist uns eus armez de cuir boilli,
En cop de [vit] si grant medecine a
Cune charrete jusqua Mes en sailli,
5 Qui engendra le seigneur de Seulli,
La Maselaine, dont uns cos se disna;
Mais uns harens touz sen desgratina,
Quant il fu mors, pour ce con li toli
La pater nostre qui li adevina
rst two lines were written only as aides-mmoire for scribes, Fatras fatrasie, 218. This would have an
impact on the editions of the two fatras set to music in the interpolated Roman de Fauvel: Rosenberg and
Tischler have inserted the eleventh line after the rst to form the rst AB, though the manuscript presents
eleven lines without a refrain, underlaying the rst and last to music: The Monophonic Songs, 1257.
111
Ed. Rosenberg and Tischler, The Monophonic Songs, 1257. Two other potential musical settings of
couplets on which fatras are built surviveone is the lowest voice in the polytextual English motet Deus
creator/Rex genitor/Doucement; the other, a monotextual setting in Wrocawsee Brewer, French Ars Nova,
1945 and A Fourteenth-Century Polyphonic Manuscript, 910, and Lefferts, The Motet in England,
845. However, it is not clear whether these are examples of fatras nots (as has been assumed), or whether
the poems and musical works share the same refrain. If the fatras was indeed conceived of in its eleven-line
form as Uhl suggests (see previous note), then the polyphonic works, which place lines 1 and 11 together, are
probably referring to a common source rather than to a fatras.
60
10 Quavec les angles, in gloria celi,
Sunt hodie dolus et rapina
112
[Presiding on thrones of time
Said one with arms of boiled leather,
There is such strong medicine in a good fuck
That it blew a cart right into Metz,
Which gave birth to the lord of Seuilly,
On the Feast of Madeleine, when he dined on the hull of a wheat germ;
But a herring scratched them all,
When he was dead because one took away from him
The our father which came upon him
With the angels in the glory of heaven,
There are days of deceit and plunder.]
113
The exact meaning of some of the lines is elusive, but the way in which the original
couplet is transformed is evident: the ideal reader of for such poetry is one who is familiar
with these lines in their original context, and is therefore all the more able to appreciate
their transformation. Such a man was Watriquet de Couvin, and such another was his pa-
tron Guy de Chtillon, for whom this and other fatras were copied into a manuscript which
also includes a miniature of all estates grooming Fauvel, placed somewhat inexplicably
before Watriquets Dit de Loyaut.
114
In addition to being the source of these obvious quotations (labeled as such in the
Roman de Deduis and marked as external in the fatras by their position within the form and
their use of the Latin language), it seems that motet texts also exerted a more oblique inu-
ence on late-medieval writers. A famous example comes from Chaucers oeuvre. As Kit-
tredge rst observed in 1915, the Book of the Duchess incorporates many translations and
112
Ed. Uhl, Fatras fatrasie, 21920.
113
I am grateful to Professor Michael Randall for his translation of these lines, which have not previously
been rendered in English.
114
BN fr. 14968; on connections between Watriquet and the Fauvel crowd, see Taylor, Le Roman de
Fauvain, 5809.s
61
paraphrases from a number of Machauts works, including several motets.
115
The following
lines, from a tirade against fortune, Borrow in at least four instances from Qui/Ha Fortune
(M8):
116
Chaucer, Book of the Duchess Machaut, Qui es promesses
For fals Fortune hath pleyd a game
Atte ches with me, allas! the whyle!
The trayteresse fals and ful of gyle, Fausse, tratre, perverse (l. 17)
That al behoteth and no-thing halt,
She goth upryght and yet she halt,
That baggeth foule and loketh faire,
The dispitouse debonaire,
That scorneth many a creature!
An ydole of fals portraiture Une ydole est de fausse pourtraiture (l. 9)
Is she, for she wil sone wryen;
She is the monstres heed y-wryen,
As lth over y-strawed with oures; Cest ens couvers de riche couverture (l. 7)
Hir moste worship and hir our is
To lyen, for that is hir nature;
Withoute feyth, lawe, or mesure. Sans foy, sans loy, sans droit et sans
She is fals; and ever laughinge mesure (l. 6)
With oon eye, and that other wepinge.
That is broght up, she set al doun.
117
Chaucers poem refers at several points unambiguously to Machauts triplum, bor-
rowing ideas (lth covered in owers), copying diction (faith, law, measure), and even trans-
lating word for word when he calls Fortune An ydole of fals portraiture. None of this is
115
Chaucer certainly knew Qui /Ha Fortune(M8), and possibly H! Mors/Fine amour (M3) and Fons/O livoris
(M9). His use of Qui /Ha Fortune in the tirade against Fortune is discussed below, and this triplum is also
relevant to Duchess ll. 81213. It has been suggested that the description of Fortune as a scorpion who
stings from behind in ll. 63640 is borrowed from the motetus of Fons/O livoris (Wimsatt, Chaucer and the
French, 73, 159), but Fortune is not mentioned in that text, and the scorpion represents Lucifer. Perhaps
the idea of a scorpion is common enough that Chaucer need not have gotten it from Machaut. Kittredge
cites lines 18 of the triplum of M3 as an inspiration for the Black knights lay of complaint (ll. 475486)
but Wimsatt cites Jugement dou Roy de Behaigne as a main source here, though he notes the similarity with
Motet 3; Kittredge, Guillaume De Machaut, 78; Wimsatt, Chaucer and the French Love Poets, 158.
116
This passage also has borrowings from Machauts Remede de Fortune and the Roman de la Rose; see Kit-
teredge, Guillaume De Machaut, 1011, and Wimsatt, Chaucer and the French Love Poets, 1589.
117
Ll. 61835; Chaucer, The Complete Works, 1:298.
62
surprising within the broader scope of the Book of the Duchess, which is full of references to
the Roman de la Rose, dits by Machaut and Froissart, and a host of other classical and me-
dieval sources.
118
What may be surprising is that the triplum of Qui/Ha Fortune (M8), along
with several other motet texts, should be included in this company of inuences alongside
the Ovide moralis, Jean de le Motes Regret Guillaume, Vergila Aeneid, and Alan of Lilles De
Planctu Naturae, to name a few.
Obviously Chaucer knew his source well. Not only does he quote directly in one case,
he also seems to have the whole of the original triplum rmly in mind. For these references
appear in reverse order in Chaucers poem, relative to their original context. This could of
course be a coincidence, but it could also be deliberate. As the last line of the quoted text
reminds us, Fortune brings down what is brought up: in other words, she causes reversals.
The backwards incorporation of a series of lines from a source about Fortune would thus
be well in line with her usual iconography; we will see her doing far stranger things to music
and poetry in Chapters Four and Six. Regardless of whether the particular order in which
lines from the motet are used here is signicant, it is interesting to see Mahchauts triplum
becoming part of the fabric of Chaucers poem. As Jacques Boogaart has shown, Machauts
motets are themselves suffused with quotations from sources ranging from the bible and
Boethiuss Consolatio to the poetry of the trouveres.
119
Here we see them in turn become the
objects of citation and allusion.
Nor is this phenomenon limited to Machaut. Andrew Wathey has recently brought
to light a number of later engagements with motet poetrymostly the work of Vitryby
118
See list in Wimsatt, Chaucer and the French Love Poets, 155.
119
Boogaart, Encompassing Past and Present.
63
fteenth-century humanists. Not only did they include motet text in their collections, as
noted above, but they also allowed these texts to serve as models and inspiration for new
poetry and moralistic literature. For example, a 1444 pamphlet by Aeneas Silvius Picco-
lomini (later Pius II) engages with the themes of the triplum of Colla/Bona and echoes its
language.
120
Another mid-fteenth century sourcea humanist miscellanyincludes a
113-line poem that begins Quondam colla iugo Veneris submisserat Hugo (Once Hugo
had lowered his neck under the yoke of Venus). It tells the story of a cleric named Hugo
who is tried for having taken a concubine. When she denounces him because he is not rich,
Hugo is driven into a mysoginistic tirade (Woman, dregs of satan, stinking rose, charming
poison, etc.), and the reader is given the moral: character, strength, spirit, possessions,
body, and honorso many goods things does the cleric loose in [turning to] women.
121
The poem is pertinent here because its opening line references the triplum of Colla/Bona,
which begins Colla jugo. Furthermore, in naming the lleric Hugo the poems author seems
to be referencing the motet Cum statua/Hugo, which is a tirade against a friar named Hugo
for having committed some unspecied hypocritical act against Vitry.
122
In combining
these two references in the poems opening lines, the writer is thus implicitly noting that
they belong together because they are borrowed from the same genre.
RECEPTION OF ARS NOVA MOTETS OUTSIDE OF ARS NOVA CIRCLES
Although our primary focus has been on motets in their initial contexts, and on
their meanings for the scribes and theorists who were among their rst interpreters, we have
120
Wathey. The Motet Texts of Vitry, 200.
121
Femina, fex sathane, rosa fetens, dulce venenum, l. 93; Ingenium, vires, animam, res, corpus, honores
/ Tot perdit bona vere clericus in muliere, ll. 11011: Huemer, Lateinische Rhythmen, 296.
122
Cum statua/Hugo is discussed in detail in Chapter Six.
64
also had occasion to see ars nova motets functioning farther out from their points of origin,
both geographically (England, Italy) and temporally (the fteenth century). When we com-
pare reception of motets by ars nova theorists and scribes with that of more peripheral audi-
ences, a number of interesting differences come to light. The most consistent of these con-
cern naming and emphasis. The fact that the Robertsbridge intabulations give the triplum
rather than the motetus text for each motet gains signicance in light of the propensity of
English music theorists to cite ars nova and even ars antiqua motets by their triplum texts.
123
And as we have seen, Chaucers borrowings from Machauts Qui/Ha Fortune (M8) also in-
volve the triplum rather than the motetus. Poets, scribes, theorists, and intabulator: each in
his own way seems to give primacy to the triplum. This implied hierarchy in turn contrasts
with continental practice, which consistently refers to the motetus as the name-giving voice
across the rst three quarters of the century. Here again, indexers and theorists agree, as
indeed do poets: both fourteenth-century citations of motet in poetry discussed above (in
the Roman de Deduis and in Watriquets fatras) use the beginning of a motetus text.
To be sure, these are not earth-shattering differences. But behind names lies ontol-
ogy. Noting that Italian treatises use triplum citations earlier in the 14th century than do
French treatises, Margaret Bent has suggested that such citations may disclose a different
view of the genre.
124
And in our case the conuences in naming and placing of emphasis
among theorists, scribes, and poets in France, on the one hand, and England on the other
indeed suggest that there are differing views of the genre at play. What these views might
be is harder to say. In most motets the triplum text is probably the more easily intelligible,
123
See Appendix 1A for examples from the Regule of Robertus de Handlo and the Summa of Johannes
Hanboys; the latter even cites an ars antiqua motet (Aucun ont trouv/Lonc tans) by its triplum.
124
Bent, A Note on the Dating, 223.
65
since its range is higher and its speed of delivery closer to that of speech. And yet the mote-
tus has historical primacy and an etymological link with the motet as genre. The tendency
of French theorists to refer to motets by their inner, less audible voice may reect a height-
ened awareness of this history and etymology, perhaps combined with a kind of inside-out
listening. Once outside of the circles for which is was written, the motet was likely heard
differently, and the more audible triplum seems to have taken the lead.
One other difference between insular and continental reception concerns the writ-
ing of contrafactanew texts to t existing melodies. This was common ars antiqua prac-
tice: Gordon Anderson has written that of the surviving corpus of thirteenth-century Latin
double motets, more than a third are contrafacta of French or bilingual motets.
125
In con-
trast, ars nova motets were never furnished with new texts within the circles for which they
were initially written. This lack of textual re-composition is probably a result of the close
structural, rhetorical, and semantic alliances between ars nova texts and musicalliances
which have been demonstrated for a number of works in the repertory, and which will be
explored in depth in later chapters of this study. It would seem that the original intended
listeners and practitioners were aware enough of these congruences that they did not wish
to re-write texts to change their language or subject-matter. But this was not the case out-
side of France. As Peter Lefferts has noted, the avoidance of the vernacular was a feature
of the motet outside a narrow but prolic Parisian orbit, and the small number of ars nova
contrafacts that survive replace French text with Latin for English audiences.
126
Thus the
125
In some cases a motet received new texts in two stagesrst for one of its upper voices, then for an-
other: Anderson, Notre-Dame Latin Double Motets, 41.
126
These begin to appear in English sources in the second half of the century, created perhaps in the wake
of Jean le Bons captivity in the battle of Poitiers and ensuing English stay, Lefferts, The Motet in England,
190. But the English preference for Latin over French goes back to the thirteenth century: see Anderson,
Notre-Dame Latin Double Motets, 37.
66
notes, but not the words of the Ivrea motet Se paour/Diex tan desir/Concupisco are preserved
in the rear yleaves of an English source from the second half of the century as Domine
quis/De veri cordis/Concupisco.
127
As Lefferts has noted, these Latin texts addressed to God
and Jesus, are less suited to the tenor concupisco (I lust) than the courtly French texts
which came before them.
128
But the use of Latin texts in this motet (and likely in several
other works from the same source) resulted in upper-voice content more suited to the over-
whelmingly sacred contexts for motet performance in England.
129
That the replacement
of French texts with Latin was fairly common practice is attested by the triplum of the
ve-voice motet Are post libamina/Nunc surgunt by Matheus de Sancto Johanne. This text
ends by actually identifying itself as a Latin contrafact of a French original: the active, dis-
tinguished Frenchman composed this song on French melodies but after he had revised it
with the Latin language it more often became sweet to the English, replacing Deo gratias.
130
* * *
In this chapter I have used various kinds of reception as evidence about how those
interacting with ars nova motets in the fourteenth century might have conceived of them.
Specically I have been interested in how the internal hierarchy of voices might have been
constituted. What emerges from this receptionscribal, theoretical, and poeticis a top-
down, or perhaps inside-out, rather than a bottom-up idea of the motet. Upper-voice texts
seem to be integral to motets, with the motetus voice giving its name to the genre and the
127
Oxford, Bodleian Library, MS Mus. e. 7, fols. 268v9.
128
The Motet in England, 280.
129
Lefferts suggets that Parce piscatoribus/Relictis retibus and Omnis terra/Habenti dabitur are also contrafacts,
ibid. 204.
130
On Matheus de Sancto Johannes authorship of this work and its broader cultural context, see Wathey,
The Peace of 13601369, 14950.
67
motetus incipit giving composition their names. This is not to downplay the fact that mo-
tets would surely have had different kinds of signicance for different kinds of musicians
and listenersrecall Murinos focus on tenor construction. But he also makes it clear that
these voices are a subset of the genretenores motettorum. More broadly, a rhetorical and
ontological emphasis on the upper voices is borne out by the layout and illumination of
manuscripts, by the focus of scribal corrections, and by the independent poetic lives motet
texts seem sometimes to have led.
This top-heavy view of motets seems to conict with several aspects of modern
motet reception. On the one hand, we have been suspicious of motet texts because of the
polytextual texture in which we encounter them: it is often assumed that when upper-voice
texts are combined, they are rendered inaudible. On the other hand, it seems that the musi-
cal form of motets arises from the tenors rhythmic repetitions, and as such the upper voices
would seem to be subordinate to the tenors in theme or form. Both views can be nuanced
considerably, and this will be the goal of the following two chapters. These in turn will pre-
pare the way for a series of analyses that read motets in ways consistent with their reception
as described above: that is, as works in which upper-voice ideas can fundamentally affect
musical texture, form, and sometimes even the tenors textual and melodic content.
68
APPENDIX 1A
Citations of Surviving ARS NOVA Motets in Treatises
The following list of motet citations from c. 13001415 is arranged in approximate
chronological order. Unless otherwise indicated, numbers in parenthesis refer to page-
numbers in the editions identied in the notes. Information on the sources and editions
of cited motets is available in the Catalog of Ars nova Motets at the end of this study.
COMPENDIUM TOTIUS ARTIS MOTETORUM (before 1340)
1
Douce/Garison in tenore de Garyson (35)
Tuba/In arboris in tenore de In arboris (35)
Super/Presidentes in uno moteto Praesidentes in tronis seculi (37)
Firmissime/Adesto in Adesto sancta trinitas (37)
in moteto Adesto sancta trinitas (37)
Mon chant/Qui doloreus in uno moteto Qui doloretus (37)
Tribum/Quoniam in moteto Quoniam secta latronum (37)
JOHANNES BOEN, MUSICA (c. 1355)
2
Floret/Florens Sic et secunda admittitur in motheto Florens vigor super verbo
Mardocheo (68)
O canenda/Rex et secunda et quarta in tenore motheti Rex quem metrorum (68)
Se grasse/Cum venerint Ex quo concludo motetum Cum venerunt (67)
SEX SUNT SPECIES PRINCIPALES... (mid-14th century)
3
Firmissime/Adesto in moteti tenore qui vocatur Adesto (67)
1
Ed. Wolf, Ein anonymer Musiktraktat, 348. On the relationship between this treatise and the ars
nova writings broadly construed, see Fuller, A Phantom Treatise, 445, and Balensuela, The Borrower is
Servant, 1214.
2
Frobenius, ed., Johannes Boens Musica.
3
Sex sunt species principales...explicit ars quevis mensurandi motetos compilata a magistro Philippo de
69
Colla/Bona in Bona condit (67)
Douce/Garison in Garison (69)
ut in tenore de Garison (69)
Tuba/In arboris ut in tenore In arboris (69)
Garrit/In nova ut in tenore de In nova fert animus (69)
DE MUSICA MENSURABILI (mid-14th century)
4
Super/Presidentes ut in tenore de Presidentes (184)
Douce/Garison ut in motecto de Garison (186)
ut in eodem de Garison exemplum de tenore dicti moteti (186)
Garrit/In nova ut in tenore de In nova (186)
Tuba/In arboris ut in tenore In arboris (186)
QUATUOR PRINCIPALIA (1351)
5
Cum statua/Hugo in moteto qui vocatur Hugo quod edidit Philippus de Vitriaco
(420)
Vos/Gratissima ut patet in tenore de Gratissima, quem idem Philippus edidit
(420)
Tant a soutille/Bien ut patet in moteto qui vocatur Tant a sotile pointure (459)
ARS (MUSICAE) (Johannes Boen, c. 135075)
6
Impudenter/Virtutibus quod in moteto Virtutibus (28)
in tenore Virtutibus (29)
Vitry Magistro in musica. This is the treatise also referred to as Ars mensurandi motetosa title derived
from the explicit. Ed. Reany et al., Philippi de Vitriaco Ars nova, 55-69. See comments in Fuller, A Phantom
Treatise, 267.
4
Ed. Coussemaker, Scriptorum 3:177-93.
5
Ed. Aluas. See commentary on motets cited, p. 1458. The last part of this treatise in its A version is
edited as De musica antiqua et nova in Coussemaker, Scriptorum 3:334-64.
6
Boen, Ars (musicae), 15-46.
70
et hoc in in primis quatuor notulis illius excellenter moteti Vir-
tutibus clare possimus contemplare
(17v)*
Apta/Flos sic et in tenore Flos virginum (29)
ut in tenore de Flos virginum (28)
Tuba/In arboris ut in tenore moteti In arboris (26)
O canenda/Rex ut in contratenore de Rex quem metrorum (27)
Rex/Leticie ut in tenore Rex Karole moteti eodem (78v)**
* In the redaction in London, British Library, Additional 23220, ff. 14v-21v.
