PAN-RATIONAL AND IRRATIONAL RHYTHM: THE HISTORY, DEVELOPMENT, AND

MODERN IMPLEMENTATION OF NON-DYADIC RATIONAL RHYTHMS IN WESTERN

MUSIC

By

JORDAN ALEXANDER KEY

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL

OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2021
© 2021 Jordan Alexander Key

2
To my partner in all things, Jason Edmond Johnson

3
 ACKNOWLEDGMENTS

For support in the completion of this document, I wish to thank my advisors Paul

Richards, Scott Lee, Jennifer Thomas, and Kevin Knudson as well as my additional mentors in

music, Laura Ellis and Paul Koonce, all of whom have graciously guided me over many years. In

addition to my present advisors, I also want to thank my past mentors in music: Jack Gallagher,

Daniel Asia, and Brian Quakenbush. I also want to thank my mother and father, Nora Neaves

Key and Wade Dillon Key whose emotional, financial, familial, and spiritual support have been

incalculably important to my success; along with my father and mother, I thank my brother and

sister-in-law, Justin Dillon Key and Lindsey Nicole Key, who have been sincere friends and

supports throughout my life. Additionally, I thank my mother-in-law, Judith Johnson for her

love. Last, but not least of all, I wish to thank my husband, Jason Johnson, who has been an

inspiration over many years, encouraging and aiding me to achieve my fullest potential.

4
 TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ……………………………………………………………………….. 4

LIST OF TABLES ………………………………………………………………………………. 8

LIST OF FIGURES ……………………………………………………………………………… 9

LIST OF MUSICAL EXAMPLES ……………………………………………………………... 11

ABSTRACT …………………………………………………………………………………….. 16

CHAPTER

1 INTRODUCTION: DEFINITIONS AND DELIMITATIONS ……………..…………….. 18

Mathematical Definitions …………………………………………………………………... 22

Divisibility and Indivisibility …………………………………………………………... 22
Rationality and Irrationality …………………………………………………………… .23
Commensurability and Incommensurability …………………………………………… 25
Dyadic and Non-Dyadic Rational ……………………………………………………. .. 26
Musical Definitions ………………………………………………………………………… 26
Time: Preliminary Axioms …………………………………………………………….. 27
Time: Scientific, Actual, and Musical …………………………………………………. 27
Musical Time: Prescriptive and Non-Prescriptive …………...………………………... 29
Rhythm: prescriptive and Non-Prescriptive …………………………………………… 30
Tempo and Speed ………………………………………………………….…………… 31
Measure and Meter …………………………………………………………………….. 32
Rhythmic Hierarchy: Mensuration and Time Signature ……………………………….. 33
Polyrhythm and Polymeter …………………………………………………………….. 44
Musical Scope and Delimitations ………………………………………………………….. 45
Document Layout ………………………………………………………………………….. 46

2 SUBTLE SIGNATURES FOR MUSIC’S TIME: RHYTHMIC INNOVATIONS

OF THE LATE MIDDLE AGES AND THE ARS SUBTILIOR …………………………. 49

The Franconian Vision: the Ars Antiqua, the Ars Nova, and the Birth of Rhythmic
Hierarchy .………………………………………………………………………………. 50
The Ars Subtilior and the First Experiments in Pan-Rational Rhythmic Notation ..…….… 59
Conclusions to Chapter 2 …………..……………………………………………………… 78
Synthesis I ....……………………………………………………………………………..… 81

3 THE DREAM OF TINCTORIS: PROPORTIONALITY IN THE RENAISSANCE

AND ADVANCED STUDIES IN RHYTHMIC COUNTERPOINT ..…………………… 86

Muris and the Proportions from which “a song will be made” …………………………….. 86

5
 Prosdocimo and the 15th Century Advent of Hindu-Arabic Fractions in Music ………...… 90
The Rise of Fractions in the 15th Century: The Road to Tinctoris …………………….…… 97
The Schools of Proportional Mensuration: Hothby, Tinctoris, and Gaffurius …………… 107
John Hothby and the Beginnings of Proportional Mensural Schools …………………..… 111
Tinctoris and Gaffurius: Revisionists of Proportional Music …………………………..… 113
Towards the Elizabethan Avant-garde: Mensuration in the 16th Century ……………...… 122
Dygon and Baldwin: The Gaffurian Legacy in Brittan and the Elizabethan Rhythmic
Avant-garde …………………………………………………………………………… 127
Conclusions to Chapter 3 ………………………………………………………………… 146
Synthesis II ………………………………………………………………………...……… 148

4 THE DECLINE AND REBIRTH OF RHYTHMIC DIVERSITY: IVES, COWELL,

NANCARROW, CARTER, AND RHYTHM IN THE EARLY 20th CENTURY ..……... 151

The Dark Age of Rhythm and The Genesis of Time Signature: 16th to 19th Century .…… 151
The 19th Century and New Horizons of Rhythmic Diversity …………………………….. 163
Charles Ives: The Rebirth of Rhythmic Diversity in the 20th Century …………………… 170
Beswick, Cowell, and Bartolozzi: The Early 20th Century and the Re-emancipation
of Rhythm ………………………………………………………………………...…... 185
Conlon Nancarrow: The Emancipation of Meter in the Mid-20th Century …..………..…. 208
Elliott Carter: Metric Modulation and the Extension of the Tuplet …………………….… 219
Conclusions to Chapter 4 ……………………………………………………………….… 234
Synthesis III …………………………………………………………………………….… 239