**In the redaction in Venice, Biblioteca Nazionale Marciana, lat. app. cl. VIII/24 (coll.
3434), ff. 73v-89v on fol. 78v.
CUM DE SIGNIS TEMPORIS (later 14th-c. copy from mid-century exemplar)
7
Firmissime/Adesto ut in moteto Adesto sancta trinitas (4v)
in moteto Adesto (4v)
Colla/Bona in Bona condit (4v)
Tuba/In arboris in moteto In arboris epyro (4v)
Garrit/In nova ut in tenore In noua fert animus (4v)
Mon chant/Qui doloreus ut in moteto Qui dolereux (4v)
Vos/Gratissima ut in moteto Gratissima uirginis species (4v)
Qui/Ha Fortune ut in moteto Qui des promesses de fortune se e (5)
Douce/Garison ut in moteto Garison selonc nature (5)
Zolomina/Nazarea ut in moteto Nazarea (5)
7
Paris, Bibliothque Nationale, lat. 14741, ff. 2r-6v. Edited in the Thesaurus Musicarum Latinarum, <
http://www.chmtl.indiana.edu/tml/14th/ANOQUAE_MPBN1474.html>. Also edited but conated with
other treatises in Reany et al., Philippi de Vitriaco Ars nova, 259, 32.
71
SEX MINIMAE POSSUNT PONI (c. 1400; Italian copy of an earlier exemplar)
8
Orbis orbatus/Vos pastores ut in Orbis orbatur (19v)
9
Firmissime/Adesto ut in Adesto uetus (19v)
in Adesto (20r)
Colla/Bona in Bona condit (20r)
Douce/Garison in moteto Garison (20r)
ut in tenore de Garison (20r)
Tuba/In arboris ut in motecto In arboris in tenore illius moctecti (20r)
ut In arboris (20r)
Garrit/In nova ut in tenore In noua sit animus (20r)
OMNI DESIDERANTI NOTITIAM... (late 14th century)
10
Tuba/In arboris in moteti tenore qui dicitur In arboris (33)
in tenore de In arboris (34)
Garrit/In nova vel in tenore de In nova fert animus (33)
ARS CANTUS MENSURABILIS (after 1376)
11
Ida/Portio in motecto Ida capillorum (212)
ut habetur in tenore Portio nature vel Ida capillorum ut hic, per
exemplum (256)
Rex/Leticie sic in illo magno motecto Rex Karole (241)
8
Rome, Biblioteca Apostolica Vaticana, Barberini lat. 307, ff. 17r-20v. Edited in the Thesaurus Mu-
sicarum Latinarum, < http://www.chmtl.indiana.edu/tml/14th/VITANV_MBAVB307.html>. Edited but
conated with other ars nova related sources treatises in Reany et al., Philippi de Vitriaco Ars nova, 2331.
See comments in Fuller, A Phantom Treatise, 256.
9
Orbis orbatus is the more verbose voice but Vos pastores is usually higher.
10
Two versions exist, one edited as the Ars perfecta of Vitry in Coussemaker, Scriptorum III:2835, and
the other in Angls, Dos Tractats Medievals, 610. The above excerpts are from the former. The latter
reads Et de omnibus veritas patet in tenore de In arboris. Innova fer anims, and two examples follow; Ibid.,
10.
11
Ed. Balensuela, Ars cantus mensurabilis.
72
TRACTATUS FIGURARUM (last quarter 14th century)
12
Apta/Flos ut patet in Apta caro
Tribum/Quoniam in motetis ipsorum magistrorum videlicet Tribum que non ab-
horruit et in aliis et cetera.
Rex/Leticie et in Rex carole, ac etiam in aliis motetis (34)
QUONIAM CIRCA ARTEM MUSICALIS SCIENCIE... (after 1400)
13
Mon chant/Qui doloreus ut monachant de morte wilhelmi (336)
Apollinis/Zodiacum ut Apollinis (336)
Degentis/Cum vix Degentis vita (336)
Tuba/In arboris in tenore Tube sacre dei (336)
NOTITIA DEL VALORE DELLE NOTE (early 15th century)
14
Sub Arturo/Fons come el tinore di certi motetti, cioe Luce clarus o Sub Arturo o
Omni habentiSub Arturo a tre taglie di valore (57)
DIFFERENTIA EST INTER MOTETOS (early 15th c., Italian)
15
Ida/Portio [I]da capillorum
Impudenter/Virtutibus Impudenter
TRACTATULUS DE CANTU MENSURALI (1462)
16
Degentis/Cum vix ut Degentis vita (16)
in cantione Degentis vita (29)
Apollinis/Zodiacum utApollinis (16)
12
Ed. Schreur, Tractatus gurarum.
13
Ed. Wolf, Ein Breslauer Mensuraltraktat, 33145.
14
Ed. Carpentyan, Notitia del valore delle note.
15
Ed. Staehelin, Beschreibungen und Beispiele, 23742.
16
Ed. Gallo, Tractatulus de cantu mensurali.
73
CHAPTER TWO
Hearing Voices
One of our most important faculties is our ability to listen
to, and follow, one speaker in the presence of others. This is
such a common experience that we may take it for granted;
we may call it the cocktail party problem. No machine has
been constructed to do just this, to lter out one conversa-
tion from a number jumbled together.
Colin Cherry, 1957
1
I saw how in a large gathering of discerning people, when
motets were sung according to the new manner, it was asked
what language the singers were using, Hebrew, Greek, or Lat-
in, for it was not understood what they were saying.
Jacobus de Montibus (olim Leodiensis), c. 1330
2
W
ITH THE GROWTH OF COMMERCIAL AVIATION in the 1950s, air trafc controllers
encountered an unexpected problem: presented with the voices of multiple pilots
coming out of the same loudspeaker, they had trouble keeping these different streams of
information separate, and were thus at risk of misunderstanding important messages. Their
difculties led to the rst research on what has since been termed the cocktail party effect
or cocktail party problemthe ability of people to isolate one sound-source in the pres-
ence of many. The brains of most humans are able to navigate complex social soundscapes
without much conscious effort. But how exactly we do this, and why it should have been
difcult for the air trafc controllersand later for computersconstituted the problem.
3
1
Cherry, On Human Communication, 278.
2
Bragard, ed., Jacobi Leodiensis Speculum musicae 7:95; this passage is discussed in detail below. On the
dating of the treatise and the evidence for Jacobus de Montibus as opposed to Jacques de Lige as the
treatises author, see Desmond, New Light on Jacobus, 2436.
3
This effect has been an object of study in physiology, neurobiology, psychophysiology, cognitive psychol-
ogy, biophysics, computer science, and engineering. For a review, see Haykin & Chen, The Cocktail Party
74
For musicologists, a parallel effector perhaps problemis inherent in the poly-
textual soundscapes of medieval motets, whose superimposition of multiple strands of text
has led us to inquire whether audiences were meant to understand any of the words during
performance. Scholars have answered this question in various ways, some insisting that the
texts of motets were not meant to be understood in performance, others suggesting that
only select words or phrases were meant to be perceived by the listener. These stances about
intelligibility have in turn led to varying views of the genre, the compositional priorities of
its composers, and the role of text in its aesthetics. In the chapters that follow, I discuss a
number of examples from the ars nova motet repertory in which compositional choices stem
from the content of the texts. These show that motet texts were meaningful to their com-
posers. But I will also suggest that many of those compositional choices resulted in musical
events that would have been audible to listeners familiar with the generic norms of motets.
These listeners could only connect these compositional choices with the semantic stimuli
that inspired them if they were able to hear and understand some considerable amount of
the motets texts. For this reason it will be necessary for me to engage here with the ques-
tion of intelligibility. I will rst address the small amount of medieval evidence that bears
on the topic, including the famous observation by Jacobus quoted at the head of this chap-
ter. I will then engage with musicological literature pertaining to this problem, especially
with the arguments put forward by Christopher Page in Discarding Images. This in turn will
lead to a consideration of the performance aesthetics that underlie assumptions about the
intelligibility of motet texts. Finally, I will return to the cocktail party effect to see what
cognitive science can tell us about the listening conditions under which motet texts might
be understood.
Problem.
75
MEDIEVAL LISTENING PRACTICES AND MODERN EARS
Before moving further with the question of intelligibility in polytextual perfor-
mance, it should be noted that some motet poetry may have also been heard in monotextual
contexts. In the case of Vitrys motets, literary circulation of the texts must have given
some the opportunity to study the poetry on the page, and perhaps to commit it to memory.
There is also the possibility that motet voices were sung separately rst and then together.
4
Indeed, Jacobus refers to such a practice in the Speculum musicae.
5
I have already discussed
this passage and its broader context in Chapter 1, but it will be worth repeating the relevant
part here:
The discanting voices can either be compared to the tenoror they can be
considered by themselves, not with respect to the notes of the tenor and sung
at the same time as them, but separately and in turn, one after the other, as
when someone sings some motetus or triplum or quadruplum by himself without a
tenor.
6
Clearly the practice of singing voices on their own was nothing too out of the or-
dinary, since Jacques is citing it here not for its own sake, but to support a broader claim
that the voices of polyphonic compositions can each be considered as independent songs
on their own terms. Further evidence of this independence comes from the interpolated
Fauvel. There, the triplum voice of the motet Floret/Florens, which survives in the Brussels
4
This approach is taken to good effect on several Gothic Voices recordings and posited by Page in a foot-
note: It must be acknowledged that a layered performance of motets (each voice being sung individually,
perhaps over the tenor) followed by a complete performance may have taken place in the 13th c., a method
which solves the problem of comprehensibility at a stroke, Discarding Images, 85n73. Page has also sug-
gested that before performing a motet, the singers might have summarized the contents of their voice for
the audience, but this conjecture is supported by no evidence pertaining to the motet; Listening to the
Trouvres, 648; Around the Performance, 351.
5
Page makes reference to this passage in Around the Performance, 351, 356n14.
6
Possunt autem voces discantus ad voces comparari tenorisvel possunt per se considerari, non per re-
spectum ad voces tenoris et ut simul dicuntur cum illis, sed divisim successive una post aliam cum aliquis per
se cantat motetum aliquem, triplum vel quadruplum, sine tenore, Bragard, ed., Jacobi Leodiensis Speculum
musicae 7:9.
76
rotulus and Cambrai fragments, is reworked into a monophonic prose beginning Carnalitas
luxuria.
7
Though some of the rhythms have been changed, the long rests that originated in
the motet remain in the prose, as do several leaps of a fourth. Alice Clark has noted that
no concession is made in the prose melody to the fact that this voice, originally the top
part in a three-part texture, is now a monophonic piece that must stand alone.
8
Indeed, the
re-working seems to prove Jacques claim. In theory, and here in practice, an upper voice is
as much a song as the tenor.
And there is evidence from the literary realm that motet voices were sometimes
sung on their own: as the narrator of Froissarts Joli Buisson de Jonece (1373) walks along in
a courtly procession on his way to see the god of Love, he gives voice to his joyful mood by
singing a new motet that he had been sent from Reims (En chantant un motet nouviel/
Quon mavoit envoiiet de Rains).
9
Troubled probably by this solo performance and the late
date for a motet nouviel from Reims, Earp has suggested that the term in this context
may refer to a new polyphonic ballade or rondeau, the sort of work that to a non-musician
would surely have sounded as complicated and learned as a motet.
10
But it seems likely
that Froissart, who himself composed in the formes xes, would have been able to tell the
difference. Rather, the passage seems to attest to the spontaneous solo performance of a
single motet voice.
7
The prose is edited in Harrison, The Monophonic Music, 18890. Floret/Florens in Sanders, The Early
Motets, 3745, and Lerch, Fragmente aus Cambrai, 2:20514. For a hypothesis on why the prose and not
the motet appears in Fauvel, see Bent, Fauvel and Marigny, 467.
8
She therefore posits that the original plan was to notate a two-voice composition, but the editors ran out
of space. Clark, The Flowering of Charnalit, 183.
9
On this source and the motets to which it may allude, see Huot, Reading across Genres. See also discus-
sion in Chapter One of the present study, and below.
10
Guillaume de Machaut, 54.
77
And yet these references to voices sounding on their own do not obviate the need
for questions about intelligibility. It is unlikely that all listeners would have had access to
all motet texts ahead of hearing the works performedindeed, motets not by Vitry seem to
have been transmitted as poetry very rarely. Nor can we assume that all performances began
with turn-by turn sing-throughs. And even if they did, it is still worth asking what audiences
heard (were expected to hear, or were able to hear, or chose to hear) in a tutti performance.
The most frequently cited medieval evidence that there might have been problems
with comprehension when motets were sung is the anecdote from the Speculum musicae
cited at the head of this chapter. Jacobus seems to refer to problems arising from polytex-
tuality when he reveals that a listener at a gathering of discerning peopleasked what
language the singers were using: Hebrew, Greek, or Latin. If the listener could not even
understand what language he was hearing, its seems impossible that he could have under-
stood the words. Thus the passage has been used to support the claim that the very nature
of double motets with their simultaneous presentation of two texts jeopardized the message
of either [text].
11
But an examination of this comment in context serves to give an entirely different
impression of its meaning. In this 46th chapter of Book Seven of the Speculum musicae,
Jacobus undertakes a comparison of two sets of thingsthe old art of composing and the
new, and also the old manner of singing and the new (comparatio veteris artis musicae
mensurabilis ad modernamet antiqui modi cantandi ad modernum).
12
There is no im-
plication that it would be possible to perform the old kind of music in the new mode of
11
Pesce, The Signicance of Text, 912.
12
Ed. Bragard, Jacobi Leodiensis Speculum musicae 7:93.
78
singing, and for the most part Jacobus keeps these two dimensionsthe compositional and
the performativecoupled. But towards the end of the chapter he turns his attention to
performing specically, and it is at that point that the discussion of text and its intelligibil-
ity takes place. The passage is worth quoting in full:
In a certain gathering in which skilled singers and discerning laypersons
were gathered together and where modern motets were sung according to the
modern manner, and some old [motets] as well, I observed that the old mo-
tets and also the old manner of singing were more adequately pleasing, even
to the laypersons, than the new. And even though the new manner [initially]
pleased in its newness, this is no longer the case, for it begins to displease
many.
Therefore let the old songs and the old manner of singing and notating be
called back to the native land of singers. Let these things come back to use,
and let the rational art ourish once more. It has been exiled, along with
its [corresponding] manner of singing, as though violently thrown down by
the fellowship of singers. But the violence need not be perpetual. To whom
does such wantonness of singing give pleasurethis curiosity in which, as
may be seen by anyone, the words are lost, the harmony of the consonances
diminished, the length of notes changed, perfection disparaged, imperfec-
tion elevated, and measure confounded?
I saw how in a great gathering of discerning people, when motets were sung
according to the new manner, it was asked what language the singers were using,
Hebrew, Greek, or Latin, for it was not understood what they were saying. Thus
the moderns, although they write many beautiful and good texts in their songs, lose
them in their manner of singing, since they are not understood.
13
The nal sentence of the quoted text, which serves as a summary of what we should get
13
Vidi ergo, in quadam societate, in qua congregati erant, valentes cantores et laici sapientes. Fuerunt ibi
cantati moteti moderni et secundum modum modernum, et veteres aliqui. Plus satis placuerunt, etiam laicis,
antiqui quam novi, et modus antiquus quam novus. Etsi enim modus novus in sua placuerit novitate, non
sic modo, sed incipit multis displicere. Ergo iam placeat cantus antiquos et cantandi notandique modum
antiquum ad patriam revocare cantorum. Revertantur haec ad usum reoreatque rationabilis ars. Exul fuit et
eius cantandi modus. Quasi violenter deiecta sunt haec a cantorum consortio. Violentum autem non debet
esse perpetuum. Ad quid tantum placet cantandi lascivia, curiositas in qua, ut aliquibus videtur, littera per-
ditur, harmonia consonantiarum minuitur, valor notarum mutatur, perfectio deprimitur, imperfectio sub-
limatur mensuraque confunditur? Vidi in magna sapientium societate, cum cantarentur moteti, secundum
modernum modum, quesitum fuit, quali lingua tales uterentur cantores, Ebrea, Greca, vel Latina, vel qua
alia, quia non intelligebatur quid dicerent. Sic moderni, licet multa pulchra et bona in suis cantibus faciant
dictamina, in modo tamen suo cantandi cum non intelligantur, perdunt ea. Bragard, ed., Jacobi Leodiensis
Speculum musicae 7:95; emphasis mine.
79
from the anecdote, is crystal clear in its meaning: its too bad that the well written texts of
the moderns are masked by the new manner of performance. The implication, then, is that
if the same good texts were delivered in the old manner of singing, they would not be lost. In
complaining that new texts are obscured Jacobus tacitly acknowledges that the clear pro-
jection of text is a merit of the older repertory that he champions. Far from proving that
medieval audiences could not understand the texts of motets in performance, then, this
passage in fact suggests that there was an expectation on the listeners part that text would
be intelligible.
Nor does Jacobus necessarily men for us to take him fully at face value. As Suzan-
nah Clark has pointed out, Jacquess complaint that they could be singing in Hebrew or
Greek is, of course, spurious, for the upper voices of motets were either in Latin or French
or both.
14
Hence the observer is making fun of a delivery of Latin text which was, to his
mind, muddled as a result of the new manner of singing. The listeners mocking comment
would hardly makes any sense unless understanding text was an aspect of appreciating the
modus antiquus cantandi.
What does this passage tell us about the ars nova and new manner of singing? Not
much. Jacobus is writing to persuade, and the relation of a fellow conservatives negative
evaluation of the performance is intended to malign the modern manner rather than to
give us insight into its aesthetics or the expectations of its adherents. It is certainly possible
that for Jacobus and similar listeners familiarity with, and a preference for older textures
and norms might have resulted in a greater facility of comprehension of texts set in the
older style. Specically, the increased use of duple meters may have made melodies and text
14
Sen dirai chanonete, 32n6. Frank Hentschel also points out the humor in the remark, seeing it to
reect back on the apparently discerning persons who listened. Der Streit um die ars nova, 120.
80
delivery seem stilted enough to obscure meaning for those accustomed to perfection. But to
an unbiased listener it is hard to believe that the words of ars nova motets would be harder
to understand than those of older works, since they feature more transparent textures.
Even if Jacobus is not a trustworthy source for reporting on ars nova performances
in any but a negative way, he clearly tells us something about contemporary expectations.
Undoubtedly he and the learned person whom he paraphrases would agree that in an ideal
situation the texts of motets should be intelligible. Furthermore, he gives us the key to un-
derstanding why they are sometimes not. For towards the end of the passage he uncouples
composition and performance, focusing on the moderns manner of singing as the cause
of incomprehension. In doing so, he reminds us of an obvious but important fact: perfor-
mances can make words audible, or they can obscure texts, causing them to become lost.
And it is here that we can turn from the new manner to a newer manner still, and consider
the effect of twentieth- and twenty-rst century performances of ars nova motets on the
intelligibility of their texts.
* * *
One of the most inuential and controversial arguments about the audiences for
medieval motets was put forward by Christopher Page in his 1993 Discarding Images. Of-
fering a new reading of the often-quoted lines from Grocheios De Musica (c. 1300), which
seem to locate motets within elite intellectual spheres, Page argued that motet audiences
included a much broader cut of society: Johannes de Grocheios clergy, and those who care
for the renements of any skill may have formed a very diverse group indeed.
15
15
Discarding Images, 111. The main tenets of this important study been widely discussed and critiqued in
print, and it is not the purpose of this chapter to rehearse those argumentsarguments about the writing
of history, periodization, and the role of scholars and performers in interpreting medieval music, broadly
81
But while diversifying the intended audience for ars antiqua motets, Page has nar-
rowed the listening experiences available to this group, downplaying the intellectual aspects
of listening and shifting the focus to sensory pleasure. Objecting to the reading of mere
contrast as opposition, he argues that in fact while motets may present us with differ-
ences in diction or topic between their texted voices, these do not result in the destabilizing
quality that much modern criticism tends to discern and explore.
16
In the process, motets
emerge as integrated structures, their separate voices merged into a unied whole by con-
gruences in theme, vowel-sound, and diction.
Such an aesthetic certainly has implications for the importance of individual
textsindeed, of any textas a salient aspect of composition or reception. Indeed, Page
suggests that part of the pleasure gained from motets by their diverse medieval audiences
may have been derived from not understanding the texts:
Some listeners in the thirteenth century may have enjoyed the sound-pat-
terns of Latin motet poetry without deriving any signicant understanding
of the sense, much as some modern listeners do today when hearing these
motets (or operatic arias, for that matter)it can scarcely be doubted that
many motets were designed to induce an exhilarating impression of words
leaving sense behind and beginning to skirl.
And later:
The existence of triple motets with three simultaneous poems that some-
times deal with subjects in two languages, French and Latin, proves that the
aesthetic of the poet is one which allows verbal communication to decline
as metrical, musical, and structural ambitions mount. This was surely not
taken to involve a sacrice of meaning but rather a gain of pleasure that was
intellectual in this sense: it produced the exhilaration of knowing that a
piece contains more than one can ever hope to hear.
17
construed. See, in order of publication, Bent, Reections on Christopher Pages Reections, Page, A Reply
to Margaret Bent, Strohm, How to Make Medieval Music Our Own, Wegman, Reviewing Images, and
Weller, Frames and Images.
16
Discarding Images, 92, 96.
17
Ibid., 856, 101.
82
Pages claims and supporting arguments pertain mostly to ars antiqua motets, which
were the subject of Grocheios observations. And it is in this repertory that he locates a
number of traits that argue against the independence or intelligibility of texts, such as the
coordination of rhyme-sounds and vowels between voices. As additional evidence against
an intellectual reception he cites what he perceives as a lack of engagement with the times
intellectual trends (the scholastic method, for example, has left virtually no impression
upon the Latin or French motets).
18
But is it is also clear that Page sees the same problems as extending beyond the ars
antiqua, to the motet:
Of all the musical genres known to the Middle Ages, it is the motet which
most candidly acknowledges the importance of verbal sound over verbal
sense by placing two or even three texts together, minimizing their intelligi-
bility but maximizing their phonic contrast.
19
In fact ars nova motets are much less likely to have the moments of sonic unity that Page
sees in those of the thirteenth century. And a wide cross-section of current intellectual
trends is indeed represented by their texts, which engage in contemporary theological de-
bates, incorporate quotations from learned sources and even treat such esoteric subjects as
metaphysics and poetic theory.
20
But in one aspect ars nova motets are open to the kinds of
18
Ibid., 84.
19
Discarding Images, 856. See also the evaluation of Machaut on p. 14: Machauts music leaves no doubt
that his sensations when composing were as indifferent to moral or intellectual persuasions as those of any
composer at any period in history when genuinely engrossed. It should be said that in general Pages argu-
ments are more persuasive for the ars antiqua motet, which is after all his subject. In some passages, especially
in the longer section surrounding the passage quoted above, he seems to shift from thirteenth-century lis-
teners to the medieval motet almost seamlessly.
20
For an argument in favor of the Assumption as doctrine, see the triplum Almifonis melos cum vocibus, ed-
ited and summarized in PMFC 5, supplement, p. 8. On classical quotations in Vitrys motets, see Bent, Po-
lyphony of Texts and Music, and Wathey, Myth and Mythography. Metaphysics is criticized in ll. 1931
of the triplum Tuba sacre dei (PMFC vol. 1, 324). On In virtute/Decens and its engagement with Horaces
Ars poetica, see Chapter Five.
83
criticisms leveled by Pagethey absolutely combine diverse texts in one work.
Some arguments that motet texts may have a semantic as well as a sonic role in
performance have been put forward. Dolores Pesce has argued that in double Latin motets,
music serves texts by giving musical prominence togures of repetition and rhetorical
expressions, and that in some cases this kind of emphasis may represent at least an attempt
to give musical expression to the words.
21
And in a recent study of the Montpellier motet
Joliement en douce/Quant voi/Je sui joliete/Aptatur, Suzannah Clark has drawn attention to a
variety of ways in which the music claries rather than obscures the meaning of the text
in performance.
22
Beginning from the premise that hearing the odd word here and there
is less a matter of chance than one of effort on the part of poet-composers to convey the
text, she has argued that some level of understanding is made possibleeven assuredby
compositional choices such as the placement or refrains, the use of musical echoes, and the
careful coordination of rests between voices.
23
And yet even here the matter of understand-
ingalbeit a rather deep understandingdepends upon grasping on the listeners part.
24
Other scholars have agreed with Pages claims about the impossibility of hearing in-
dividual texts while nevertheless preserving a place for an intellectual dimension. Margaret
Bent, the main advocate for the importance of text to the analysis of motets, concedes to
an impossibility of understanding [motet texts] at a rst or unprepared hearing, but argues
that there is space for the prepared listeneran experienced listener who, like Boethiuss
musicus, exhibits the faculty of forming judgments according to speculation or reason and
21
Pesce, The Signicance of Texts, 94, 97.
22
Clark, Sen dirai chanonete, 31.
23
Ibid., 34.
24
Ibid., 589.
84
is already familiar with the composition: instant understanding is impossible, prepared
understanding possible; this applies to many kinds of music.
25
There is no doubt that a listening experience such as the one suggested by Clark
full of intertextual and intermelodic referenceswould be meaningful. No doubt also that
the understanding of a piece by a listener prepared in Bents sense of the termintimately
familiar with a motets texts, proportions, and numerical structures, as well as the liturgi-
cal context of its tenorwould be deep. But must instant understanding of the text always
be impossible for the audiences of a polytextual work? Reminded by Jacobus how much
depends on the manner of singing, I would like to evaluate one aesthetic underpinning of
Pages incredulity about intelligibility: the question of timbre.
TIMBRE
Pages debt to recordingsespecially to his own work with Gothic Voicesis self-
avowed:
The primary inspiration for [Discarding Images] has been provided by per-
formancerecordings are sometimes superseded by advances in knowledge,
and are often vanquished by changes in taste, yet innovative or challenging
performances can none the less disturb a wide range of preconceptions that
we may unwittingly hold about the interest and scope of a repertory.
26
And of course he is not alone in this. I am among those for whom Pages beauti-
ful recordings with Gothic Voices have felt deeply authentic. And there can be no doubt
that they are closer to medieval performances than many previous modern re-creations in
one sense: inspired by his research into the question of instrumental versus vocal perfor-
25
Bent, Reections, 632, and Words and Music, 387. Reading the latter article makes us a listener
prepared, in Bents sense, to hear Machauts Fons/O livoris (M9).
26
Discarding Images, p. xx.
85
mance of medieval song, Pages recordings allowed us to reconceive of a wide repertory of
music as primarily vocal.
27
Lawrence Earp evokes the feelings of many when he writes that
the recordings of Gothic Voices are consistently of the highest quality, and put into prac-
tice the most advanced and solidly grounded views of performance practice at the time of
recording.
28
But if in the respect of instrumentation Pages re-imagination of the medieval
sound was fresh and radical, another aspect of his aestheticthat which pertains to tim-
breseems to have been inherited perhaps too readily from the English choral tradition.
This is not to say that Page was not thoughtful about questions of timbre. If, as John
Potter has argued, the position of authority [held by the British early music sound] has
been achieved with the minimum change to the techniques that the singers learned to gain
their choral scholarships, this is perhaps less a matter of laziness (certainly in Pages case)
than of agenda. For Page was reacting to the bright, mixed sound of instrumental perfor-
mance current in the rst stage of the early music revivala sound, in his words, of boldly
individualized lines collaborating in a polyphony which is extrovert and almost heraldic in
colour, which is candid, even naive, in its directness of address to the listener.
29
The ascetic
English choral sound evocatively described by Daniel Leech-Wilkinson as freshly cleaned
Anglicanism must have seemed like the perfect antidote.
30
This pure sound with its equal timbres, matched vowels, and lack of vibrato could
not but inuence Pages conception of the motet. As his performance aesthetic blends
sounds together into a unied whole where sonority reigns supreme, so his arguments about
27
See Page, Machauts Pupil Deschamps, and The Performance of Songs.
28
Earp, Review of The Medieval Romantics, 289.
29
Page, The Performance of Songs, 441.
30
Leech-Wilkinson, The Modern Invention of Medieval Music, 206.
86
ars antiquaand ars novaaesthetics stress unity over contrast. For me, synchronization
of vowels and verbal identity that lead to sonorities of exceptional clarity and forth-
rightnesscharacteristics which Page argues are inherent to the compositional fabric of ars
antiqua motetsbring to mind an Oxford college choir, or the Tallis Scholars.
31
Theirs is
a sound in which we may [enjoy] the sound-patterns of Latin motet poetry without deriv-
ing any signicant understanding of the sense.
32
And in the phrase there is variation and
contrast, but there is neither conict nor tension, which Page uses to describe the relation-
ship between the motetus and triplum in a motet, I cannot help but read also a description
of the sound-world cultivated by Gothic Voices.
33
Margaret Bent has pointed out the danger
of drawing circular conclusions from particular styles of performance that emphasize some
rather than other features of the music.
34
And indeed, Pages arguments about the priori-
ties of the genre could double as an aesthetic manifesto: The complexity is purely phonic
and not semantic. As I read them, the texts of this motet are designed to be so similar to
one another that there is little or no contrast between them.
35
The texts, or the voices that
execute them?
Page has himself cautioned that his reconstructions of medieval sounds are not
wholly objective in part because they are based uponexperience with professional Brit-
ish singers, most of whom received their training in the choir of some cathedral or collegiate
31
Discarding Images, 103, 105.
32
Ibid., 86.
33
Ibid., 97.
34
Bent, Reections on Christopher Pages Reections, 630. On the question of authenticity and its role
in Pages argument see also Wegman, Reviewing Images, 26970.
35
Page, Discarding Images, 110.
87
establishment in England.
36
But we are to understand that the reconstructed sound-world
is at least mostly objective, and Page has cited various kinds of evidence to support the idea
that the late-medieval sound might be similar to the modern British one. For example, the
similar range of motet voices is cited as evidence that motets were performed by identical
(or by nearly identical) voices:
Many of the motets in the Bamberg manuscript (probably dating from c.
1285) exploit a range of around eleven notes from top to bottom, while
some individual voice parts also employ this compass. It would therefore
seem that 13th-century motets were conceived to exploit the resources of
two, three or four equal voices; one singer, in other words, could normally
perform any part in a motettenor, duplum, triplum or quadruplum
without undue difculty. The result would have been a well-blended sound
with each part distinguished by its text rather than by its colour.
the evidence of voice ranges suggests that blend, and not contrast, is the
guiding colour principle of the Ars Antiqua motet. This is quintessentially
music for singers, designed to exploit the sounds of Latin and Old French
as put into song by voices of very similar type singing in very much the same
kind of way.
37
Such a reading of the evidence blatantly conates range with tessitura. The compass of a
part is hardly indicative of where that part lies for most of a composition, but it is the latter
information that would indicate which singer is to be placed on which voice. Furthermore,
even if singers had similar usable ranges, it seems likely that some would be more used to
singing tenors, and some, upper parts. This distinction would be cemented in the following
century with the emergence of function designations such as tenorista.
38
But most importantly, nothing guarantees that two medieval singers, even if they
36
The Performance of Ars Antiqua Motets, 149.
37
Ibid., 1534.
38
On tenorista as function rather than range, see Wegman, From Maker to Composer, 4458. In con-
trast, contratenorista seems to have been a range designation for bass voices. I am grateful to Alejandro
Enrique Planchart for discussing this issue with me.
88
had the same range, would necessarily have voices of very similar type. The unity of sound
achieved by English singers is the result of a huge system of choir schools and choral schol-
arships, with many professional performers emerging from either Oxford or Cambridge.
39
Arguably, only such an infrastructure can produce singers so similarly trained as to per-
form music in this way. This is not to say that those singing motets in the thirteenth and
fourteenth centuries were not experienced the degree of specialized knowledge needed
to sing mensural notation was great, and all singers able to execute motets would probably
have begun as choirboys or, in the case of the nobility, would have been taught by clerics.
40
But while the performers would all have been schooled, they would by no means have been
products of the same school. We can imagine a scenario where a Latin motet was sung by
three university men, each from a different part of Europe and with a different way of
pronouncing Latin. Would their vowels be matched, or their training be similar? We know
that even different regions within France had their own local ways of singing: the canons at
Lyons cathedral, for example, were known to shout.
41
On the international level, these dif-
ferences were apparently worthy ofand subject torivalry and ridicule. Thus an Italian
rhetorician wrote in 1226 that
the Greeks say the Latins bark like dogs and the Latins say the Greeks growl
like foxes. The French claim that the Italians groan the Italians, on the
other hand, say that French and Germans emit tremulous sounds like some-
one suffering from fever, and that they sing so loudly they must think God
is deaf.
42
39
Potter, Vocal Authority, 11617.
40
Certainly there was some non-clerical performance of mensural polyphony, as indicated by the depic-
tions of lay singers in the Machaut manuscripts (e.g. the singers gathered around a barrel, discussed below).
In the Voir dit the narrator sends notated music to the noble Toute Belle, expecting her to perform it. For
further evidence that women performed mensural polyphony, see Page, A Treatise on Musicians.
41
Page, Around the Performance, 348. See also the contemporary comments about French and Italian
singers collected in McGee, The Sound of Medieval Song, 178.
42
Vecchi, Musica e scuola, 26673; trans. Page, Around the Performance, 348.
89
But motets were performed in precisely those places where singers from different corners
of Europe would have mingledin universities, at courts, and during occasions of state.
Would not some performances have had in them, then, a trace of those exaggerated differ-
ences? If not exactly emitting howls, groans, and feverish wails, it would nevertheless be
surprising if such diverse performers could be found singing in very much the same kind
of way.
Other scraps of evidence support the idea that the sound-world of motets may have
been a varied one. One unlikely source of insight comes from a description of dog-song in
Gace de la Buignes Roman de Deduis (c. 135977). In arguing that hunting with dogs is
more noble than hunting with falcons, the allegory Love of Dogs alludes to the pleasure of
hearing dogs singin fact, he describes them singing a motet:
Les uns vont chantans le motet,
Les autres font double hoquet.
Les plus grans chantent la teneur.
Les autres la contreteneur.
Ceulx qui ont la plus clere guile
Chantent les tresble sans demeure,
Et les plus petis le quadouble
En faisant la quinte sur double.
43
[Some of [the dogs] go singing the motet; the others make a double hocket:
the largest [dogs] sing the tenor and the others the contratenor. Those that
have the clearest voice sing the triplum without delay and the smallest [sing]
the quadruplum, making a fth on the duplum.]
44
To be sure, this is hardly informative from a performance practice perspective. But the vari-
ety of look and sound cultivated in this passage is still compelling. The parody evokes a dis-
parate group producing sounds that vary both in timbre and ambitus: the dogs singing the
43
La Buigne, Le Roman de Deduis, ll. 8081-8; Page has cited this passage in support of all-vocal perfor-
mance of chansons: Machauts Pupil Deschamps, 487.
44
Trans. Leach, Sung Birds, 1812.
90
triplum have penetrating voices, the big dogs sing tenor, and the quadruplum is left to the
puppies. This variety is the more signicant when we remember thatas Page has noted
many motets are relatively narrow in range, often covering no more than a twelfthand
this in turn can be the range of a single voice in a more virtuosic work such as a chasse.
45
Thus Love of Dogs presents us with a more varied picture than we might expect (if we have
any expectations of canine motet performance).
This variety is not dissimilar from that cultivated in the image of performance
which opens the motet section of Machauts manuscript A (see Figure 2.1). Here a mix of
clerics and laypersons is gathered around a wine cask, drinking wine and singing a motet
from a scroll.
46
Their diverse ages, different backgrounds, and presumed state of inebria-
tion (have they drained the barrel?) all but guarantee that their voices will not sound similar.
This group is a far cry from the singing angels of van Eycks Ghent altarpiece, whose young
identical faces seem to imply the kind of sound we associate with the English aesthetic. This
diverse group of drinkers hints instead at a smorgasbord of timbres and volumes.
45
See Kgle, The Manuscript Ivrea, 153n134.
46
That the composition being performed is a motet is clear not only from the miniatures position at the
head of the motet section, but also from the words tenor and dame that are visible on the scroll. This
image is usually interpreted as a group of singers (and drinkers)see descriptions in Huot, From Song to
Book, 3001, Earp, Guillaume de Machaut, 187, Wimsatt, Chaucer and His French Contemporaries, plate 13,
and Leo, Authorial Presence, 44. Kgle has argued that the four men on the right are singing, while those
on the left (a layman, a cleric, and a servant) are drinking and listening; Die Musik des 14. Jahrhunderts,
361 (note that the image descriptions and images are reversed, so that the description labeled bild links
pertains to the motet image, which is on the right). Kgles analysis is convincing as pertains to the direction
of the scroll, but the nobleman at left does seem to be keeping tactus on the arm of the cleric to his left, and
the latters holding the scroll seems to imply his performative participation.
91
Figure 2.1: A group of men around a wine keg singing a motet
47
Granted, this too is not the kind of evidence we would like. It is certainly possible
that rather than recording anything akin to a performance, the image is serving as an al-
legorical depiction of motets, with their mixture of the sacred and secular.
48
And Love of
Dogs is at least as biased an observer as Jacobus de Montibus. But choices were made in both
casesthe choice to use a motet as the example to be sung; the choice to depict a diverse
group as singing it. Also noteworthy is the joyful, easy mood that characterizes both depic-
tions. Nothing could be more natural, Love of Dogs wants us to believe, than this singing,
and there isnt a responsory or an alleluia, even if it were sung in the Kings chapelthat
would cause the same tremendous pleasure that one experiences in hearing such a hunt.
49
In this context we may also return to the above-mentioned passage from Froissarts Joli
47
Machaut MS A, fol. 414v.
48
The same iconographic motifsingers around a wine kegis also used to illustrate the singing of a
rondeau in Machaut MS E (fol. 16). Here the singers seem all to be laypersons, but they still vary in age and
class. See Kgles discussion of the differences between these two images in Die Musik des 14. Jahrhun-
derts, 361. Dominic Leo also discussed this image in Singing Around a Wine Keg.
49
Trans. Leach, Sung Birds, 181. Though the sound is natural, it is not rational, and Leach has argued that
this passage is part of a broader discourse about vox confusa and vox discretathe dogs are cantores but not
musici; ibid., 2129.
92
Buisson de Jonece, in which a lover joyously sings a motet while walking to meet the god of
love. As he sings, the lover is ideally situated in mood and position:
En cheminant sus ce voyage
En pais en joie et en reveil
En chantant un motet nouviel
Quon mavoit envoiiet de rains
Premiers nestoie ne darrains
Mais en mi lieu par grant solas
Pares duns noes solers a las
Ensi quamant vont a la velle
On me boute et lors je mesvelle.
[As I traveled along on this excursion, in peace, joy, and gaiety, singing a new
motet that had been sent to me from Reims, I was neither in the front nor in
the back, but very comfortably in the middle, dressed in new lace-up shoes,
the way lovers go for a late night out.]