5 NEW HORIZONS FOR RHYTHMIC DIVERSITY: RHYTHMIC EXCEPTIONALISM

AND ITS NOTATION IN THE LATE 20th CENTURY ……..………………………….. 244

The Limits of Tuplet Notation and Hyper-rhythmics in the Late 20th Century ...………… 244
Aggregate Tuplets and Stratified Mensurations in the Late 20 th Century ……………..…. 272
A New Look at Time Signature and Tuplets in the Late 20 th and Early 21st Century ….… 283
Conclusions to Chapter 5 ……………………………………………………………….… 298
Meta-synthesis: What makes a good system of Notation? …………………………….…. 301

6 NEW RHYTHMIC RESOURCES: A PERFORMATIVE SYSTEM OF PAN-RATIONAL

RHYTHMIC PRESCRIPTION ……………………………………………………...…… 315

Adoptions, Abandonments, Adaptations, and Additions …………………………….…… 315

Notation of Pan-Rational Time Signatures and Their Rhythmic Hierarchies ………….… 318
Perfect-Regular Rational Rhythmic Hierarchies ………………………………….….. 318
Imperfect-Regular Rational Rhythmic Hierarchies ……………………………….….. 322
Applications of the Pan-Rational Rhythmic Notation System …………………….…. 337
Demonstration 6.01 …………………………………………………………….…. 339
Demonstration 6.02 …………………………………………………………..…… 340
Demonstration 6.03 ……………………………………………………….………. 342
Demonstration 6.04 ………………………………………………………….……. 344
Practical Examples of Pan-Rationalism …………………………………………...….. 346

6
 Notation of Indivisible and Irrational Mensurations Under Pan-Rationalism ………..…... 366
Irregular (Indivisible) Rational Time Signatures/Mensurations ………………......….. 366
Irrational (Incommensurate) Mensurations under Pan-Rationalism ………………..… 372
Conclusion …………………………………………………………………………..……. 376
Problems in Key’s System of Pan-Rational Rhythmic Notation …………………..…. 378
Benefits to Key’s System of Pan-Rational Rhythmic Notation …………………….… 384

APPENDIX

A COMPLETE MUSICOLOGICAL EDITIONS FROM CHAPTER 4 …………………… 388

B NOTED ERRORS AND DISCREPANCIES IN MILTON BABBITT’S PHILOMEL

AND JORDAN KEY’S RENOTATION OF BARS 1-23 OF PHILOMEL ……………... 420

Noted Errors in Babbitt’s Manuscript and Norton’s Edition of Philomel ……………...… 420
Jordan Key’s Renotation of bars 1-23 of Milton Babbitt’s Philomel …………………….. 422

C ORIGINAL HYPER-RHYTHMIC AND PAN-RATIONAL COMPOSITIONS

BY JORDAN ALEXANDER KEY ……………………………………………………… 425

Discursus Anachronismus (“Discursivities against Time”) for quintet ………………….. 426

Passacaglia Mensuras, 'pater meus bac(c)h(us) est' …………………………………. 427
Machaut, Mitter, Messiaen, 'A Secret Labyrinth' …………………………………….. 434
Fantasia & Fugue, 'Sumite Karissimi' ………………………………………………... 443
Saxophone Quartet No. 1 : Polyptych, ‘DOCH, nichts ist genug‘ ...................................... 464
Wie doch die Zeir vergeht... (Chanson) ......................................................................... 466
Das ist doch keine Kunst! (Chaconne) ........................................................................... 471
Nimm doch etwas Kuchen (Fugue Fantasia) .................................................................. 478
Doch, es ist mein Fernweh... (Chorale) ......................................................................... 486
Haette ich es doch nur geusst! (Canon) ......................................................................... 492
Versuch’s doch mal! (Fantasia) ..................................................................................... 496
Octet: Threnody on the Death of Children ……………………………………………….. 502
To Say Pi: A Black-MIDI Ballet for fixed media …………………………………………. 554
Bicinia: Three Pieces on Rhythmic Proportions for clarinet duo ………………………… 556
Bicinia, Version 1, Tuplet Realization ………………………………………………... 558
Bicinia, Version 2, Tempo Realization ……………………………………………….. 576
Bicinia, Version 3, Pan-Rational Realization (Mvt. 1 only) ………………………….. 592
Trio, Festive March for the 12th Hour for clarinet, violin, and piano …………………….. 598
Nachi no Taki (那智滝), On the Inkjet Scrolls of Tomohiro Muda for fixed media ……... 636
Viola Sonata for solo viola ……………………………………………………………….. 639
Toccata ad Sancta Caecilia for solo organ ……………………………………………….. 649
Rhythmomachia: Icosihexaplex for solo violin ………………………………………….... 657