50
We can almost hear this performance. As the lover looses his song, it is with joyful ease, al-
lowing the sound of his voice to reect his general euphoria.
51
I think it is fair to say that some of the joy and spontaneity that seems to have been
an aspect of motet performance is lost when a unied sound is the goal. Indeed, Page has
written that singing motets should be difcult, even uncomfortable for the singer:
Something is often to be gained by pitching these motets in a way that lifts
the singers on the upper parts away from the comfortable range where they
can coast or croon. When the music is lifted above this range the relatively
higher pitch brings many things under closer control because the vocal cords
are vibrating more quickly and because the singers, recognizing that danger
is only about a tone away, begin to work hard in a fashion that is always ben-
ecial to this music.
52
Elsewhere, the same author has written that in the Middle Ages to succeed [at composi-
tion] was to create something that singers could control and dominate, but here the music
50
Ed. and trans. Figg and Palmer, Jean Froissart: An Anthology, 4623.
51
The lovers ease and happiness here are all the more important because in the next line he is woken up
from his happy dreamsee the discussion in Chapter One.
52
Page, The Performance of Ars Antiqua Motets, 158.
93
is instead allowed to control the singers, coaxing a unied sound from them with the threat
of nearby danger.
53
The reason that all of this is pertinent to an argument about intelligibility is that
timbre is one of several variables that allows the human ear to focus on one sound-source
in the presence of others. A number of other aspects of performancesincluding acoustics,
an audience members ability to see the performer, and the role of recording technology
also bear on this issue. In what follows I will not suggest that performance conditions which
maximize intelligibility are either authentic or desirable. I will only review the evidence
provided by psychology on the acoustic and cognitive phenomena that allow a person to
navigate a busy soundscape whether it be a cocktail party or a gathering of discerning
people.
THE COCKTAIL PARTY PHENOMENON
Given the high likelihood that a motets intended audience would hear multiple
performances of the same work, the question of whether it is possible to hear both texts of
a polytextual motet at the same time wanes in importance next to a more basic inquiry: Is
it possible to hear any text in a polytextual setting, or does the act of combining more than
one text in one sonic space necessarily render any and all of those texts unintelligible?
54
53
Cf. Jerome of Moravias advice that a song never be begun very high, especially by those possessing head
voicesbut they should always begin moderately, that is sing; that is so that the song may not rule the voice,
but the voice rule the song; beautiful notes cannot be produced otherwise, trans. Rosenfeld in McGee, The
Sound of Medieval Song, 28.
54
Given the relatively small size of the repertory it seems likely that works would have been performed
multiple times. Occasional works would perhaps have been an exception to this, but they too seem to have
been reused, as in the case of Servant regem/O Philippe and Servant regem/Ludovice. Robertson has also argued
that Machauts Bone pastor/Bone pastor, though seemingly an occasional piece, was in fact intended to be an
archbishop motet in the widest possible sense, in which Guillerme could be replaced with another name;
Guillaume de Machaut and Reims, 60.
94
Rephrasing the question claries the link between the psychological cocktail party phenom-
enon and the ars nova motet. Whether in a busy room where the person behind you may
be speaking louder than your collocutor, or in a motet where the triplum is singing higher
and faster than the motetus which you are perhaps desirous of hearing, the cognitive task
is the same: to isolate a sound-source of interest in the presence of other, sometimes more
prominent ones. But the assumptions have been very different in the two elds: in psychol-
ogy, it is readily agreed that the cocktail party phenomenon is one of our most important
facultiessuch a common experience that we may take it for granted.
55
And indeed,
walking into a busy restaurant and shifting attention from one conversation to another
will quickly remind us of our considerable powers of segregating sounds.
56
In musicology,
the burden of proof has been on the other sidemotet texts are inaudible until proven au-
dible, simply because they are joined by other motet texts. Given this discrepancy, it will be
worth summarizing the results of some studies that have investigated how humans are able
to solve this cocktail party problem that continues to bafe machines and musicologists.
A listeners ability to understand speech in a complex environment depends on the
segregation of simultaneously occurring sounds into streams originating from different
sourcesa process known as Auditory Scene Analysis.
57
The ability to do this success-
fully rests on a number of connected factors, some of which are the result of the sound
sources while others pertain to the environment in which listening is taking place. Among
the former, differences in the fundamental frequencies and spectral proles of soundsor
55
Cherry, On Human Communicaton, 278.
56
Though the experiment has the side-effect of making one feel rather nosy.
57
Bregman, Auditory Scene Analysis, 36.
95
we might say differences in timbre are vital to the segregation of concurrent sounds:
58
Sounds of similar timbres will group together, so that the successive sounds of the oboe
will segregate from the harp, even when they are playing in the same register.
59
And when
sounds are separated by range as well as timbre, lower and higher sounds will naturally
segregate themselves into groups, or perceptual streams. Claude Alain has shown that an
observers ability to identify two different vowels presented simultaneously improved by
increasing the fundamental frequency separation between the vowelsspecically that si-
multaneous sounds at least 4 semitones apart in pitch can be more easily parsed than those
closer that 4 semitones.
60
These timbral and registral aspects of sound-sources are so-called monaural cues
that is, they can be perceived and separated with one ear. But even more crucial to solving
the cocktail party problem is our faculty of localizing sound, which depends on binaural
hearing. The distance between the ears allows humans to accurately identify the location of
a sound-source along a horizontal axis, since the sound waves enter the two ears at different
times. Using a more subtle set of cues in a process known as head-related transfer func-
tion, we are also able to localize sounds on the other axes.
61
It has long been recognized
that spatial hearing plays a major role in the auditory systems ability to separate sound
sources in a multiple-source acoustic environment and thus in our ability to solve the
58
Ibid., 92103. Though timbre is an inexact term (Bregman calls it an ill-dened wastebasket cat-
egory, that can include fundamental frequency, spectral component frequency, the balance of the spectrum,
and possibly individual peaks in the spectrum, ibid., 92), it is also an intuitive one.
59
Ibid., 19-20, 856.
60
Alain et al., Hearing Two Things at Once.
61
See summary in Haykin & Chen, The Cocktail Party Problem, 18789.
96
cocktail party problem.
62
In parallel to the observations that increased pitch differences
can aid in sound segregation, a number of studies have found that an increase in distance
between the sources of simultaneous sounds aids in their auditory differentiation.
63
Some circumstantial factors also seem to play a role. Room acoustics can have an
adverse effect on the cocktail party phenomenon if excessive reverberation interferes with
auditory scene analysis, though the auditory system is powerful enough to suppress even
moderate amounts of echo and focus instead on the source sound.
64
On the other hand,
visual cues such as facial expression and especially the movement of the lips in speech can
aid with comprehension. Even under ideal listening conditions, lip-reading can help us by
resolving certain ambiguities of speechfor instance by giving visual cues to aid in the
rather small phonetic difference between b and a v. In noisy situations such cues become
much more important, and an ability to see the speaker is thus a signicant asset in under-
standing what s/he is saying.
65
Together these monaural and binaural processes enable us to carry out auditory
scene analysis, and this subconscious process is the rst step to solving the Cocktail Party
Problem. Once sound sources are localized, attention comes into play. Attention is, in
William Jamess classic denition, the taking possession of the mind, in clear and vivid
form, of one out of what seem several simultaneously possible objects or trains of thought.
Focalization, concentration of consciousness are of its essence. It implies withdrawal from
62
Ibid.
63
Most recently in Drennan et al., Perceptual segregation of competing speech sounds.
64
Haykin & Chen, The Cocktail Party Problem, 1885.
65
Summereld, Lipreading and Audio-Visual Speech Perception. See also Wei et al., Lip-Reading Aids
Word Recognition.
97
some things in order to deal effectively with others.
66
In the context of the cocktail party problem, several kinds of attention have been
studied. One is selective attention, in which the listener chooses to focus on some source
and ignore others. In this case the comprehension level of the primary stream is high, but
the secondary stream is hardly attended to. In an early study in which a different text was
played into each ear (dichotic listening) and listeners were asked to speak the right-ear text
back while they listened (a process referred to as shadowing), Colin Cherry found that the
rejected text got very little attentionindeed, listeners did not notice when that left-ear
text switched to German.
67
But more recent studies have shown that while shadowing or
listening actively to one text a listener may still hear aspects of the other, and may be able
to recall some words from the rejected stream.
68
The listener may also be subconsciously
affected by text which they consciously choose not to listen to.
69
Another paradigm is divided attention, in which the listener focuses on more than
one sound source, or switches between channels (switched attention). These different types
of focus can be produced by telling the listener what to pay attention to, though certain
sonic events in a rejected stream, such as hearing ones own name, can produce shifts of
attention even during selective attention. Furthermore, factors that interfere with auditory
scene analysis can make some kinds of attention more difcult to sustain. For example, as
Wood and Cowan have noted, selective attention depends heavily on physical (e.g. voice)
66
James, The Principles of Psychology (1890), 4034.
67
Cherry, Some Experiments on the Recognition of Speech, 978.
68
Wood & Cowan, The Cocktail Party Phenomenon Revisited.
69
This is demonstrated in a number of studies using priming, for example DuPoux et al., Lexical Access
Without Attention?
98
differences between channels and thus dichotic studies which use similar monotone voices
in both ears may not allow listeners to fully concentrate on one channel or to fully ignore
the other.
70
If there are any similarities we can count on between ourselves and medieval listen-
ers, they are in the makeup of our auditory systems and the cognitive mechanisms which
support the listening process.
71
For the cocktail party problem precedes cocktailsit is in
fact a basic necessity for human society. But these similar biological systems and processes
receive radically different musical stimuli today than they did 600 years ago. Listening to
medieval music in the twentieth and twenty-rst centuries often means listening to a re-
cording. The list of factors allowing the human listener to solve the cocktail problem gains
in relevance to my broader argument when we consider how much relevant information is
lost when we listen to a recording.
Most importantly, the three-dimensional localization of sound sources usually car-
ried out by head-related transfer function is rendered ineffective by the attening of the
auditory scene to only two channels, and even the information potentially contained in
these is often erased due to editing and mixing. Auditory scene analysis is further impeded
by often excessive reverbeither inherent in the recording space (often a large church) or
added to evoke a medieval sound. Such surroundings and post-production choices muddy
the waters, making it even more difcult to pick out the texts of individual voices. The re-
cent Hilliard Emsemble recording of Machauts motets is a good example.
72
A large amount
70
Wood & Cowan, The Cocktail Party Phenomenon Revisited, 255.
71
The claim of a common biological makeup is, I think, less subject to accusations of universalism than in-
vocation of a trans-historical humanness or trans-historical aesthetics; see Page, Discarding Images, 190,
and Wegmans response in Reviewing Images, 270.
72
Hilliard Emsemble, Motets: Guillaume de Machaut.
99
of reverb cloaks each motet with a blanket of overtones and echoes. As one reviewer wrote,
the engineers beautifully capture the glow of the church acoustic.
73
Indeed they do, but
these are secular worksmostly love songswhich as far as we know would not have been
sung in a church, but in a chamber or small hall.
74
The problem of reverb can be compounded with those of timbre discussed above to
create recordings in which is it impossible to tell voices apart by any cues other than range.
And range can be a ckle indicator. Consider the Hilliards recording of Machauts Dame/
Fins cuers doulz (M11). Listening to their rendition of mm. 1128 (Track 9, 0:20:54) the
impression is of a tenor and two texted voices almost identical in timbre and separated by
anywhere from a 3rd to a 6th. The upper voice seems to sing a number of neighbor-note
gestures in semibreves (A-G-A, E-D-E, G-F-G, etc). But a look at the score reveals that a
series of voice-crossings occur in this passage, so that the voices are in fact trading off this
motif (See Example 2.1). Eventually it is possible to teach oneself the difference between
these two very similar voices, and to begin to hear the crossings of their lines. But this is
a task requiring focused work and a score. How can we begin to make decisions about our
attention or to listen for a texts meaning when the rst stepauditory scene analysisis
made so difcult? And of course, the visual aspects of performance, which would be a great
aid in a situation where competing sound-sources are vying for attention, are taken away
in any recording. In this case trying to understand the text of a given voice is like trying to
hear your neighbor at a party held in a vaulted hall where the guests wear opaque masks over
their faces, while their voices are re-mixed and projected through two speakers at the front
73
Dan Davis, editorial review for Amazon.com, accessed 5 May 2010.
74
The recording includes eighteen motets of which four can be classied as ceremonial, and two more as,
perhaps, paraliturgical. These latter works might have been sung in a larger space, but this is a matter of
conjecture.
100
of the room with extra reverb added by the capricious hosts.
Of course, not all recordings are the same. The 2002 album of Machauts motets
by the French group Ensemble Musica Nova uses two discernibly different voices to render
the same work.
75
The triplum is sung by a bright tenor with forward resonance, while the
motetus is breather and has a slight buzzing quality; these distinct timbres make the voice-
crossings in mm. 1128 much more discernible (disc 2, track 2, 0:17:39). A relatively
dry acoustic adds to the intelligibility of these voices, which are pleasantly resonant on their
own. The same can be said of several Machaut motets recorded by the American ensemble
Liber unUsualis.
76
More reverberant than Ensemble Musica Nova, the performance here
is nevertheless closely miked, so that it is easier to separate the two upper voices. Quant/
Amour (M1) is sung by a triplum with a slight, fast vibrato and focused tone, and a motetus
with a slightly breathier, more luminous sound and a slower, less prominent vibrato. These
well-matched but distinct timbres allows us to keep track of each individual voiceand its
texteven through a hocket section in which the vocal ranges are suddenly equal (Track
5, :2736).
75
Ensemble Musica Nova, Guillaume de Machaut, Intgrale des motets.
76
Liber unUsualis, Unrequited.
101
Example 2.1: Machauts, Dame/Fins cuers doulz (M11), mm. 730
77
But evenor rather, especiallyin these latter cases, how much more is gained by
listening to a live performance, when the ability to distinguish sound-sources from each
77
Reproduced from PMFC 2:1445.
102
other is rendered all but automatic by their location in three-dimensional space, and when
the faculty of sight can aid our comprehension of texture and text, showing us who is sing-
ing what, when. Of course we do not always have access to a live performance (though many
scholars have the means to make a live performance happen when they need one). But it is
vital to recognize the differences between recorded and live performances, especially as they
pertain to questions of intelligibility and text-music relations more generally. Too often,
performance is taken as synonymous with recordingsindeed, Page demonstrates the
easy slippage between the two in speaking of the inuences for Discarding Images:
The primary inspiration for these chapters has been provided by performance.
The chance to hear medieval music in recorded performances is one of the
most obvious ways in which the musicological opportunities available to the
modern scholar exceed those of previous generations. Recordings are some-
times superseded by advances in knowledge...
78
Similarly, when Daniel Leech-Wilkinson makes an argument for analysis of performance
and studying music as sound in his second analysis of Rose, lis, he in fact focuses on three
recordings, two of them by Gothic Voices.
79
Signicantly, the kinds of observations he makes
focus onand, I would suggest, stem from the difculty of isolating voices:
When we listen to the rst phrase of Rose, lis we generally concentrate our
listening on the tune and the harmony that goes with it. In a three-part
performance that is very easy to do; with all four parts it is a little trickier,
because the tune is not always on top. The four-voice version is useful
for this reason, that it divides our attention, taking a bit of it away from
the tuneand tempting us to hear tune and top voice, at least, as separate
but more nearly equal. The extent to which that happens dependson our
willingness to try it rather than just to listen to the highest pitch all the
time, and on the performance. In the four-part performance issued on disc by
Gothic Voices in 1983it is not easy to pick out the cantus...
80
Indeed it is not, and it is no surprise that the ensuing analysis focuses on the interac-
78
Page, Discarding Images, p. xx; emphasis mine.
79
Leech-Wilkinson, Rose, lis revisited, 253.
80
Ibid., 254; emphasis mine.
103
tion of musical sonorities and vowelsvowels are audible in this recording. Now, Leech-
Wilkinson is careful not to make any claims about the historicity of his approachquite
the opposite, in factand the problem here is not one of authenticity. But by conating
a recording with performance, with the sound of a piece, we risk distorting priorities,
whether they be our own or Machauts. For instance, it is difcult to take at face value the
assertion that one is much more aware of the sound of the text when one sings it oneself
than when one listens to it being sung while listening to the rendition of Dame/Fins cuers
doulz by Ensemble Musica Nova.
81
None of this is to argue that modern performance is a sensuous and provocative
inducement to study medieval music and one that must be left behind as that study becomes
progressively more engaged and serious.
82
But this music in its recorded form is perhaps
that sensuous and provocative inducement which, though there is no reason to leave it be-
hind, must at least be viewed critically. I submit that in a live performance where each singer
is allowed to sing with his or her own voice, and where each understands the text enough to
be able to project some of that understanding in their facial expressions and declamation,
an audience member familiar with the language in question should have no problem pick-
ing one voice out and focusing on it. If that same audience member had the opportunity to
hear this piece again, then s/he would perhaps focus on a different voice. Certainly there are
times when one voice or another begs for attention by a bold gesture of some kindhere
a listener has the option of heeding the call of that gesture, or of staying their attention in
one place. In such a performance, a motet can be explored like a painting; and the listener
81
Ibid., 258. Conversely, in some recordings it is not clear that the performers themselves are any more
aware of the text than their listeners.
82
Page, A Reply to Margaret Bent, 132.
104
has control over their gaze. To suggest that it would be impossible to pick out various ele-
ments of that painting seems rashespecially if we perceive it most frequently through
lower-resolution, low-contrast reproductions.
How can we explore such experiences practically? A number of avenues seem open.
I have experimented with translating motets into English for English-speaking audiences,
in order to remove the linguistic barriers set up by middle French and Latin. The result
was a greater intelligibility of the text, though needless to say the translation was inelegant
compared to Machauts original. And the mode or performancea recording in which
the same singer sang both voices to a MIDI tenoris open to every objection I have out-
lined above.
83
One can imagine multi-channel recordings that take advantage of surround-
sound technology to place the listener at the center of a performance in which auditory
scene analysis is not futile, and shifts of attention can be voluntary. One can even imagine
a visual element added to this. Perhaps a trio of laptops, each with the video of one singer
(and subtitles?) could execute these complex pieces. Or we can go analog: we can perform
these works more, in small venues, with a variety of vocal types, perhaps even repeating the
same work several times in a row, so as to give the audience a chance to explore. We can re-
institute those gatherings of discerning people.
* * *
The goal of this chapter has not been to argue that motet texts, whether heard indi-
vidually or in tandem, would have been necessarily or immediately intelligible to medieval
listeners. Nor do I hope to have convinced anyone that hearing or understanding these
texts in a live performance would be automatic, even if one knew the language. But focused,
83
I played this recording to accompany my Lies, Damned Lies, and Hockets.
105
mindful attention on the part of the listener will always pay off in the case of ars nova mo-
tets, and in the case of most other types of music. The limited medieval evidence seems
indeed to place a premium on claritythat much we can glean from Jacobuss comments.
Other contemporary echoes of performance seem to suggests that motets might well have
been performed by different-sounding voices. And whatever kind of evidence this is, there
is none to the contrarythat there was any ideal of poetic, cultural, or sonic homogeneity
associated with the ars nova motet. That is not to say that medieval performances would
have been ideal in terms of intelligibility. We cannot know that. But we can recognize that
the idea that texts in a polytextual setting are unintelligible de facto is a modern prejudice,
born in part of 1980s performance aesthetics and supported in large part by the digital
mediation to which much of our listening is subject.