LIST OF REFERENCES ………………………………………………………………...…… 666

BIOGRAPHICAL SKETCH ……………………………………………………………...….. 684

7
 LIST OF TABLES
Table page

3-1 Rhythmically eccentric works within the Baldwin Commonplace Book …………. 144

4-1 Examples of quintuple, septuple, and hendectuple time by Romantic

composers ………………………………………………………………………… 169

4-2 Relationship of canon ratios and tempo markings in Nancarrow

studies ……………………………………………………………………….……. 213

4-3 Comparison of Longy-Miquelle (Carter) and Hindemith (Read) tuplet

notations ……………………………………………………………………….…. 234

6-1 Tempo values by first subdivision on two smallest pqr-time signatures …….…... 323

6-2 Table of first-order subdivision values of lowest six pqr-time signatures

and lowest three pqrs-time signatures ………………………………………….… 324

8
 LIST OF FIGURES
Figure page

1.1 Classification system of Rational Time Signatures …………………………………36

2-1 Notes shapes in the superius voice of Rodericus’ “Angelorum psalat tripudium” ... 74

3-1 Thomas Morley, Table of Proportions from Plaine and easie Introduction to
Practicall Musicke ………………………………………………………...….…... 137

3-2 Proportions given in Nathaniel Giles’ duo on Miserere ………………………….. 145

4-1 Sample survey of 15th-16th century repertoire in perfect time ……………….…… 156

4-2 Triple meter signatures as given by Giovanni Maria Bononcini in 1673 ………… 158

4-3 Triple meter signatures as given by Lorenzo Penna in 1684 ……………………... 159

4-4 Rhythmic proportions found in Charles Ives’ In Re Con Moto et al, bars 59-61 …. 184

4-5 Laura H Beswick’s free tripartite subdivision of the whole note …………….…... 189

4-6 Example given by Beswick of intermingling of dyadic and triadic subdivisions … 189

4-7 Table of note-shapes and rhythmic values in Henry Cowell’s Fabric ………...…. 195

4-8 Bruno Bartolozzi’s complete table of rhythmic notation …………………………. 206

4-9 Elliott Carter, Triple Duo: tempo relationships ………………………………..…. 230

4-10 Diagram of System Quantization Classification ………………………………….. 241

5-1 P0, I0, and I8 for the tone matrix of Babbitt’s Philomel …………………………… 250

5-2 Creston’s non-dyadic time signatures with “primary” and “extrametrical”

units ……………………………………………………………………………….. 289

5-3 Creston’s “Table of Common Note Values” from Rational Metric Notation ……. 290

k
6-1 Time signature tree for signatures in the form of , where k is some integer
p
partitioning of the FTU, and p is the only prime factor of the FTU where 2 ≤ p .. 319

k
6-2 Rhythmic hierarchy for signatures in the form of , where k is some integer
pq
partition of the FTU, and p and q are the distinct prime factors of the FTU
where 2 ≤ p < q .…………………………………………..………………….…. 326

9
6-3 Time signature tree of corresponding rest values for signatures in the form of
k
, where k is some integer partitioning of the FTU, and p and q are the
pq
distinct prime factors of the FTU where 2 ≤ p < q …………………………..…. 329

6-4 Expanded pan-rational rhythmic hierarchy tree, with combinatorial values …..…. 331

6-5 Rhythmic hierarchy for pentadectadic time signature (p = 3 and q = 5) ……….… 333

6-6 Rhythmic hierarchy for dectadic time signature (p = 2 and q = 5) …………….…. 334

6-7 Dectadic Rational Rhythmic Hierarchy (“Decaplex”) ………………………….… 335

6-8 Imperfect Triplex Mensural Modulation via a third order pivot ……………….…. 339

6-9 Perfect Triplex Mensural Modulation via a third order pivot ……………….……. 340

6-10 Perfect Pentaplex Mensural Modulation via a third order pivot …………….……. 340

6-11 Partitioned Modulation by partitioned addition to the 3 rd-order (equally,

subtraction from the 2nd-order) …………………………………………………… 341

6-12 Partitioned Modulation by partitioned addition to the 3 rd-order via

non-dyadic rational time signature ………………………………………………... 342

6-13 Partitioned Modulation methodologies with non-dyadic rational rhythmic

hierarchy for k/6 time signatures …………………………………………….…… 343

6-14 Partitioned Modulation by partitioned subtraction to the 1st order (equally,

addition to the 2nd order) …………………………………………………….……. 344

6-15 Partitioned Modulation by partitioned subtraction to the 1st order via non-dyadic
rational time signature ………………………………………………………….…. 344

6-16 Partitioned Modulation by partitioned addition to the 2nd order (equally,

subtraction to the 1st order) …………………………………………………….…. 345

6-17 Partitioned modulation by partitioned addition to the 2nd order via

non-dyadic rational time signature …………………………………………….….. 345

A-C1 Traditional vs. Pan-Rational notation of triplets intermixed with quarter notes ….. 651

A-C1 Traditional vs. Pan-Rational notation of triplets intermixed with quarter notes ….. 651

A-C1 Traditional vs. Pan-Rational notation of triplets intermixed with quarter notes ….. 652

10
 LIST OF MUSICAL EXAMPLES
Example page

2-1 Perotin, Alleluia, Nativitas Gloriosae (Montpellier Manuscript, ca.

early 13th century ………………………………………………………………...… 52

2-2 Anonymous Franconian Motet (Montpellier Manuscript, ca. late 13th century) .….. 53

2-3 Franconian style “Agnus Dei” from The Mass of Tournai (early 14th century) ….... 56

2-4 Ars Nova isorhythmic motet “Se grasse” from The Mass of Tournai
(early 14th century) …………………………………………………………………. 56

2-5 Opening portion of by Johannes Cuvelier’s Se Genevre from the

Chantilly Codex ……………………………………………………………….…….67

2-6 Johannes Ciconia, Le ray au soleyl with canonic inscription ……………….………70

2-7 Johannes Ciconia’s Le ray au soleyl from the Lucca Codex ………………………. 71

2-8 Rhythmic mapping of Johannes Ciconia’s Le ray au soleyl from the

Lucca Codex ……………………………………………………………………….. 72

2-9 The opening of Rodericus’ (S Uciredor’s) “Angelorum psalat tripudium” ……..…..75

2-10 Rodericus’ (S Uciredor’s) “Angelorum psalat tripudium” from Chantilly

Codex (f. 48v) ………………………………………………………………….……76

3-1 Johannes Stockem, Ave maris stella from the Materanensis Manuscript ..……….. 388

3-2 John Hothby, Ora pro Nobis from the Faenza Codex ……………………...…….. 392

3-3 Adam, De tous biens playne a2 from the Segovia Manuscript …………...………. 395

3-4 Tinctoris, Le Souvenir a2 from the Segovia Manuscript ………………...……….. 396

3-5 Obrecht, Regina Caeli a2 from the Segovia Manuscript ………………...……….. 397

3-6 Anonymous, Le serviteur a2 from the Materanensis Manuscript ………………… 399

3-7 Mensural tricinia from Dygon’s Proportiones practicabiles secundum

Gaffurium …………………………………………………………………………. 402

3-8 Mensural tricinia from Morley, A Plaine and Easie Introduction …………...…… 403

3-9 Nathaniel Giles, Miserere from the Baldwin Commonplace Book ………………. 414

11
4-1 Charles Ives Over the Pavement (1910); bars 59-61 ………………………..……. 174

4-2 Charles Ives, Scherzo: Holding Your Own (1903); bars 22-28 ……………..……. 175

4-3 Charles Ives, Tone Roads No. 3 (c. 1915); bars 18-21 ……………………..…….. 176

4-4 Charles Ives, “The Housatonic at Stockbridge” from Three Places in New
England (c. 1911); bars 1-4 ………………………………………………...…….. 177

4-5 Charles Ives, The Unanswered Question (1908, rev. 1935); bars 47-53 …………. 178

4-6 Charles Ives, Largo Risoluto No. 1 (1909); bars 10-22 ………………………….. 179

4-7 Charles Ives, Largo Risoluto No. 1; bars 10-22 ……………………………...…… 180

4-8 Charles Ives, mvt. II “Argument” from String Quartet No 2 (1913);

bars 17-30 …………………………………………………………………...……. 180

4-9 Charles Ives, The Unanswered Question; bars 14-19 ……………………......…… 181

4-10 Charles Ives, In Re Con Moto et al (1916); bars 7-12 ……………………….…… 182

4-11 Charles Ives, In Re Con Moto et al; bars 24-25 ……………………………….….. 182

4-12 Charles Ives, In Re Con Moto et al; bars 42-45 ……………………………….….. 183

4-13 Charles Ives, In Re Con Moto et al; bars 59-61 ………………………………..…. 185

4-14 Henry Cowell, Quartet Romantic, bars 10-15 ……………………………….…… 197

4-15 Henry Cowell, Rhythmicana, bars 10-12 …………………………………….…… 197

4-16 Henry Cowell, 6 Ings, mvt. 1 “Floating,” bars 1-11 ……………………………… 198

4-17 Henry Cowell, 6 Ings, mvt. 6 “Seething,” 5-9 ……………………………………. 198

4-18 Conlon Nancarrow, excerpt from Study for Player Piano No. 1 …………………. 212

4-19 Approximated rendition of third canon from Nancarrow’s Study No. 33

for Player Piano. Transcription by Paul Usher ………………………………..…. 216

4-20 The opening of the third canon in Nancarrow’s Study No. 33 ………………...….. 218

4-21 The closing of the third canon in Nancarrow’s Study No. 33 ………………..…… 218

4-22 Elliott Carter, String Quartet (1950 - 1951), mvt. 1, bars 36 – 43 …………..…… 224

12
4-23 Elliott Carter, Double Concerto, bars 1099-114 ……………………………..…… 228

5-1 Norton Edition of Milton Babbitt’s Philomel, Part 1 with time-point analysis .…. 251