106
CHAPTER THREE
What is a TALEA? Upper-voice Periodicity in ARS NOVA Motets
M
ANY ASPECTS OF THE PRACTICE KNOWN AS ISORHYTHM have recently come under
review. Margaret Bent has questioned the appropriateness of the term, especially
of iso- (the same) in describing processes that are more often transformational than
simply repetitive.
1
The techniques role as generic determinant has also been subject to revi-
sion, as focus has shifted from what Ernest Sanders once characterized as the isorhythmic
Ars nova motet, (a reduction of the bewildering variety of motets ourishing at the turn of
the [fourteenth] century to one denitive type) to, in Bents words, the variety rather than
the sameness of strategies of motet composition in the 14th and 15th centuries [which
can] even less be accommodated within a single model than can sonata forms.
2
And Anna
Maria Busse Berger has questioned the purpose of isorhythmic practice, arguing that struc-
tural regularity often seen as an end in itself is in fact the result of mnemonic practices
which helped composers to write motets and performers to memorize them.
3
Within this revitalized discourse, one basic aspect of the narrative remains unques-
tioned, even though exceptions to the apparent rule are known: the tenor continues to be
viewed as the primary shaping force in the construction of isorhythmic works. The idea
that the tenors layout dictates those of the other voices and that a motets primary structur-
ing device is that of the tenor affects the ways in which 14th-century motets are edited and
1
Bent, What is Isorhythm? and Isorhythm. See also Sanders, Isorhythm, 351, and Kgle, Isorhyth-
mie, 121920.
2
Sanders, The Medieval Motet, 556, emphasis mine; Bent, What is Isorhythm?, 135. See also com-
ments in Everist, The Horse, the Clerk, 1389, and Bents account of the emergence of the isorhythmic
motet as generic label in What is Isorhythm?, 1218.
3
Busse Berger, Medieval Music, 21252.
107
interpreted. In fact, a number of motets have upper-voice structures different from those
of the tenor. These upper-voice units, which have sometimes been referred to as Grota-
lea or super-talea, have been observed in the analyses of individual pieces, and in at least
one caseMargaret Bents pioneering analysis of Vitrys Tribum/Quoniam the presence
of such upper-voice structures has been used to question the tenors supremacy.
4
The ef-
fects of supertaleae are also incorporated by Bent and others into a number of analyses of
individual pieces, especially, as we shall see, of Machauts motets Helas/Corde mesto (M12)
and Amours/Faus Semblant (M15). More general remarks about different or multiple ta-
leae reside in Heinrich Besselers footnotes, Friedrich Ludwigs editorial notes, and Ursula
Gnthers asides, which anticipate many of the observations below.
5
And Karl Kgle has
discussed some of the formal irregularities to which I will draw attention from a comple-
mentary analytical stance.
6
4
Bent argues that noticing the tenors structure alone in Tribum/Quoniam will give only subsidiary atten-
tion to the amazing interlocked tripartite structure, with its own internal identities, that is counterpointed
against the two identical tenor color statements, Polyphony of Texts and Music, 92. Bents argument has
been very inuential in the writing of this chapter. At the same time, I wish to question the notion that
the upper-voice structures in Tribum/Quoniam and other motets discussed below are counter-isorhythmic
(ibid.); as I will argue below, they are isorhythmic, but with units or repetition which are dictated by the upper
voices rather than the tenor. I hope that the other motets discussed here will serve to place the upper-voice
periods of Tribum/Quoniam in context.
5
In the case of In virtute/Decens, Besseler points out that the upper voices in the motets rst section are
organized into taleae twice as long as those of the tenor (Studien zur Musik des Mittelalters II, 222n9)
and other upper-voice congurations that do not match those of the tenor are included in the footnotes
to the table on his pp. 2224; these will be referenced individually below. Interestingly, Besseler notes the
presence of hockets in his main table as though they were a property of tenor rather than of upper-voice
construction. Gnther also mentions in passing many of the motets discussed in this chapter: As to some
extent with Vitry, in some [of Machauts] motets (nos. 1, 8, 10, 12 and 15) it takes two or three Tenor pe-
riods to make up one period of the upper voices. In the other 15 works the tenor divisions occur at the same
place as the upper voice onesit is normal in [Ivrea] too for the upper voice periods to coincide with those
of the lower voices. The appearance of two short Tenor taleae against one upper voice period occurs only in
motets 9, 13, 22 and 71. The rst part of no. 15 is exceptional insofar as here the reverse process occurs,
so that two upper voice periods are the equivalent of one Tenor talea, The 14th-Century Motet, 30, 37.
6
Although he does not invoke the possibility that upper-voice taleae might in fact be of different lengths,
Kgle has drawn attention to the results of supertaleic organization. For example, he notes that in the rst
108
But no discussion exists of supertaleae as a phenomenon. This is not surprising,
since it sits between general and specic procedures. Where analysis is focused around a
single motet (usually in order to demonstrate its depth and uniqueness) the presence of
similar procedures in other works might be inconvenient or irrelevant. And in broader
surveys, these motets stand as exceptions, and are thus not worth dwelling on in great detail.
But a collection of exceptions can change the rules. I have brought together the
analyses that follow because I believe that they illustrate a single point: in a number of
14th-century motets upper-voice structure is to some extent independent of tenor talea
structure. The idea that some motets may have forms into which the tenor is made to t,
rather than forms built by that tenor, opens the space to richer and more meaningful analy-
sis. But before turning to these irregular works it will be useful to consider the conse-
quences of allowing the tenor unquestioned structural (and semantic) primacy.
THE TENOR AS A FOUNDATION?
It is by no means unreasonable to call the tenor a fundamental element in many ars
nova motets, and at rst glance the theoretical evidence seems strongly to support such a
view. Color and talea, the only terms used in the middle ages with reference to the procedure
later called isorhythm, are usually dened as the repetition, in the tenor, of melody and
rhythm respectively. Johannes de Grocheio, writing around 1300, famously compared the
tenor of a motet to a buildings foundation and to the bodys skeletal frame:
half of Colla/Bona, phrase joints occur at the beginningof every second talea statement; The Manuscript
Ivrea, 98.
109
The tenor is that part upon which all the others are founded, in the same way
as the parts of a house or building are, upon its foundation. And the tenor
regulates them and gives them quantity, just as the bones do to the other
parts [of the body].
7
This metaphorical description is supported by Egidius de Murino, whose De modo
componendi tenores motettorum is the only detailed account of compositional process for
ars nova motet tenors. He indicates that a composer should rst rhythmicize a tenor, then
a contratenor (if there is to be one) and nally the upper voices.
8
Thus there is medieval
precedent for seeing the tenor as central to motet composition; Bents suggestion that most
of these works would better be called tenor motets than isorhythmic motets rings true.
9
But the theoretical evidence is also contingent. For example, Murino warns that
his instructions are written for the teaching of children and may therefore be too sim-
plistic for the subtilis cantor.
10
And Ugolino of Orvieto provides evidence that, by the early
fteenth century, the terms color and talea could refer to the upper voices as wellan ex-
panded usage that Daniel Leech-Wilkinson reasonably suggests must date back to the four-
teenth century.
11
Most importantly, Grocheio, an oft-cited witness to the tenors primacy,
was writing before the developments of the Ars nova. A comparison of his statement with
Murinos is instructive. Writing in the middle of the fourteenth century, Murino advises
his reader to take the Tenor from some antiphon or responsory or another chant from the
7
Tenor autem est illa pars, supra quam omnes aliae fundantur, quemadmodum partes domus vel aedicii
super suum fundamentum. Et eas regulat et eis dat quantitatem, quemadmodum ossa partibus aliis, ed.
Rholoff, Die Quellenhandschriften, 1467.
8
Ed. Coussemaker, Scriptores 3:1248, partial edition and translation in Leech-Wilkinson, Composi-
tional Techniques,1823.
9
What is Isorhythm?, 138. Clark has also referred to the tenor as the primary voice of the motetquite
naturally, given the focus of her study; Concordare cum materia, 2.
10
Ed. and trans. Leech-Wilkinson, Compositional Techniques, 19, 22.
11
Ibid., 15.
110
antiphonal, and the words should concord with the materia of which you wish to make the
motet.
12
Scholars sometimes read this as a statement equivalent to Grocheios foundation-
al metaphor, interpreting it as evidence that composers worked from the bottom up; that is,
they took a segment of chant and used it as a scaffold for the added upper voices.
13
But in
fact, it is clear from Murinos description that somethingmateriaprecedes the choice of
tenor. What is this materia? The word is unhelpfully elusive: it can mean substance, topic,
subject matter, even building material. Leech-Wilkinson reasonably interpreted it as the
message of the upper-voice texts, but several other commentators have linked materia with
the very tenor whose choice depends on it. Alice Clarks discussion of Murinos statement
exemplies the problems of interpretation presented by the text:
The tenor is thus the melodic, harmonic and rhythmic point of origin for
the motet, though it is in fact preceded by the motets materia, on whose ba-
sis it is chosen. In his sketchy account, it is easy to overlook this statement
concordare cum materia, and perhaps even Egidius is more interested in
talea and color formation than in tenor selection, but he makes it clear that
such selection is not a random act. Egidius does not, however, specify just
how the tenor is to relate to the matter of its motet. That relationship is
generally thought to center around the textual incipit given to the tenor in
the manuscript: that is, the tag is appropriate for a tenor on that subject.
14
On the one hand, materia clearly precedes the tenor; on the other, materia is so vague and
unhelpful a term that it is much easier to think of the tenor, or perhaps its text in the manu-
script, as the motets point of origin.
12
Primo accipe tenorem alicuius antiphone vel responsorii vel alterius cantus de antiphonario et debent
verba concordare cum materia de qua vis facere motetum, Ibid., 18, 21.
13
Robertson, Remembering the Annunciation, 287.
14
Clark, Concordare cum materia, 6. Compare also Anne Walters Robertsons interpretations of this pas-
sage, which read it variously as indicating bottom-up composition and a more exible approach: Compos-
ers worked from the bottom up; that is, they took a segment of chant and used it as a scaffold for the added
upper voices. This practice was summed up by music theorist Egidius de Murino, Remembering the An-
nunciation, 287; for Egidius a motet grows from top down and from bottom up simultaneously, Guillaume
de Machaut and Reims, 146.
111
If some modern accounts of isorhythmic composition unquestioningly adopt Gro-
cheios foundation metaphor rather than dwelling on the nature of Murinos materia, it
is also perhaps because of an implicit symmetry between the progress of music history and
a bottom-up order of composition. Plainchant is the oldest material, and thus composers
must have begun with it. The repetition of rhythm independently of melody occurred rst
in the tenors of thirteenth century motets and only later in their upper voices; thus it fol-
lows that isorhythm must have rst been applied to the tenor voice of a motet, and only
later to the upper voices. In music history and in a single composition (which here acts as
a microcosm) isorhythmso the logic perhaps goesbegins in the tenor and moves up-
wards, crystallizing in the upper voices with geological propriety, spreading to them (like
a disease), and once there causing acts of violence.
15
The most bruised specimens (and
therefore, the logic sometimes goes, the latest) are called Panisorhythmic motetsworks
in which every rhythmic value in every voice is subject to periodic articulation.
16
From this progressionapparently compositional as well as historicalstems the
reason most often given for the presence of upper-voice isorhythm: to make the tenors rep-
etitions more audible. In the textbook explanation, upper voices may also be organized iso-
rhythmically, in whole or part, to emphasize the recurring rhythmic patterns in the tenor.
17
15
It was felt necessary to introducestrophically recurring isorhythmic passages into the upper voices.
Such isorhythmic parallelisms at rst crystallized mainly around phrase endings, Sanders, Isorhythm,
352. Isorhythms started in the tenor line but spread complexly to other parts, Delahoyde, Medieval Mu-
sic: Ars nova <http://www.wsu.edu/~delahoyd/medieval/ars.nova.html>. The new principle [on the con-
trary] even extends the application of isorhythm to the upper voices (since the tenor is constructed in this
manner, and the complexity of its rhythm, which repeats several times, requires in compensation a rhythmi-
cally unied layout of the upper voices as well). This necessarily leads to acts of violence; Ludwig, Studien
ber die Geschichte, 224, cited and trans. Bent, What is Isorhythm?, 122.
16
The term pan-isorhythm was coined by Apel; see Remarks About the Isorhythmic Motet, 139.
Gnther used the extent of upper-voice isorhythm to date motetsa practice that has been criticized by
Bent; see Gnther, The 14th-Century Motet, and Bent, What is Isorhythm?, 1278.
17
Grout, Burkholder & Palisca, A History of Western Music, 121. The sentiment behind this textbook
112
The highlighting of structure, we are to understand, is an end in itself.
18
The result is something of a contradiction. On the one hand, upper-voice rhyth-
mic repetition has interested scholars, who have used it for dating and for articulating,
exalting, and ultimately condemning a compositional ethos whose smallest details were
foreordained, and to which any sort of lyric sentiment was as foreign as to the numbers
that determined the form and dimension of the work.
19
On the other hand, the purpose
of all this tightly controlled compositional material is simply to highlight the repetition of
rhythmic patterns in the tenor repetition which in and of itself was by no means new in
the fourteenth century. Despite this contradiction the unspoken assumption that ars nova
motets have no structure independent of that of their tenor persists, and it has important
implications.
For not only upper-voice rhythmic repetition, but also the structure of upper-voice
texts are said to amplify the tenors rhythmic cycles. In his edition of the Ivrea and Chan-
tilly motets, Frank Harrison included a table entitled Relation Between Sections of Poems
and Taleae in which he assigned a letter grade, from a+ to d, to each voice of each motet
based on the degree of coordination between poetic and musical design.
20
Although he
denition is well represented in scholarship of the past century. To take just two examples, Ludwig wrote of
upper voices whose purpose is die Isorhythmie der einzelnen Tenorabschnitte ebenfalls ganz scharf aus-
prgen, Geschichte der Mensuralnotation, 622. Later Sanders claimed that rhythmic correspondences
between successive [upper-voice] phrases or phrase groups [were put in] evidently to lend emphasis to the
works structure, Isorhythm, 351.
18
There is however 14th-century evidence that tenor taleae are not meant to be heard, as when Johannes
Boen warns composers not take too much time in the rhythmic organization of tenors because [talea] is
presented more to the sight than to the hearing. Cited and trans. Busse Berger, Medieval Music, 223.
19
de Van, ed., Guglielmi Dufay opera omnia, 2:i. Sanders comments on this discrepancy: Isorhythm is often
dened with reference only to the tenor taleae (recurring rhythmic units) of 14th-century motets [although]
it is the growth of isorhythm in the upper voices of 14th-century motets that is characteristic and signi-
cant, Isorhythm, 352.
20
Harrison, ed., Motets of French Provenance, xii.
113
notes that these value-symbolsadmittedly cannot be completely objective, his very be-
lief in the exercise speaks to the analytical primacy given to tenors.
The burden of semantic congruence also lies with the upper voices. Even though
Murino indicates that the tenors words are chosen to concord with some pre-existing thing,
the tenor is often given interpretive priority. In Anne Robertsons words:
What is evident [from Murinos treatise] is that a certain amount of knowl-
edge was embedded a priori into a composition on a cantus rmus simply
by virtue of the choice of tenor. The tenor thus assumes a crucial role in the
interpretation of a motet: it holds the key to certain aspects of meaning in
the work; it may indicate the way in which the piece was used in the liturgy;
and it may help explain the relationship of the voices to one another.
21
Jacques Boogaart reads the passage similarly, conating the choice of materia with the choice
of tenor: According to Egidius de Murinothe composer must rst dene the materia with
which the work deals by choosing a suitable fragment from plainchant; in other words the
idea at the base of the whole construction must be xed rst of all.
22
Tenor-centric readings of Murino tend to produce religious interpretations of ars
nova motets, since they take the tenors liturgical origin as a starting point. Almost inevi-
tably, the upper-voice texts are relegated to the status of accessory and called upon only to
support interpretations which have their basis in the tenor.
23
Rarely is the reverse standard
applied: just as an inverse of Harrisons table does not exist, so it is not asked whether a par-
21
Remembering the Annunciation, 287.
22
Loves Unstable Balance, Part I, 5.
23
This is the approach taken by Robertson in her study of Machauts motets, where the tenors do funda-
mentally shape and ultimately guide the interpretation of Motets 117pieces which are only seemingly
secular, Guillaume de Machaut, 82. As Everist has pointed out, while Robertsons account of the ordering
of Machauts tenors, where the gist of the story of Motets 117 is rst revealed (p. 84), has a strong evi-
dential base in the mystic theological texts that she adduces, this base is much less clear and more subject to
ad hoc interpretative strategies in her accounts of the story as told in the upper voices of the 17 motets,
The Horse, the Clerk, 139.
114
ticular tenors text is well-suited to the message of the upper voices. This is partly because
tenor texts are only a few words long, and thus easily bent in many directions, but more
because the tenors perceived structural supremacy seems to preclude the questionsince
the tenor is a foundation, a starting-point, and a melodic, harmonic and rhythmic point of
origin for the motet, its propriety to the piece is beyond reproach.
24
And yet, an entire building must be planned before the foundation can be laid.
Nor is it always possible to tell from looking at a foundation how many oors the eventual
structure may have, or what materials it might be constructed from, or for what purpose.
25
The elusive materia that precedes the tenor may include any number of signicant musi-
cal and semantic decisions which may directly impact the choice of tenor and aspects of
text-music relations. Nowhere is the tenors foundational role more open to question than
in the works I wish to discuss here: motets in which the upper-voice taleae are of different
lengths than those of the tenor. Some of these motets exhibit structural peculiarities that
are noted in existing analyses and editions. Others will be new to the discussion. But taken
as a group they have the potential to complicate and enrich our understanding of ars nova
compositional processes by casting that basic compositional building block the taleain
a new light.
24
Clark, Concordare cum materia, 6.
25
As Abigail Klima has kindly explained to me in a private correspondence, foundations are more condi-
tioned by the climate, soil, and earthquake potential of a region than by the materials of the structure to be
placed upon them. Understanding the soil condition, an architect would be able to make a ballpark guess
about the height of a building based on the foundation.
115
THE EVIDENCE FOR INDEPENDENT UPPER-VOICE TALEAE
The rst half of the Ivrea motet In virtute/Decens exemplies the phenomenon of
independent upper-voice structures. Its tenor and contratenor are organized in 10-breve
taleae, within which both lower voices are fully isorhythmic. But the upper voices have no
rhythmic congruence across these sections, except for a longa in the motetus in breves 56
of each talea (see Example 3.1, where isorhythm is shaded). Accordingly, Harrison catego-
rized the upper voices in this part of the motet as not isorhythmic.
26
In fact there is more rhythmic repetition in the upper voices than rst meets the
eye. While the odd-numbered taleae are still largely not isorhythmic, the upper voices above
tenor taleae A2, A4, and A6 have a notable amount of congruence. Or, to put it another
way, the periods being articulated in the upper voices are twice as long as the tenor taleae
(see Example 3.2). The resulting 20-breve units (marked a13 in Example 3.2) look
more like normal upper-voice taleae, in which isorhythm is more often found in the second
half than in the center of the talea.
27
These upper-voice units of periodic repetition are not
marked in this motets only edition, which labels only the tenors repetitions. But they are
important to our understanding of the work in which we can now, at the very least, observe
a greater degree of isorhythm.
26
See the evaluations of Motet #18 in Tables II and IV in Harrison, ed. Motets of French Provenance, 202,
204.
27
This is obvious in cases where there are hockets, but it applies more broadly, as Boogaart has observed:
Normally the degree of isorhythm increases towards the end of the tenor talea, Loves Unstable Balance,
Part 1, 1318.