5-2 mm 15-16, Norton Edition of Milton Babbitt’s Philomel ………………………… 254

5-3 mm. 15-16, Babbitt Manuscript of Milton Babbitt’s Philomel …………………… 255

5-4 mm. 19-22, Key Edition of Milton Babbitt’s Philomel ……………………….….. 255

5-5 Brian Ferneyhough, Second String Quartet, bars 122-125 …………………….…. 262

5-6 Ferneyhough, In Nomine, bars 8-13; rhythmic reduction of oboe, bars

8-10 ………………..……………………………………………………..……….. 263

5-7 Rhythmic reduction, simplification, and metrical clarification of Ferneyhough’s

oboe part in measure 9 of In nominee a 3 ……...…………………………….…… 264

5-8 Michael Finnissy, English Country Tunes. Excerpt, page 6 ……………………… 266

5-9 Michael Finnissy, English Country Tunes. Excerpt, page 50 …………………….. 267

5-10 Roger Redgate’s Genoi Hoios Essi measures 69-71 ……………………………… 267

5-11 Jonathan Dawe, Liber de arte contapuncti 1477/2001, bars 73-76 ………………. 269

5-12 Ben Johnston, String Quartet 4, page 10 (1st-4th system) ………………………… 274

5-13 Ben Johnston, String Quartet 4, page 15 (4th system) -16 (1st system) ...………… 275

5-14 Ben Johnston, String Quartet 4, page 16 (3rd-4th system) ………………………… 275

5-15 Ben Johnston, Sextet, rhythmic skeleton of opening sections ……………….…… 276

5-16 Michael Gordon, Four Kings Fight Five, bars. 231-234 …………………………. 278

5-17 Akira Nishimura, String Quartet No. 4, mvt. 3, rehearsal 102 …………………… 279

5-18 Akira Nishimura, String Quartet No. 4, mvt. 3, rehearsal 111-112 ………………. 279

5-19 Karen Khachaturian, Trio for Strings, rehearsal 7 ………………………...……… 280

5-20 Ben Johnston, String Quartet No. 5, page 15-16 ………………………….……… 282

5-21 Ben Johnston, String Quartet 10, mvt. III, bars 1-4 ……………………………… 283

13
5-22 Carlos Chavez, Piano Sonata No. 3, page 19, final system ………………………. 284

5-23 Thomas Adès, Traced Overhead, II Aetheria ………………………………...…... 291

5-24 Thomas Adès, Piano Quintet ………………………………………………...…… 293

5-25 Reduction of Rehearsal Mark 4, Adès, Piano Quintet, 5-7 ……………………….. 293

5-26 Thomas Adès, Piano Quintet, rehearsal mark 3 ……………………………..…… 294

5-27 Thomas Adès, Piano Quintet, excerpt from the Piano …………………………… 294

5-28 Belling’s alternative piano part, 5 after [3] in Thomas Adès’ Piano Quintet …….. 294

5-29 Adès, Violin Concerto, violin solo, rehearsal 14, two notations by Adès …….….. 295

5-30 Adès, Violin Concerto, violin solo, Huw Belling’s second alternative
notation ………………………………………………………………...…………. 296

5-31 Thomas Adès, Tevot, after rehearsal mark F ……………………………..………. 297

5-32 Thomas Adès, Totentanz, after rehearsal mark 2 …………………………...…….. 297

5-33 Peter Alexander Thoegersen, Hypercube, first page …………………..………..... 299

6-1 Melodic variations in dyadic-, triadic-, and pentadic-rational time …………...….. 320

6-2 Comparison of A) Key’s rhythmic notation with perfect-regular time signatures,

B) by use of tempo changes with tuplets, and
C) by use of tuplets exclusively ...……………………..…………………………. 321

6-3 Melody from 6-02 given in traditional “simple” time signature ………………… 321

6-4 Key’s Viola Sonata No. 1, Ceol Mor, mm. 8-11 showing hexaplex ……………… 347

6-5 Key’s Viola Sonata No. 1, Ceol Mor, mm. 26-31 showing hexaplex …………….. 347

6-6 Key’s Viola Sonata No. 1, Ceol Mor, mm. 16-19 showing decaplex ………….…. 348

6-7 Key’s Viola Sonata No. 1, Ceol Mor, mm. 70-76 showing decaplex ……….……. 348

6-8 Key’s Toccata ad Sancta Caecilia for solo organ, mm. 1-9 showing
tetradecaplex ……………………………………………………………………… 350

6-9 Key’s Toccata ad Sancta Caecilia for solo organ, mm. 10-25 showing
tetradecaplex ……………………………………………………………………… 350

14
6-10 Key’s Toccata ad Sancta Caecilia for solo organ, mm. 62-72 showing
tetradecaplex ……………………………………………………………………… 351

6-11 Key’s Octet, page 10, demonstrating the use of partitioned decaplex ……………. 352

6-12 Key’s Octet, page 14, demonstrating the use of partitioned decaplex ……………. 352

6-13 Key’s Octet, page 24, demonstrating two uses of partitioned decaplex ………….. 353

6-14 Key’s Octet, page 38-40, demonstrating extensive use of partitioned

hexaplex ……………………………………………………………………….….. 355

6-15 Key’s Octet, page 30, of pan-rationalism w/o pan-rational signatures ...………… 356

6-16 Key’s Bicinia, first movement, second-third page, proportional tuplet

realization ……………………………………………………………………….… 359

6-17 Key’s Bicinia, first movement, second-third page, proportional tempos ………… 360

6-18 Key’s Bicinia, first movement, second-third page, pan-rational realization …...… 361