116
Example 3.1: In virtute/Decens mm. 160, arranged according to tenor taleae;
isorhythm shaded
1
1
7
Example 3.2: In Virtute/Decens mm. 160, arranged according to upper-voice periods; rhythmic congruence shaded
118
Terminology and Diagrams
In analyses of individual motets it has been unnecessary to name the phenomenon
with which we concern ourselves hereit has been sufcient to comment, as Gnther does
in discussing Ida/Portio, that 1 color equals 4 taleae, 2 of which correspond to one musical
section of the upper voices.
28
Similarly Bent, in describing Tribum/Quoniam, which has
twelve repetitions of the tenor talea, species that there are thrice two blocks of music.
29
Discussing the phenomenon in several works, Georg Reichert used the term Grotalea to
describe these upper-voice constructions; Jacques Boogaart has called them Supertaleae.
30
Although Reicherts term is more commonly used, supertalea avoids the problems of plu-
ralization presented by the former (Grotaleen? Grotaleae?). Unless we read super- as
above, both terms restrict themselves etymologically to upper-voice taleae that are longer
than those of the tenor. They are thus less useful in describing a work like Flos/Celsa, in
which, as I shall argue, the upper voices have shorter periodsKleintaleen?than those
of the tenor. In still other cases, I will suggest that even where the musical periods articu-
lated by repeated rhythms are the same length in the tenor and the upper voices, the units
of repetition are not aligned. Perhaps there is no need for an umbrella term to describe
this variety of approaches to upper-voice form, but terminology, even if slightly imprecise,
can help us to think symbolically about related processes. I will use the terms supertalea,
upper-voice talea, and upper-voice period interchangeably.
Most editions of Ars nova motets label the tenor talea and color using Roman num-
28
The Motets of the Manuscripts, p. lx.
29
Hence, sixbecause every upper-voice talea includes two tenor taleae. Polyphony of Texts and Music,
912.
30
Reichert, Das Verhaltnis, 202; Boogaart, O series summe rata, 107.
119
bers for the former and letters for the latter.
31
Where color breaks also fall at talea breaks,
this results in alpha-numberic labelling, so that AII would be the second talea of the rst
color, BIII the third talea of the second color, and so forth. Sometimes editors switch from
Roman to Arabic numerals to describe taleae in diminution, but this practice is not con-
sistent. Since adding upper-voice periodic construction as a third layer has the potential
to confuse matters, it begs for some modication of this scheme. In the diagrams below,
I through-label tenor taleae without regard of which color they belong to, though the tra-
ditional distinction between Roman and Arabic numbering is preserved, so that II is the
second talea in the motets rst section, and 2 is the second diminution talea. Integer valor is
indicated by straight lines (
), diminution by wavy lines(
), and breaks between taleae
by vertical strokes( | ). Greek letters refer to successive upper-voice taleae, which are labeled
according to their isorhythmic region rather than the color to which they belong (a only in
a unipartite work, a and b in a motet with two sections, etc.) Measure numbers are included
to the right of each diagram to allow comparison with available editions. In the enclosed
editions I have also used Greek letters to label the beginnings of upper-voice taleae where
these differ from the tenors taleae by length or position.
Supertaleae as Grouped Tenor Taleae
Reicherts 1956 study of structural relationships between text and music in Mach-
auts motets is the rst to explore supertaleic construction . There he observed that upper-
voice periods in Hlas/Corde mesto (M12) and Amours/Faus Semblant (M15) encompass
31
This is the practice followed by Schrade in PMFC1, Harrison in PMFC5, and Gnther in The Motets
of the Manuscripts. Ludwigs Musikalische Werke uses Roman numerals for both color and talea, the former
larger in size.
120
several taleae.
32
In discussing Motet 12 he coined a term to describe the phenomenon:
Every such Grotalea covers three tenor periods and coincides exactly with the course of
the tenor [color] which sounds a total of three times.
33
Thus while the tenor talea repeats
nine times, the upper voices repeat their longer period three times (see Figure 3.1). Reichert
locates the raison dtre of this super-structure in the motetus text, which is split into three
strophes, and each of these strophes is now made the basis of a Groperiod in constant
overall disposition.
34
upper voices: |supertalea a1 |
mm. 154
tenor: |talea I |talea II |talea III |
|supertalea a2 |
mm. 55108
|talea IV |talea V |talea VI |
|supertalea a3 |
mm. 109162
|talea VII |talea VIII |talea IX |
Figure 3.1: Upper- and Lower-voice taleae in Hlas/Corde mesto (M12)
In Amours/Faus Semblant (M15) he also saw a connection between upper-voice
text structure and isorhythmic super-structure. This time, the triplums text is in control.
Though it is divided into four strophes which reect the tenors four taleae, there is com-
plete metrical afnity only between alternate strophes. And the musical setting exhibits
complete rhythmic replication in the upper voices only between the material corresponding
with tenor taleae I and III, and II and IV (see Figure 3.2):
32
Reichert also observed ein hnlicher Ansatz zur Grotalea in the motetus of Motet 21, but I have been
unable to replicate his analysis, Das Verhaltnis, 203.
33
Jede solche Grotalea umfat drei Tenorperioden und deckt sich genau mit einem Ablauf der Tenormel-
odie, die insgesamt dreimal erklingt (= 3 Colores), ibid., 202.
34
Und jede dieser Strophen ist nun in gleichbleibender Gesamtdisposition zur Grundlage der Groperi-
ode gemacht, ibid.
121
upper voices: |supertalea a1 |
mm. 160
tenor: |talea I |talea II |
|supertalea a2 |
mm. 61120
|talea III |talea IV |
Figure 3.2: Upper- and Lower-Voice Taleae in Amours/Faus Semblant (M15)
Here the supertaleae are particularly stark: tenor taleae II and IV but not I and
III contain hockets, and since hockets are always isorhythmic, the arrangement into two
halves is inevitable (though unmarked in Ludwigs and Schrades editions). Ernest Sand-
ers, Ramon Pelinski, Agostino Ziino and Karl Kgle have all mentioned the bi-partite ar-
rangement of the upper voices in Amours/Faus Semblant, and Margaret Bent and Jacques
Boogaart have carried out more extensive analyses linking the works structure with ideas
expressed in its texts.
35
As Bent has noted, the upper-voice taleae of Motet 15 are also highly repetitive
within the space dened by tenor taleae. In Figure 3.3 the solid gray shading indicates
rhythms shared by all four taleae, while the patterned shading is for rhythms shared only by
alternate taleaethe rhythms which dene the supertaleae.
36
Hlas/Corde mesto (M12) thus
serves as a better example of the supertaleae I would like to examine, in that the upper-voice
35
Sanders, The Medieval Motet, 558n257: The analytical layout of Nos. 12 and 15 in Ludwigs edi-
tion does not take their phrase structure into account. This was observed by G. Reichertwho recognized
the form of these motets on the basis of the poetic structure, but failed to identify the subdivisions of the
musical form, with which the poetry is congruous; Pelinski, Zusammenklang und Aufbau, 69: Auch der
Tenor-Cantus rmus der Motette Nr. 15 wird in vier Taleae gegliedert, die paarweise (Taleae 12 und 34)
den Oberstimmenperioden entsprechen; Ziino, Isoritmia musicale, 450: il Triplum ed il Motetus invece,
a mio avviso, sono costituiti da due sole talee e non da quattro, come risulta sempre dalledizione di Ludwig
(e da quella di Leo Schrade); see also Kgle, Die Musik des 14. Jahrhunderts, 370. Besseler makes no
mention of the upper-voice structure in the footnotes to his chart, but his characterization of the triplum
and motetus as streng isorhythmisch implies that he must have observed the supertaleae, Studien zur
Musik des Mittelalters II, 222n1, 223. More extensive analyses of the motets form are available in Bent,
Deception, Exegesis, and Sounding Number, 2022 and Boogaart, O series summe rata, 1446.
36
Figure 3.3 follows Bents analysis of isorhythm in this work, as summarized by shading in her Example
1, Deception, Exegisis, 169.
122
taleae are considerably more reective of their own scheme than they are of the tenors.
Figure 3.3: Isorhythm in Amours/Faus semblant (M15); Correspondences on the level
of tenor Talea shaded in solid gray; those on the level of Supertalea in checks and stripes.
In both Machaut motets discussed so far, and in In vitrute/Decens, the upper-voice
taleae encompass several tenor taleae (three in Hlas/Corde mesto; two in the others). This is
the most common relationship between upper- and lower-voice taleae in cases where they
do not coincide, and it is present in several more motets. Machauts Qui/Ha! Fortune (M8)
consists of four supertaleae, each of which encompasses three repetitions of the tenors talea
(see Figure 3.4).
37
In Vitrys Tribum/Quoniam, the twelve tenor taleae are superstructured by
three supertaleae in the upper voices (see Figure 3.5).
38
37
Noted in Besseler, Studien zur Musik des Mittelalters II, 224n19. Hoppin also observed that By
treating three of the rhythmic patterns as one talea, [the upper voices] make the form of the tenor be-
come 3[colores]=4[taleae], Medieval Music, 41213. See also Boogaart, O series summe rata, 1:134, and
Haaburda, Fortuna in weltlichen mehrstimmigen Kompositionen, 110, 218. Schrades edition indicates super-
taleae for this motet, labeling them as though they were taleae (The Works of Guillaume de Machaut, 1346);
in this he follows Ludwig, who however indicates in his editorial notes that although the upper voices are
made up of four periods of 9 longae each, im Tenor T. 13, 46 und 79 rhythmisch gleich sind, Guil-
laume de Machaut: Musikalische Werke, 3:32. Sanders reads the passage in a similar way, pointing out that
this work and also motets 1 and 10 are based on tenor taleae with internally repetitive rhythmic patterns,
The Medieval Motet, 558.
38
Figure 3.5 excludes the six-breve introitus that precedes the tenors entrance. Compound upper-voice
123
upper voices:
|
supertalea a1
|
mm. 127
tenor: talea I |talea II |talea III
|
supertalea a2
|
mm. 2854
talea IV |talea V |talea VI
|
supertalea a3
|
mm. 5581
talea VII |talea VIII |talea IX
|
supertalea a4
|
mm. 82108
talea X |talea XI |talea XII
Figure 3.4: Upper- and lower-voice taleae in Machauts Qui/Ha! Fortune (M8)
upper voices:
|
supertalea a1
|
mm. 730
tenor: talea I |talea II |talea III |talea IV
|
supertalea a2
|
mm. 3154
talea V |talea VI |talea VII |talea VIII
|
supertalea a3
|
mm. 5578
talea IX |talea X |talea XI |talea XII
Figure 3.5: Upper- and lower-voice taleae in Tribum/Quoniam
upper voices:
|
supertalea a1
|
mm. 124
tenor: talea I |talea II
|
supertalea a2
|
mm. 2548
talea III |talea IV
|
supertalea a3
|
mm. 4972
talea V |talea VI
|
supertalea b1
|
mm. 7384
talea 1 |talea 2
|
supertalea b2
|
mm. 8596
talea 3 |talea 4
|
supertalea b3
|
mm. 97108
talea 5 |talea 6
Figure 3.6: Upper- and lower-voice taleae in Hareu/Helas (M10)
124
Upper-voice supertaleae composed of multiple tenor taleae are also used in several
motets with diminution sections, where two different talea lengths are already in play. In
some cases, there are supertaleae in both parts of the motet, so that the proportions between
the integer valor and diminution sections are preserved. An example is Machauts Hareu/
Helas (M10). Here, the tenor talea is twelve breves long (aa. )becoming twelve semibreves
in diminution (-aaa-). In the upper-voices, the talea length is doubled in both sections of
the motet, resulting in three supertaleae of 24 breves followed by three more of twelve (see
Figure 3.6).
39
Ida/Portio by Edigius de Pusiex follows the same plan in a more complex fashion.
This motet, preserved in the Ivrea codex but also in later sources such as Chantilly, is
quadripartite, with successive diminution in the proportions 6:4:3:2. Each section con-
sists of four tenor taleae, which are interpreted in the upper voices as two supertaleae (See
Figure 3.7).
40
Here again, the talea length is doubled, but the proportions of diminution
are preserved. The upper-voice isorhythm (present in the triplum within each of the four
taleae noted in Besseler, Studien zur Musik des Mittelalters II, 223n12. Bents analysis does not privilege
this arrangement; her example 4.1 lines up the isorhythmic section of the motet in four rows of three taleae,
and Figure 4.1 is arranged according to color, thus 2 rows of 6 taleae each. But the structure diagramed here
plays a large role in the analysis: thrice two block of music arranged over twice three identical places in the
tenor [create] a great hemiola of threefold form arranged over a twice-stated color, Polyphony of texts and
Music, 92. The kind of supertaleic structure at work in Tribum/Quoniam is not unique to that workIf
Machauts Motet 8 were arranged as 3 blocks of four taleae (that is, according to its color), instead of the 4
blocks of three (the shape dictated by its upper-voice isorhythmic structure), we might there too see interest-
ing patterns of non-coordination created by rhythmically identical blocks of music.
39
Observed by Besseler in Studien zur Musik des Mittelalters II, 224n18. Schrades edition (The Works
of Guillaume de Machaut, 10811) ignores the tenor taleae and arranges the voices according to the upper-
voice supertaleae, in which he follows Ludwig, who mentions the tenor in his notes (im tenor T. 14 und
58 rhythmisch gleich sind) but marks only supertaleae in his edition, Guillaume de Machaut: Musikalische
Werke, 3:3740. Only Boogaart indicates both levels of organization in his edition, calling the tenor taleae
Ia, Ib; IIa, IIb, etc.; O series summe rata, 1:2534, 2:6048.
40
Noted by Besseler, who further points out that the motetus ceases to be isorhythmic before the triplum
does, Studien zur Musik des Mittelalters II, 222n8.
125
upper voices:
|
supertalea a1
|
mm. 136
tenor: talea I |talea II
|
supertalea a2
|
mm. 3772
talea III |talea IV
|
supertalea b1
|
mm. 7396
talea 1 |talea 2
|
supertalea b1
|
mm. 97120
talea 3 |talea 4
|
supertalea g1
|
mm. 12138
talea i |talea ii
|
supertalea g2
|
mm. 13956
talea iii |talea iv
|
supertalea d1
|
mm. 15768
talea1 | talea 2
|
supertalea d2
|
mm. 16980
talea 3 | talea 4
Figure 3.7: Upper- and lower-voice taleae in Ida/Portio
upper voices: |supertalea a1 | mm. 130
tenor: |talea I | talea II |
|supertalea a2 | mm. 3160
|talea III | talea IV |
|supertalea a3 | mm. 6190
|talea V | talea VI |
|supertalea b1 mm. 91108
|talea 1 | talea 2
|
|supertalea b2 mm. 109126
|talea 3 | talea 4
|
|supertalea b3 mm. 127144
|talea 5 | talea 6
|
|partial supertalea b4 mm. 145159
|talea 7 | talea 8
|
Figure 3.8: Upper- and lower-voice taleae in Vos/Gratissima
126
sections and in the motetus in sections a and b) is so prevalent that Harrison organizes
this work according to supertaleae in his edition, making no mention of the tenor taleae.
41
In both parts of Vitrys Vos/Gratissima too, tenor taleae are doubled to make upper-
voice supertaleae. As Besseler has noted, while the tenor has six taleae in the rst section and
7 1/2 taleae in the second, the upper voices are grouped in three larger sections followed
by four shorter ones, the last of which is truncated (see Figure 3.8).
42
At what point in
the compositional process might composers have decided to use supertaleae? For several of
Machauts motets, the answer is provided by the poetic structures of the upper-voice texts,
which mirror the supertaleae more closely than they do the tenor taleae. This suggests that
the tenor rhythms were here concieved as half-periods, rather than as building blocks in
themselves.
43
The same is true for Vos/Gratissima, where the rst half of the triplum text
is arranged in six-line strophes with the structure a
7
a
7
b
8
a
7
a
7
b
8
. In section a, each of these
strophes is set to one upper-voice supertalea, while tenor taleae correspond with the smaller
3-line half-units (see Table 3.1). Thus the triplums rhyme scheme beautifully mirrors the
doubled-up taleae; certainly the three-line unit has structural validity, just as the tenors ta-
lea is a building block in the works form. But only the six-line stanza is poetically complete.
41
Motets of French Provenance, 249, 193.
42
Studien zur Musik des Mittelalters II, 223n13. Kgle has also described the results of the larger units
in the upper voices during the second half of the piece in terms of differences between adjacent taleae: iso-
rhythmic passages (without hocket) and isorhythmic hocket passages between triplum and motetus alternate
at the beginning of each talea in the second segment, similar to the layout of the diminution section in Flos/
Celsa, The Manuscript Ivrea, 102. Vos/Gratissima is a slightly unusual case, in that the tenor is re-rhythmi-
cized in the second section of the motet, rather than being diminished. This approach has sometimes been
called pseudo-diminution, though there is no diminution here to speak of. While other works with freely
re-rhythmicized tenors often use smaller note-values in the talea rhythm of the second section into order to
mimic the effect of diminution, Vos/Gratissima uses perfect and imperfect longs, breves, and breve rests in
its tenor in both sections. But the length of the taleae decreases from 15 breves in section a (..a.) to 9
in section b (.aa). The resulting 5:3 proportion (one that would be impossible to achieve with diminution)
is preserved in the doubled supertalea lengths of 30 and 18 breves.
43
Reichert, Das Verhaltnis, 2023.
127
Table 3.1: Allocation of the Triplum text in Vos/Gratissima, section a
triplum text rhyme supertalea tenor talea
Vos, quid admiramini.
virgines, si virgini
pre ceteris eligende
dignati fuerimus
nubere? dum nupsimus
tamquam valde diligende:
a
7
a
7
b
8
c
7
c
7
b
8
a1 A1
A2
Ista pulcra spetie.
humilis manerie
ac opere virtuosa;
turpis vestrum altera.
ausu numis aspera.
necnon virtutes exosa.
d
7
d
7
e
8
f
7
f
7
e
8
a2 A3
A4
Ista lux, vos nubila
ista velox aquila.
vos, colubres gradientes!
Ista super ethera
regnat, vos in misera
valle languetis egentes.
g
7
g
7
h
8
j
7
j
7
h
8
a3 A5
A6
More Intricate Upper-Voice Arrangements
In Hareu/Helas, Ida/Portio, and Vos/Gratissima, all sections of the motet take the
same approach to building supertaleae, so that the upper-voice periods ultimately preserve
the proportions and effect of the tenors diminution. But at least two other works, Vitrys
Colla/Bona and Flos/Celsa (also ascribed to Vitry), use supertaleic structure to render all of
the upper-voice periods the same length, in spite of the tenors diminution. The tenor of
Colla/Bona is arranged in the rather complicated scheme of seven-and-a-half twelve-breve
taleae in integer valor, followed by seven six-breve taleae in the diminution section (Figure
3.9a). But the short tenor taleae clearly form larger units, doubling up in periods which are
128
12 breves long the same length as the integer valor taleae (see Figure 3.9b).
44
Thus the -
nal six taleae in diminution combine to form three supertaleae . Supertalea 8 becomes a com-
pound period, combining a fragmentary integer valor talea with rst of the short diminished
taleae. Although the text-setting is far from regular, Vitry treats the compound talea 8/9
as a unit: in the motetus, he sets one line nunquam saporiferaacross the diminution
break. In the triplum, two lines are set to this compound talea (aulici sunt opere/semper
adulari), just as two lines are set to the ve integer valor taleae before it (see Example 3.3).