6-19 Key’s Rithmomachia: Icosihexaplex, mm 8 – 16 ……………………………...…. 364

6-20 Key’s Rithmomachia: Icosihexaplex, mm 26 – 31 …………………………..…… 364

6-21 Key’s Rithmomachia: Icosihexaplex, mm 40 – 46 ………………………….……. 364

6-22 Key’s Rithmomachia: Icosihexaplex, mm 122 – 128 ……………………….……. 365

6-23 Khachaturian’s Trio, indivisible polyrhythmic stratification, notated

using layered classical time signatures ……………………………………..…….. 369

6-24 Key’s Saxophone Quartet No. 1, indivisible polyrhythmic stratification,

notated using layered classical time signatures ……………………………….….. 369

6-25 Key’s Octet, indivisible polyrhythm, notated using pan-rational

mensuration ……………………………………………………………………..… 370

15
 Abstract of Dissertation Presented to the Graduate School
Of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

PAN-RATIONAL AND IRRATIONAL RHYTHM: THE HISTORY, DEVELOPMENT, AND

MODERN IMPLEMENTATION OF NON-DYADIC RATIONAL RHYTHMS IN WESTERN
MUSIC

By

Jordan Alexander Key

May 2021

Chair: Paul Sidney Richards

Major: Music

Three times in the history of Western Music - at the end of the 14th century, the end of the

16th century, and the beginning of the 20th century – has there been a flowering in the

development of non-dyadic rational rhythmic hierarchies. Only in the last of these occurrences

has this development persisted continuously to the present; each time before, rhythmic

complexity collapsed into a system dominated by dyadic- and/or triadic-rational rhythmic

hierarchies. By the 17th century, even triadic-rational rhythmic hierarchies had totally

disappeared from musical discourse to be supplanted by our modern system of dyadic-rational

time signatures. Even into the 21st century, dyadic-rational time signatures remain predominant,

despite work by composers like Henry Cowell and Conlon Nancarrow, which suggested the

possibility of a rhythmic paradigm shift during the early- and mid-20th century.

Despite the persistent hold of dyadic-rational time signatures, developments in

prescriptive rhythmic complexity during the 20th century have continued to the present,

persisting over multiple generations of composers and forming distinct schools of musical

discourse popular in contemporary classical music today. Among these composers are not only

Charles Ives, Henry Cowell, and Conlon Nancarrow, but also Thomas Ades, Brian Ferneyhough,

16
Michael Gordon, Karen Khachaturian, Milton Babbitt, Elliott Carter, and Jonathan Dawe

among others. Within their oeuvre, each of these composers has encountered the need for a

broader exploration, development, and notation of rhythmic structures beyond our current

dyadic-rational system, allowing into their musical systems pan-rational time signatures,

irrational time-signatures, and/or indivisible rhythmic hierarchies – all these rhythmic

prescriptions either not seen since the 16th century or never before seen in Western music.

Given the present state of our system of music notation and rhythmic prescription within

it, what are we doing and what can we do now in the 21st century with the rhythmic tools

developed in the past one hundred years? By thoroughly understanding the history of

prescriptive rhythmic experimentation in Western Music, we can possibly better understand why

certain systems of rhythmic notation have persisted while others have been forgotten; through

such better understanding of the history of rhythmic notation we might fashion a notational

system today that overcomes our present limitations in rhythmic prescription better than previous

failed models. To this end, I will trace the historical development of systems of rhythmic

hierarchy from Medieval music to Modern music, focusing on music with exceptional

prescriptive, precise, mathematically defined rhythmic structures, excluding aleatoric and

spatially based rhythmic notation. In doing this, we will gain a historical contextualization of the

rise of pan-rational systems of rhythmic notation. Following this, we will survey a variety of

modern compositional methods that expand standard prescriptive rhythmic notation, beginning

with Charles Ives and Henry Cowell and ending with living composes like Thomas Ades. Last,

this dissertation will address my own compositional work in the context of pan-rational systems

of rhythmic hierarchies and propose a new addition to the lexicon of rhythmic notation that will

emancipate the composer from traditional dyadically-rational rhythmic notation.

17
 CHAPTER 1
INTRODUCTION:
DEFINITIONS AND DELIMITATIONS

This dissertation examines instances wherein rhythm is exceptionally contextualized

beyond what might be considered “typical” contemporary rhythmic practice, represented simply

by the vast majority of rhythmic paradigms in any generation. This document does not propose

any exhaustive theory of rhythm and does not attempt to “analyze” any particular piece,

composer, or style. Rather, this document will attempt to historically contextualize exceptional

rhythmic practices, demonstrating the efficacy of various experimental systems of prescriptive

rhythmic notation in Western Classical Music from the Middle Ages to the beginning of the 21st

century. In so doing, we hope to not only better understand how rhythmic notation and

composition has come to its present state, but also understand how and why “dyadic-rational

rhythmic hierarchies” came to supremacy in Western music and route a better path forward

toward a renewed diversification of rhythm beyond dyadic rhythmic prescription and to a “pan-

rational” rhythmic notation. To this end, we aim to propose a new addition to the current system

of rhythmic notation, reinterpreting time signature to allow for current rhythmic symbols to pivot

their meanings between dyadic and non-dyadic rhythmic values and hierarchies. We will also

present an extension to our unidimensional hierarchy of rhythmic subdivision and combination,

allowing for a pan-rational implementation of rhythmic values concurrently. To accomplish these

proposals, we must first establish some standard and new definitions of musical terms along with

some useful mathematical descriptors. Following the establishment of definitions and

delimitations of this new system, we will conduct a survey of past notational reforms that

attempted to similarly diversify rhythm in order to understand the strengths and weaknesses of

these past systems so as to find the optimal middle ground between tradition and innovation,

18
semiotic diversification and unification, and notational flexibility and comprehensibility in

regard to rhythmic notation in music.

Over the course of the past millennium, since the dawn of modern Western music

notation to the present, numerous systems of rhythmic prescription have arisen, most of which

have been either wholly abandoned or only partially absorbed (usually in progressive

modification) into the predominant Franconian system first popularized in the early 14th century

(see Chapter 2). Why is it that so many systems failed to advance rhythmic notation? Why is it

that we still live in the shadow of Franco of Cologne nearly 1000 years since his innovation of

hierarchical rhythmic delineation by note shapes? Generally speaking, only a very few

modifications to this system have brought it to us from the High Middle Ages to the present.

Why did those particular innovations hold? Why were most innovations abandoned? Why do we

fundamentally rely on a dyadic rhythmic hierarchy in our current notation rather than triadic,

pentadic, or (more logically) some combination of these and all possibilities? Who and what

steered this aesthetic course of music notation and when and why was this particular system

adopted over the other possibilities offered by theorists and composers throughout history?

Kurt Stone (1911-1989), American music editor and musicologist, writes in his forward

to Gardner Read’s (1913-2005) book Sourcebook of Proposed Music Notation Reforms (1987)

that

…there is a rather prosaic reason why the many new suggested notational systems have
not been found acceptable by the musical community. Most musicians, after spending
years since childhood to become experts in dealing with the complexities and
shortcomings of traditional notation, are quite unwilling to learn new, unproven
additional systems, especially since common-practice notation has enabled them to
perform a vast range of music perfectly well.1

1
Read, Source Book, pg. x (in the forward by Kurt Stone).

19
Stone goes on to recognize that any notational innovation only “becomes desirable” or

“necessary” when “music becomes so much more complex or so completely different in

approach and technique that traditional notation can no longer cope with it….”2 Since the first

half of the 20th century, even before Henry Cowell’s publication of New Musical Resources (see

Chapter 4), Western music has demonstrated its inability to “cope” with the rhythmic complexity

demanded of it by many contemporary composers. However, no solution to our present system’s

inability to cope with the new demands of rhythm, particularly those outside the purview of

dyadicism, has attempted to systematically solve this shortcoming, and any partial attempt that

has been made (such as Henry Cowell’s) has never yet succeeded.

Gardner Read astutely recognizes the crux of the continual failure of revisionists,

acknowledging after his exhaustive survey of music notation reforms from 1700 to the mid-20th

century that “most revisionists seem to have been unaware that certain basic conditions must be

met before any proposed reform can succeed.” Such conditions, according to Read, are

evidenced by a clearer understanding of the history of notation itself, both the successes and

failures. Read recognizes and cautions that, for at least the two centuries he surveyed, “reformers

had assiduously pointed out the imperfections of conventional notation and then – uninformed of

proposals prior to their time — were utterly convinced they alone could devise a better method.”3

However, by Read’s estimation, no one will likely devise a better system of notation than one

progressively evolved overtime by wide consensus. Composer and music scholar Erhard

Karkoschka (1923-2009), author of the influential book Das Schriftbild der neuen Musik (“The

2
Ibid.
3
Ibid., 4.

20
Notation of New Music,” 1965), parallels Read’s assessment, asserting that our present system of

notation

has survived centuries of attempts at reform, and justifiably. In spite of certain

inadequacies, it has supported a many-sided musical culture, especially because it
presented pitches and durations to the eye in such a way that the sound corresponded
directly with the visual event even if the notation was not completely perfect.4

Logically then, by Read’s estimation, successful notational reform “should not stray too far from

the time-tested values of traditional notation,”5 and rather achieve a “synthesis”6 with it. Indeed,

for the very reason that nearly every reform of rhythmic notation does not achieve Read’s

“essential synthesis,”7 these innovations failed to make a lasting impact on music discourse and

the development of its notation.

While we do not assume that any system proposed herein will have a lasting impact on

music notation, recognizing the exceptional nature of pan-rationalism in rhythm and its relative

difficulty in execution, we do hope at least to propose a system more well-thought through by

means of thorough and critical historical contextualization, logical and mathematical

reevaluation and delineation of musical rhythmic objects as given (i.e. time signature), and

careful synthesis with current notational practice where effective. Thus, by our careful

formulation, we hope our notational system will have more merit than any previously proposed.

Before outlining this document and its intent further, it will be useful to establish some

crucial definitions, both mathematical and musical. These definitions, particularly those

regarding music, might vary slightly from conventional denotations. Furthermore, there are some

4
Karkoschka, Notation in New Music, 15.
5
Read, Source Book, 5.
6
Ibid.
7
Ibid.

21
musical terms which do not yet appear in the musical lexicon, and I propose that they should

replace inaccurate or misrepresentative terms currently in use.