The result is much simpler than it sounds: with its combination of integer valor tale-
ae, diminution supertaleae, and the hybrid talea 8/9, the upper voices of Colla/Bona emerge
as eleven (super)taleae of twelve breves each, plus a nal longa which stands outside of the
isorhythmic scheme.
Where Colla/Bona uses double-length supertaleae in its diminution section to cre-
ate periodic equality between the motets two sections, Flos/Celsa, another motet probably
by Vitry, takes the opposite approach to achieve the same effect: it uses half-taleae in its
integer valor section. The tenor talea is 24 breves long in the rst sectionby far Vitry's
longest. This is repeated three-and-a-half times, and then sung again in diminution (12
breves long) three-and-a-half more times (see Figure 3.10a). But throughout the motet the
upper-voice rhythms recur at a rate of 12 breves.
45
In other words, the tenor talea is divided
in half in the rst section of the motet (see Figure 3.10b).
The tenors rhythm seems to have been designed with such a division in mind. The
full pattern . . .. ..splits easily into two rhythmically similar halves of 12
44
Noted in Besseler, Studien zur Musik des Mittelalters II, 222n3.
45
Again, noted by Besseler, ibid., 222n7.
129
(a) Tenor:
|
talea 1
|
|
talea 2
|
|
talea 3
|
|
talea 4
|
|
talea 5
|
|
talea 6
|
|
talea 7
|
|
half-talea 8
|
|
talea 9
|
|
talea 10
|
|
talea 11
|
|
talea 12
|
|
talea 13
|
|
talea 14
|
|
talea 15
|
(b) Upper Voices:
|
talea 1
|
|
talea 2
|
|
talea 3
|
|
talea 4
|
|
talea 5
|
|
talea 6
|
|
talea 7
|
|
half-talea 8
|
talea 9
| = supertalea 8
|
talea 10
|
talea 11
| = supertalea 9
|
talea 12
|
talea 13
| = supertalea 10
|
talea 14
|
talea 15
| = supertalea 11
Figure 3.9: Tenor (a) and Upper-voice (b) periods in Colla/Bona
Example 3.3: Colla/Bona, mm. 8596 (reproduced from PMFC1, 86)
130
|talea I | mm. 124
|talea II | mm. 2548
|talea III | mm. 4972
|half-talea IV | mm.73-84
|talea 1 | mm. 8596
|talea 2 | mm. 97108
|talea 3 | mm. 10920
|half-talea 4| mm. 121-6
Figure 3.10a: Tenor taleae in Flos/Celsa
|a1
}
= tenor talea I
mm. 112
a2 | mm. 1324
|a3
}
= tenor talea II
mm. 2536
a4 | mm. 3748
|a5
}
= tenor talea III
mm. 4960
a6 | mm. 6172
|a7 | = tenor half-talea IV mm. 7384
|b1 | = tenor talea 1 mm. 8596
|b2 | = tenor talea 2 mm. 97108
|b3 | = tenor talea 3 mm. 109120
|b4 | = tenor half-talea 4 mm. 1206
Figure 3.10b: Upper-voice taleae in Flos/Celsa
131
breves each. The maxima of the rst half is replaced in the second by two longae, and the
rst halfs nal longa becomes rests: . . | .. ...
46
And over these similar tenor
halves the upper voices sing similar phrases which in no way hint at the tenors larger struc-
ture.
47
So far we have considered two ways in which multipartite motets use supertaleae. In
some, such as Ida/Portio, all sections combine upper- and lower-voice periods in the same
way, for instance by doubling the tenor with respect to upper-voice supertaleae. In others,
supertaleae appear in only one section, making the length of upper-voice periods throughout
the work equal even though the length of the tenor color changes (e.g. Colla/Bona). Machauts
Quant/Amour (M1) apparently combines these two approaches. Like the former group of
motets it uses supertaleae in both its integer valor and its diminution sections, with the result
that the tenors twelve taleae (six in each section) become six (see gure 3.11).
48
46
Kgle has commented on the similarity of these two hemistichs of twelve breves each, differentiated
from each other only by the substitution for a long of a longa rest on breves 910 in the second phrase
(= ), The Manuscript Ivrea, 100. The sonic effect is perhaps more similar still, since in performance
a maxima does not sound any different, from a textural point of view, to two longae, though the former
guarantees more tonal stability.
47
The analytical shape of Flos/Celsa is not as neat as that of Colla/Bona, since it does not square up
completely: a nal half-talea in diminution seems external to the scheme of repetition, and its presence is
unexplained by upper-voice rhythmic concordances. I will return to this point below.
48
This was rst noted in Besseler, Studien zur Musik des Mittelalters II, 224n18. Gombosi also noted
that the tenor of each talea is a double period, seeing in this an off-center kind of symmetry, Machauts
Messe Notre-Dame, 220.
132
upper voices:
|
a1
|
mm. 136
tenor: talea I |talea II
|
a2
|
mm. 3772
talea III |talea IV
|
a3
|
mm. 73108
talea V |talea VI
|
b1
|
mm. 10920
talea 1 |talea 2
|
b2
|
mm. 12132
talea 3 |talea 4
|
b3
|
mm. 13344
talea 5 |talea 6
Figure 3.11: Supertaleae in Machauts Quant/Amour (M1)
Numerically the three nal supertaleae (b13) add up to one upper-voice supertalea of the
integer valor section, since the latter are 36 breves long, and the former, 12. This is not a
coincidence, but rather a result of the 3:1 ratio produced by diminution. Yet the setting of
the triplum text suggests that Machaut was thinking of the entire diminution section as
a unit equivalent to an upper-voice supertalea. The triplum is divided into four nine-line
stanzas, of which the rst three are set to sections a1, a2, and a3, and the nal one to the
diminution section:
a1 Quant en moy vint premierement
Amours, si tres doucettement
Me vost mon cuer enamourer
Que dun regart me st present,
Et tres amoureus sentement
Me donna aveuc dous penser,
Espoir davoir mercy sans refuser.
Mais onques en tout mon vivant
Hardement ne me vost donner;
133
a2 Et si me fait en desirant
Penser si amoureusement
Que, par force de desirer,
Ma joie convient en tourment
Muer, se je nay hardement.
Las! et je nen puis recouvrer,
Quamours secours ne me veut nul prester
qui en ses las si durement
me tient que nen puis eschaper;
a3 ne je ne weil, quen atendant
sa grace je weil humblement
toutes ces doleurs endurer.
Et sAmours loyal se consent
que ma douce dame au corps gent
me weille son ami clamer.
je scay, de vray que aray, sans ner.
joie quAmours a n amant
doit pour ses maus guerredonner.
b1 Mais elle atent trop longuement
et jaimme si follettement
que je noze merci rouver.
b2 car jaim mieus vivre en esperant
davoir merci procheinnement
que Refus me veingne tuer.
b3 Et pour ce di en souspirant:
Grant folie est de tant amer
que de son doulz face on amer.
The triplum texts arrangement invites us to view the diminution section as, in a
sense, section a4. Quant/Amour is thus poetically reminiscent of Colla/Bona, which squares
up its diminution section to equate it with the integer valor. Meanwhile, the motetus text
acknowledges the motets structure as it stands, with two lines of text assigned to each of
the supertaleae, whether in integer color or in diminution. The result is a much more densely
texted motetus in section b of the motet, since the same number of lines is squeezed into a
section a third the length.
This is as good a place as any to add that though I have opposed upper-voice
134
structures to tenor ones, it is by no means impossible for a motetus and a triplum to suggest
slightly different isorhythmic structures. In Colla/Bona, for instance, the actual patterns of
upper-voice rhythmic recurrence are a little more complicated than Figure 3.9b suggests,
since the motetus, but not the triplum, hints at upper-voice supertaleae during the rst four
tenor taleae. Meanwhile the text-setting for the motetus supports the scheme graphed in
Figure 3.9b, while the triplum sets approximately the same amount of text to each tenor ta-
lea, without regard for supertaleae. The kinds of structural differences I have observed above
between the tenor and the upper voices may well apply between voices, on a smaller scale.
* * *
It is one thing to argue that upper-voice supertaleae structure the relationship be-
tween text and music in a motet, and another to wonder why they should do so. What is the
point, for Machaut, of these poetic and isorhythmic congruences in Quant/Amour? And
why does Vitry use half-taleae in Flos/Celsa? Is he playing a numerical game? Maybe so.
But I think that there is a more interesting explanation for the presence of supertaleae in
the motets I have discussed so faran explanation which has its roots in mnemonic pro-
cedures.
ISORHYTHM AND MEMORY
It is generally agreed that isorhythm was a compositional toola way for compos-
ers to organize their works and keep track of rhythm, which as the most recently updated
dimension of notation was probably still something of a conceptual challenge in the mid-
fourteenth century.
49
In the received view, composition took place on some kind of writ-
49
See, for instance, Gnthers observation that generally complete isorhythm appears in the Machaut mo-
tets in all bars distinguished by more interesting rhythms, The Fourteenth-Century Motet, 31.
135
ing surface. For example, Daniel Leech-Wilkinson suggested that a composer might have
sketched a model talea to help him make his way through a motet.
50
More recently, Anna
Maria Busse Berger has brought the emerging eld of medieval memory studies to bear
on motet composition. As the culmination of her study of the ars memorativa in the teach-
ing, transmission, and composition of chant, music theory, and medieval polyphony, Busse
Berger focuses on the isorhythmic motet, arguing that isorhythmic structures assist the
memory both for the composer and the performer.
51
She argues that motets were sung by
heart and that composers used mnemonic schemesespecially of the architectural vari-
etyto compose. Composers and singers are not separable for Busse Berger, who suggests
that isorhythmic structures were useful both in learning and composing: The memoriza-
tion of old texts and the creation of new ones are closely intertwined: they both use the same
tools. The importance of this point cannot be overstated.
52
Arguing that singers memorized text and music separately, Busse Berger goes on to
show how texts and music are separately memorable.
53
Learning texts is made easy by their
regular meters and simple rhyme-schemes.
54
And musical lines can be memorized due to
the presence of taleae in the tenors and of regular phrase lengths, recurring rhythmic gures,
and isomelism in the upper voices. Bringing it all together is the manuscript page, on which
50
Compositional Techniques, 601, 113, 14041, 154.
51
Medieval Music, 228.
52
Ibid., 214.
53
Citing Fallowss ndings about disorderly text-underlay in fteenth-century chansons, Busse Berger
suggests that the evidence leaves little doubt that singers did not learn music and text at the same time,
Medieval Music, 213. However, fteenth-century song-texting practices differ signicantly from those for
14th-century motets, where text-setting is often syllabic and remarkably consistent between sources (see
Chapter 1). It would seem likely that in this repertory the music is more memorable for the presence of text,
and vice-versa.
54
Ibid., 2259.
136
both composer and performer can see the entire motet:
I suggest that composers of isorhythmic motets chose to organize their piec-
es in tightly organized structures because it allowed them to work out the
pieces in their mind and make them memorable to performers. Ars nova
notation allowed them to see the entire musical structure on an opening, an
important condition for mnemonic structures.
55
Busse Bergers main argumentthat composers visualized compositions in their minds
during the compositional processis compelling. But manuscript openings do not high-
light musical structure. They do not, as a rule, separate tenor taleae with bars or ag upper-
voice isorhythm, either by means of markings or layout. Nor is it easy when looking at an
opening of, for example, the Ivrea Codex, to see isorhythm (except perhaps in the tenor).
There can be no doubt that the ability to zoom out and view the entire memorized object at
once is a crucial aspect of mnemotechnics. But just because performer and composer see the
piece in one glance, the former on parchment and the other in his minds eye, can we assume
that they were seeing the same thing?
I contend that pieces with supertaleae offer some compelling hints about what the
composer might have seen when he zoomed out. In all cases described above, supertaleae
transform more complicated structures into less complicated ones. Perhaps the best ex-
ample of this is Colla/Bona. Based on the tenor only, the motets structure is an unwieldy
seven taleae followed by a fragmentary eighth, and then seven more taleae which are half the
size of the rst seven. However, the diminution section is organized in supertaleae which are
the same length as the taleae in the integer valor section. The half-length eighth fragmen-
tary talea folds in with the rst diminution talea, and the others pair off. The piece can
now be conceived as eleven phrases of equal length. It goes almost without saying that this
55
Ibid., 254.
137
arrangement of the upper voices is more memorable than that of the tenor. Where in the
tenors scheme there are two lengths of talea, the upper-voice units are all the same length
(12 breves), and the motet can be summarized in the simple formula 11x12one we would
use to summarize the dimensions of a rectangle.
Flos/Celsa is also easier to picture when the long taleae of the integer valor section
are re-organized by the upper voices into half-taleae. Instead of keeping three different
lengths of talea in mind (24, then 12, then 6 breves), the motet is re-inscribed as ten 12-
breve taleae, with a fragmentary eleventh of 6 breves (10x12+1x6). And even in Machauts
Quant/Amour(M1), the triplum text encourages us to think of the nal three supertaleae as
combined together, making a rectangle with the three integer valor supertaleae.
To call the motet a rectangle reeks of a-historicism. It suggests that the composer
would imagine the parts as aligned, perhaps in score, and that he would further engage in
some kind of mental stacking of sections, associating like with like, perhaps even beginning
in the top and ending in the bottom of some mental space. I am not implying that compos-
ers thought in modern score layoutbut I would argue that they did not think in choirbook
format either. Several factors point to the possibility that a composer might indeed visual-
ize his work in a geometric manner.
Although the use of score notation would not be common until the sixteenth cen-
tury, examples of score (or, rather, of what has been called quasi-score) were known in the
14th century.
56
In England, it was commonly present in the ubiquitous and often virtuosic
56
Owens denes quasi-score as an arrangement in which individual voices each occupy a single staff and
are superimposed one above the other, not necessarily in the order high to low, and without barlines or verti-
cal alignment, Composers at Work, 35. See also the discussion in Lowinsky, Early Scores in Manuscript.
138
discant repertory.
57
In France, although songs and motets were notated in parts, sources
of Parisian organum continued to be available, as indicated by inventories.
58
Addition-
ally, theoretical treatises copied in the fourteenth century sometimes included examples
in pseudo-score, on grids, or on large staves.
59
For instance, the anonymous De varietate et
modo discantandi (late 14th c.) contains two-part contrapuntal examples notated on staves
from six to ten lines high and separated by vertical bars (see Figure 3.12).
Figure 3.12: Example on a ten-line staff from De varietate et modo discantandi
60
Even without recourse to the concept of a score, it does not seem drastic to posit
that composers thought of voices as progressing together through time. Regardless of how
these works were composed, the three or four voices that made up any polyphonic ars nova
composition were written to belong together. The rhythms of the upper voices especially beg
to be viewed in tandem in hocket, where they are carefully constructed to be complementary
57
Facsimiles of sources notated in score and quasi-score are collected in Summers, English Fourteenth-
Century Polyphony.
58
For inventory mentions of Notre-Dame organum sources, see Haggh & Huglo, Magnus liberMaius
minus, Appendix I kindly made available by Professor Haggh on her website, < http://www.music.umd.edu/
Faculty/haggh-huglo/>.
59
Some of the earliest notations of polyphony use vertical alignment of simultaneous pitches in different
voicessee the famous examples cited in Apel, The Notation of Polyhonic Music, 20513.
60
Bibliotheek der Rijksuniversiteit, incunabulum n 70, with facsimiles reproduced in Schreurs, Anthologie
van muziekfragmenten, 269.
139
(see Example 3.4). Nor need we think in score to bring these parts together; a superimposi-
tion of all three voices on one staff is easily achieved, and more readily brings to light their
closeness in range and rhythmic complementarity (see Example 3.5).
Example 3.4: Hocket in Quant/Amour (M1), mm. 946
Example 3.5: Hocketed notes in Quant/Amour (M1), mm. 946, with voices superim-
posed; triplum in pink, motetus in green, tenor in blue.
As for the idea that composers would mentally stack similar sections of composi-
tions, this is a matter of conjecture, but perhaps an unproblematic one. Medieval diagrams
and calendars regularly stack data in columns, and illuminations often present the same
subject on several vertical levels, divided by lines.
61
In the writing down of verse, lines of the
same length are placed over each other. The practice in musical manuscripts of underlaying
successive verses of a strophic piece one under the other also hints at such vertical align-
61
Other arrangements, notably circular ones, were also common. See Carruthers, The Book of Memory,
24850. For an example of repeated depiction of the same subject see the lions in the Peterborough besti-
ary, reproduced in Otto Pcht, Book Illumination in the Middle Ages (New York: Harvey Miller), 26.
140
ment of like with like.
Further evidence that composers might have conceived of their works in geometric
terms comes from mnemonic treatises. As Mary Carruthers has argued, things which are to
be remembered were thought in some way to occupy physical space which can be surveyed
by the minds eye.
62
The physical space so occupied was often described as square or rect-
angular. Thus Thomas Bradwardine writes in his De memoria articiali (c. 1333) that
the location into which texts to be memorized can be placed should be like a four-sided
oblong (that is, a rectangle or square) made of stacked rows.
63
Similarly, Jacobus Publicius,
writing in the late 15th century, explains that the memorial matrix will follow the pattern
of a square.
64
Even the dimensions of these squares are indicated. Bardwardine rcom-
mends four times ve backgrounds, or perhaps ten of them; Jacobus gives an example
with ve by ve spaces.
65
Busse Berger has noted that, like Bradwardine and other writers on memory, music
theorists from the fourteenth century use verbs of seeing to talk about rules of counter-
point and composition. Johannes Boens discussion of color serves as especially compelling
evidence that that composers might have thought this way about tenors. Busse Berger ex-
plains:
His division of the thirty tenor pitches into 6x5 or 3x20 notes is reminis-
cent of what we readin memory treatises. If one were to make a diagram
of the isorhythmic structure described by Boen, it would be very similar
62
The Book of Memory, 27.
63
Trans. ibid., 281.
64
Cited and trans. in Carruthers and Ziolkowski, The Medieval Craft of Memory, 238.
65
Trans. Carruthers, The Book of Memory, 282.
141
to the structures depicted by BradwardineBoens discussion leaves little
doubt that the composer planned and visualized the composition before he
wrote it down.
66
Boen discusses two motets, of which one, Impudenter/Virtutibus, is unipartite and the
other, Apta caro/Flos virginum, has a second color in diminutionanother thirty that would
make one half the length of the others.
67
Neither of motets uses supertaleae in its upper
voices, thus to describe the tenor is to describe the layout of all three voices. But is it pos-
sible that in a motet like Colla/Bona one of the mental images used by the composer to plan
his composition was the plan of the upper voices?
Certainly the emphasis on four-sided oblongs as desirable spaces for mnemonic
storage could help to explain the shape of that motet when its upper-voice isorhythm is
taken into account. Though their tenors present awkward collections of taleae and partial
taleae of various lengths, the overall schemes of both Colla/Bona and Flos/Celsa are ren-
dered rectangular by the supertaleae. (Or nearly so: Flos/Celsa contains an extra half-talea
to which I shall return below.)
What of the simpler examples, such as Tribum/Quoniam and Qui/Ha! Fortune, where
all taleae are the same length to begin with, and supertaleae serve only to change the dimen-
sions of the rectangle? There too the upper-voice structures may point to the presence of
a mnemonic component in composition. As noted above, Bradwardine mentions that the
memorial matrix might be some four by ve blocks, or perhaps ten by ve, or somewhat
fewer, unless a man should want to make unheard-of marvels.
68
This accords with recent
66
Busse Berger, Medieval Music, 224.
67
Quoted and trans. ibid., 223.
68
Trans. Carruthers, The Book of Memory, 282.
142
ndings that there is a capacity limit to working memory, that is, to the number of indi-
vidual things we can keep in our head at the same time. George Miller famously dened this
limit in 1956 as The Magical Number Seven, Plus or Minus Two.
69
In this light it is notable that many of the motets discussed above are far taller
than nine taleae when we look at their tenor structure only. Qui/Ha! Fortune and Tribum/
Quoniam would each require the composer to keep twelve separate units in mind. Just try to
picture twelve distinct units without grouping themreader, youll get lost. But the number
of distinct units decreases dramatically when these works are rearranged according to their
upper-voice periods. The point is not that Vitry and Machaut were reading Bradwardine,
but rather that where the memorial faculty is involved in the construction of a structure,
some limits are necessarily imposed.
Of the works catalogued by Besseler in his 1927 article (a list that includes almost
all ars nova motets) only two that have more than ten taleae do not use supertaleae.
70
And
of these, onethe Fauvel motet Firmissime/Adestohas no upper-voice isorhythm at all,
so we have no way of knowing whether its composer was indeed combining its very short
taleae in his head. The prevalence of supertaleae in motets whose structures would otherwise
exceed the limits of working memory adds weight to Busse Begers argument that the com-
position of isorhythmic works had a signicant mnemonic component.
69
Miller, The Magical Number Seven. Though Carruthers and Ziolkowski caution that the limits of me-
dieval memory are spatial rather than temporal, they observe that the psychological phenomenon of seven
plus or minus two can certainly be observed to apply to ancient and medieval mnemotechnic, The Medieval
Craft of Memory, 12.
70
Besseler, Studien zur Musik des Mittelalters II, 2224. Both are Fauvel motetsFirmissime/Adesto
and O canenda/Rex. The former consists of eight integer valor taleae three longae in length, and eight more
one-longa taleae in which the tenor is freely re-rhythmicized (in pseudo-diminution); the latter is eight
taleae in integer valor plus four more in diminution.
143
For our purposes, memory and writing are of course inseparable. While texts could
be composed entirely in the mind and then written down, they could also be memorized
from books (Hugh of St Victor recommended always using the same copy of a text); and
some treatises recommended writing on wax tablets to train the memory.
71
As Busse Berger
has noted, isorhythmic motets... could not have come about without writing.
72
But while it is misleading to rigidly oppose literacy and orality, invocation of lit-
eracy alone tends to skew our priorities. Carruthers has argued that as a concept, literacy
privileges a physical artifact, the writing-support, over the social and rhetorical process that
a text both records and generates, namely, the composition by an author and its reception
by an audience.
73
That is where a consideration of mnemonic techniques can help identify
both the object of analysis and the most appropriate means of analytical representation.
While the manuscript opening does indeed present the motet at a glance, to privilege it is
to over-emphasize the physical artifact at the expense of the processesliterate, rhetori-
cal, mnemonic, culturalat work in its composition. Although the examples above, and
even more so the examples to follow, are arranged in score and stacked in accordance with
modern paradigmatic approaches to analysis, I suggest that they can still serve as evidence
of a kind. For it seems in keeping both with late-medieval rhetorical and compositional
practices and with the internal salient properties of these works themselves that compos-
ers might have visualized their motets as collections of similar things ordered spatially and
sequentiallyas matrices, grids, or shapes.
* * *
71
Carruthers, The Book of Memory, 9, 86.
72
Medieval Music, 254.
73
The Book of Memory, 11.
144
Once we allow ourselves a sort-of birds-eye-view of motets by assigning shapes to
them for analytical purposes at least, the question follows: were these shapes always regular?
In the examples presented above, supertaleae serve only to change the dimensions of repeat-
ing blocks, in all cases but one making larger units out of smaller ones and rendering the
object more memorable. But the next series of examples suggests that the shapes evoked by
some motets may be more complicated.
SHIFTS BETWEEN UPPER- AND LOWER-VOICE TALEAE
Doubling, tripling, quadrupling, and halvingthe ways in which we have so far seen
upper-voice periods interact with the tenors taleaeall have in common the fact that talea
breaks in the tenor and in the upper voices regularly coincide. Here the terms supertalea (or
subtalea) are, I think, unambiguous. But there are other cases in which upper-voice periods
align less regularly with tenor taleae. One extreme example of this is a case like Machauts
Trop plus/Biaute (M20), in which the tenor is in the form of a rondeau, its two statements
repeated in the order ABAAABAB, while the upper voices move freely above it. The upper-
voice hockets, however, are isorhythmic with each other, though they occur irregularly over
the course of the motet.
74
Indeed, each of the three hockets is placed over a different tenor
rhythm, so that the only isorhythm is between upper voices (see Figure 3.13 and Example
3.6).
This approach exemplies a kind of upper-voice independence that is easy to miss.
74
The rst and third hockets sound over sections of the A and B parts of the tenor which are almost
rhythmically identical (a- a----a), differing only in the nal note (a perfect breve the rst time around,
and an imperfect breve the second time). But the second hocket falls on a part of the tenor that bridges the
repetition of the A section, underlaying an entirely different tenor rhythm (-- -a. a-a-) to the isorhythmic
hocket in the upper voices. Bent has argued that this motet has three irregularly placed blocks of complete
isorhythm in all three parts, at mm. 811, 2528, and 4245 (What is Isorhythm?, 131), but the only
exact matches between all three voices are between measures 810 and 424.
145
|
Hocket 1
|
|
Hock-
-et 2
| |
Hocket 3
|
Figure 3.13: Hocket placement relative to the tenor in Machauts Trop plus/Biaute
(M20), Ferrell-Vog fol. 280.
Example 3.6: Hockets in Trop plus/Biaute (mm. 811, 258, 425, reproduced from
PMFC 3:1112), isorhythm shaded.
Hocket 1:
Hocket 2:
Hocket 3:
146
As Bent has noted,
Inmotets by Machaut and others, signicant rhythmic and melodic rep-
etitions have often been overlooked because they are not aligned with the
tenor talea. It is often the case that only regular repetition is signaled, to the
end of charting a qualitative crescendo towards tidy and mechanical rhyth-
mic identity. The result is that subtler kinds of variation are neglected.
75
But in exploring such subtler kinds of variation, it is worth asking whether regular rep-
etition is always aligned with the tenor talea. A counter-example appears in Sub Arturo/
Plebs, the nal section of which features three upper-voice taleae that are out of sync with
the tenor: the upper voice periods consist of 10.5 breves of (63 minims) but the tenor
takes up 16 breves in (64 minims). Thus the tenor begins one minim into talea g2 and 2
minims into talea g3 (See Figure 3.14).
76
|
upper voices g1 | mm. 12131
tenor talea C1
|upper voices g2 | mm. 13242
|tenor talea C2
|upper voices g3
|
mm. 14353
|tenor talea C3
Figure 3.14: Alignment of upper-voice and tenor taleae in Sub arturo/Plebs (nal section)
Sub arturo/Plebs may seem like an extreme case, but it is also part of a later genera-
tion of motetsnot transmitted in either Trmoille or Ivrea, it survives in Chantilly and
even Bologna Q15. As such I think its structure is the culmination of the independent
upper-voice schemes already implicit in much earlier works. But where in Sub Arturo/Plebs
75
Ibid., 129.
76
This is not reected in Gnthers edition (The Motets of the Manuscripts Chantilly, 52) but it is in Harri-
sons (Motets of French Provenance, 1767) and Bents (Two 14th-Century Motets, 17). See also the discus-
sion in Bent, What is Isorhythm?, 129n41. Both Harrison and Bent indicate tenor talea repetitions but
not upper-voice isorhythmic regions.
147
periods of different length get progressively further misaligned, in the earlier motets the
periods of repetition in the tenor and upper voices are of the same length, but their bound-
aries are misaligned.
Now, this necessarily gets us into a realm of some conjecture, since if something is
looped over and over, who is to say where its cycles begin and end? In fact, motet taleae are
not identicalonly rhythmically similar in parts. And these similarities may begin to ap-
pear at different points in different voices. For example, the opening and closing measures
of the work are less likely to be isorhythmic in the upper voices than corresponding mea-
sures in internal taleae. This by no means surprising: the aesthetic decisions involved in be-
ginning and ending a piece are different from those made in structuring measures internal
to a work. But does it follow that the rst or last talea is simply less isorhythmic than others,
or is it possible that openings or closings may sometimes stand outside of the upper-voice
isorhythmic structure?
77
A case in point is Flos/Celsa. As noted above, its rst three-and-a-half tenor taleae
are split in half to support seven upper-voice subtaleae equivalent in length to the periods
in diminution that follow (diagramed in Figure 3.10b, above). But this arrangement does
not adequately describe the integer valor section (see Example 3.7). To begin with (and un-
surprisingly), the opening two measures are not isorhythmic. Furthermore, the upper-voice
isorhythm is clustered mostly at the beginning of each subtaleaa somewhat unusual cir-
77
This is more obviously the case in works that begin with an introitus (so called in the Machaut sources)
a section that may include the lower voices of a motet but does not gure in the isorhythmic structure that
follows. However, Michael Allsen has convincingly shown that these sections are often related to the iso-
rhythmic form in terms of length, and thus are affected by that form even though they do not participate in
it (or perhaps they question the notion of what it means for some section of a composition to participate in
its form). Allsen, Style and Intertextuality, 24651. I am grateful to Professor Allsen for sharing unpub-
lished work with me on this subject. I plan to address introitus as term and practice in a forthcoming study.
148
Example 3.7: Flos/Celsa, mm. 184, arranged according to upper-voice subtaleae
149
cumstance since upper-voice rhythmic congruences are more likely to occur toward the
end of periods.
78
Finally, the line-ends of the upper-voice texts are consistently misaligned
with the talea breaks. Though the text-setting seems consistent (four triplum lines and two
motetus lines per upper-voice talea), the nal syllable of each group of lines falls each time
at the beginning of the next taleaa circumstance which earned Flos/Celsas triplum and
motetus a C+ and a C on Harrisons report-card.
79
However, if we take the non-congruence of the opening two measures as a clue that
they stand outside of the works isorhythmic structure, we can shift the upper-voice units of
repetition by two breves with respect to the tenor. Now the triplum and motetus line-ends in
the integer valor section correspond with upper-voice talea ends and the end of each period
is marked by several breves of upper-voice isorhythm (See Example 3.8, section a). As a
result the internal divisions and repetitions highlighted in the analysis are more closely
aligned with line-ends, and thus with the experience of both composing and performing
the work.
Since the two-breve shift does not seem to continue into the diminution section,
I have proposed in Example 3.8 that measures 85-6 function in both parts of the mo-
tetthat its two rectangles overlap slightly.
80
This is in part borne out by the rhythmic
evidence. Although these measures correspond exactly with the rhythms which open the
78
Boogaart, Loves Unstable Balance, Part I, 138.
79
PMFC5, Table IV, item 7, p. 204.
80
Kgle has described the same phenomenon in terms of phrase differential between the upper voices:
This phrase differential is reversed at the beginning of the diminution section; the motetus, formerly end-
ing after the triplum, now leads the triplum, ending its phrases one breve before it (phrase lengths: triplum
eight, twelve, twelve and ten breves; motetus seven, twelve, twelve, and eleven breves). Simultaneously with
the change in phrase alignment, isorhythmic hockets of four breves in length appear at the outset of every
tenor talea, The Manuscript Ivrea, 100.
Example 3.8: Periodic Alignment in Flos/Celsa
Section a: 7
x
6L
Section b: 3 x 6L; 2-breve overlap between sections
8 breves external to periodic structure (2 at the beginning,
6 at the end)
Line-breaks in the text are marked with horizontal lines
(|).
150
151
subsequent diminution taleae (a correspondence indicated by yellow shading), the tenors
longa at this point (the diminished version of the maxima which begins the tenors integer
valor talea) corresponds with tenor longae found at the ends of alternate subtaleae in the in-
teger valor section (e.g. mm. 134, 378, and 612). Thus the tenor and one measure of
the triplum in this section (shaded in green, as the intersection between blue and yellow)
correspond with both the a and the b material. Such a rhythmic link might well have been
planned in the tenor: a longa is the only rhythmic value that is shared between the original
and diminished versions of the tenors talea. The others (maximae and breves) are unique to
tenor parts a and b respectively.
Finally, the two measures which stand outside of the structure at the beginning of
the motet invite us to see the motets last six breves not as a partial talea (as gure 3.10b
would have it), but as closing material standing outside of the repetitive structure. After all,
neither an analysis nor a pre-compositional plan needs to take account of every note.
In a sense, my rearrangement of Flos/Celsa is a matter of pushing breves around
for purely analytical aims: I have not affected the edition in any way, nor have I changed
any aspect of text-music relations. But there is nothing pure about analysis. Not only do
our analytical perceptions of a piece affect judgements about its quality (as with Harrisons
gradesthe upper voices would surely deserve As now?), they also gure prominently in
the kind of hermeneutic endeavor with which the following chapters concern themselves. If
I am to argue (as I shall) that motet texts can inuence large-scale structural decisions, it is
important to be aware of the range of possible structures. The idea that upper-voice struc-
tures are both more varied and perhaps more complicated than those of the tenor opens up
new avenues of interpretation. But it may also preclude certain kinds of argument, since it
152
can alleviate what might otherwise be perceived as formal tensions between the tenor and
the upper voices. That is, if it is not unusual for the upper voices to have their own struc-
ture, then we might view their departures from the tenors form with less expectation of a
hermeneutic explanation. A comparison of several analyses of Machauts Sil estoit/Samour
(M6) will illustrate this point.
UPPER-VOICE STRUCTURES AND HERMENEUTICS IN SIL ESTOIT/SAMOURS (M6)
Sil estoit/SAmours is unique among Machauts works in being bipartite without ei-
ther diminution or renotation of the tenor.
81
Only its rst color is given in the manuscript
sources, with a repeat sign. But the tenor as written is one breve short of a perfection, so
that the second color begins on the third breve of a modus grouping. This results in mensu-
ral transformation upon the repetition of the notated material, since different breves are
altered the second time through:
82
Notated tenor:
aa
.
aaaaaa
.
aa
.
aaaaaa
.
aa
.
aaaaaa
.
aa
.
:||:
Altered rst time:
Altered second time:
Figure 3.15: Altered Breves in the tenor of Sil estoit/Samour (M6)
A further unusual aspect of the tenor is that it appears to end with a fragmentary
talea. The color is 29 notes long, but each talea only uses nine notes, hence 27 notes for
81
Gnther has argued that the work may be considered unipartite rather than bipartite (The 14th-
century Motet, 30); Bent suggests the helpful term homographic as a descriptor of the tenors unity. See
also objections to Gnthers claim in Apel, Remarks about the Isorhythmic Motet, 143; Sanders, The
Mediaeval Motet, 2623n279; and Bent, What is Isorhythm?, 131.
82
Cf. Bent, What is Isorhythm?, 129, who suggests that this example of mensural transformation with-
out diminution leads to a modus change, so that the second [tenor statement] neither renotated, nor
rewritten, nor specied, is in imperfect modus with unaltered breves.
153
three taleae. The two extra notes appear at the end as two breves followed by two breves of
rest, seeming to make up a partial fourth talea (and a partial eighth talea, when repeated; see
Figure 3.16).
|talea 1 |talea 2 |talea 3 |talea 4? :||:
Figure 3.16: Sil estoit/SAmours, tenor (Ferrell-Vog, fol. 266)
To be sure, fragmentary taleae exist in other motetsnotably Colla/Bona, Flos/Celsa
and Vos/Gratissima, all discussed above. But as we have seen, these may be reinterpreted by
upper-voice isorhythm (as in Colla/Bona) or stand outside of the isorhythmic structure (as
in Flos/Celsa). And in all three cases, the partial taleae are substantial, taking up at least half
the length of a full talea. By contrast, the extra bits in Sil estoit/Samours are short in com-
parison with the fteen-breve taleae, lasting only three breves in the rst section, and six in
the second. For all of these reasons the short taleae have occasioned comment. In Newman
Powells evaluation,
When one considers the neat, highly organized mathematical structures
found in the tenor parts of other motets by Machaut, this periodic struc-
tureor, rather, this analysis strikes one as being strangely lopsided, un-
convincingly chopped off, irregular, and somehow inadequate.
83
Nor does everything seem quite right with the upper voices. Specically, attention
has been drawn several times to the so-called Phasendifferenz between the upper voices
and tenor. Specically, the triplums phrases (musical as well as poetic) consistently end
83
Fibonacci and the Gold Mean, 242.
154
three breves later than the tenor taleae (see Example 3.9). This combination of unusual
factorsa color that changes without being renotated or diminished, a talaeic structure that
seems formally excessive, and texts that lag behindhave produced a number of fruitful
analyses.
Those analysts progressing, as it were, from the top down, have explained the tenors
fragmentary taleae from a text-layout point of view. Thus Reichert argued that the three
excess measures at the end of the rst color are there to make up for the lagging of the
triplum text, so that it is not compressed into a small space. For this reason the fourth talea
is broken off after three measures, to wit exactly with the completion of the strophe, at
which point the rst talea of the new color begins.
84
Lawrence Earp has also noted that the
staggered phrasing [of the triplum text] is compensated by a talea fragment of three breve
measures at the end of the integer valor section and at the end of the diminution section.
85
Such upper-voice-oriented explanations differ from the analytical ndings based
on bottom-up approaches, which seek to explain the tenors fragmentary extra bits and
the apparent asymmetry of its two halves without recourse to non-structural elements. This
was the approach of Gombosi, who argued that despite appearances to the contrary the en-
tire tenor is in fact symmetrical, its rhythms strictly concentric, with the pivot of the rst
talea of the second part serving as center of the whole composition.
86
Gombosis analysis
is, to borrow Gnthers evaluation, somewhat forced, in that the symmetry he proposes
involves the interpretation of some rhythmic cells, but not of others, as retrograde, in a
84
Reichert, Das Verhltnis, 208.
85
Guillaume de Machaut, 371.
86
Machauts Messe Notre-Dame, 221.
1
5
5
Example 3.9: Sil estoit/SAmours (M6), arranged by tenor talea (continued on following page)
The edition is based on the version of the motet preserved in Ferrell-Vog.
1
5
6
Example 3.9, continued.
157
not altogether methodical manner.
87
But Gombosis approach prompted a more detailed
treatment of symmetry in the tenor by Powell.
88
The two analyses have in common a focus
on mm. 4951 as the central pivot within the motets symmetrical scheme, but differ in
their interpretations. For Gombosi, the beginning of this talea [the rst in the diminution
section] serves as a transition, and therefore does not participate in the form, which is
completed by a corresponding half-talea at the end (his diagram is reproduced as Figure
3.17).
89
Figure 3.17: Gombosis analysis of the tenor of Sil estoit/SAmours (M6)
90
Powell places his center at the same place, but makes these measures count as part of both
halves (his analysis is reproduced as Figure 3.18):
The [.. a] that is in the absolute center will be seen to complete a grouping of
ve longae in the established periodization of the rst half and at the same
time to begin the periodization of ve longae for the second half, just as this
same perfect longa both completes the rst color and begins the second.
91
87
Gnther, The 14th-century Motet, 30n16. See also Dmlings critique of Gombosis approach in
Isorhythmie und Variation, 26n9.
88
Fibonacci and the Gold Mean, 24252. Earp has summarized this approach as Gombosis sugges-
tions carried to an extreme, Guillaume de Machaut, 371.
89
Gombosi, Machauts Messe Notre-Dame, 221.
90
a and b respectively represent the rhythms which read hqqqq and q