Mathematical Definitions8

This document will deal intimately with the following mathematical concepts:

1. Divisibility and indivisibility

2. Commensurability and incommensurability9
3. Rationality and irrationality10
4. Dyadic and non-dyadic rational

Divisibility and Indivisibility

In mathematics, the divisor of an integer n (also known as a “factor” of that integer n) is

some non-zero integer m that can produce the integer n when multiplied by some other integer

a.11 If such a condition holds, then we may say that n is a multiple of m, and that n is divisible by

m. Under such conditions, dividing n by m leaves no remaining part, called the “remainder,”

only an integer value.

Simply stated, divisibility is the capacity of a dividend (n in the above definition) to be

exactly divided by a given number, or divisor (m), to produce an integer valued quotient (a).

Take for example the fraction 24/8; 8 is the divisor and 24 is the dividend, which is divisible by 8

8
These definitions, while equivalent to the standard mathematical definitions, are adapted to the purposes
of music composition and musicology and might therefore not be entirely comprehensive, only supplying a
necessary surface understanding of the term.
9
Also referred to as commensurate and incommensurate magnitudes.
10
Since here we are concerned with mathematical definitions, this should not be confused with any
psychological/neurological/philosophical denotation or connotation of these words.
11
In this document we will not consider the undefined case where the divisor is zero. Nor will we consider
the trivial case where the integer n (the dividend or numerator) is zero.

22
exactly 3 times (quotient).12 If an integer n cannot be equally divided into integer parts by its

integer divisor m (e.g. 26/11), then n is said to be indivisible by that divisor m.13

Although the above definitions will not change, for simplicity we will not consider

negative integers in this document, as there is generally no consideration of negative number in

rhythm.14

Rationality and Irrationality

In mathematics, a number c is considered to be rational if that number can be expressed

by the ratio of two integer numbers a and b, such that a/b = c, where b is non-zero. If this does

not hold, then we generally consider than number to be irrational. Rational numbers include not

only the whole numbers and the integers, but also fractions of integers, even those with repeating

decimal expansions, such as 2/3 and 3/11.

Let us consider rational and irrational numbers geometrically. Take for example a right

angle with side lengths three and four; by the Pythagorean Theorem we can deduce the length of

the hypotenuse is 5.

𝑎 = 3, 𝑏 = 4

12
Note that we refer to “dividends” and “divisors” when the two numbers in a fraction are equally
divisible, or when the dividend is a multiple of the divisor. However, when there is no assumed divisibility (n and m
might or might not be divisible), we can generally call the number to be divided the “numerator” and that number by
which we are dividing the “denominator.”
13
Note that we refer to 1, −1, n and −n as the “trivial divisors” of n. Any divisor of n that is not a trivial
divisor is known as a non-trivial divisor. It is interesting and important to note that the units −1 and 1 and the prime
numbers have no non-trivial divisors.
14
While some tentative uses of “negative rhythms” or “negative time signatures” have been proposed in the
past decade, albeit not in any academic literature, no regular use has yet been made of them and no uniqueness has
been given to them such that their nature cannot also be expressed using already extant musical signs. The most
common proposed use of negative time-signature suggests that such signatures would signal the inverted accenting
of the measure as traditionally understood. For example, while 4/4 would typically accent beats 1 and 3, -4/4 would
accent beats two and four. Such an inverted accenting could also simply be notated using accent marks under a
typical 4/4 time signature. See Neely, Adam. (2018, March 12). Negative Time Signatures [Video file]. Retrieved
from [Link]

23
 𝑎2 + 𝑏 2 = 𝑐 2
32 + 42 = 𝑐 2
9 + 16 = 25 = 𝑐 2
√25 = √𝑐 2
5=𝑐

There is no number here which cannot be expressed as either a whole number or the ratio of two

other whole numbers. However, let us consider two sides equal to one.

𝑎=𝑏=1
𝑎 + 𝑏2 = 𝑐 2
2

12 + 12 = 𝑐 2
1 + 1 = 2 = 𝑐2
√2 = √𝑐 2
√2 = 𝑐

This cannot be simplified any further. There is no precise square root of two. We could

approximate it by decimal expansion as 1.4142, but this is only an approximation since there is

no repeating pattern to the decimal expansion.

To demonstrate this approximation, let us square 1.4142 (multiply the number by itself).

1.4142
𝑥 1.4142
1.99996164

While 1.99996164 is close to two, it is not precise. In fact, there is no rational number that we

can multiply by itself such that we will get 2. We can prove this through contradiction, by

assuming such a rational number does exist. Thus, assume there exists a rational number in

𝑎 𝑎 2
lowest terms such that = √2. This then implies that (𝑏 ) = 2 → 𝑎2 = 2𝑏2 (squaring and
𝑏

multiplying both sides by b). Given that 𝑎2 = 2𝑏2 , we can assume that 𝑎 must have been

𝑎2
divisible by 2 since = 𝑏2 (dividing both sides by 2). Then, we understand 𝑎 as equaling some
2

24
other integer multiplied by two; let us call this unknown integer 𝑘. Thus, 𝑎 = 2𝑘 for some 𝑘.

4𝑘 2
Now, we may substitute 2𝑘 for 𝑎 back into our original premise, giving us = 2 → 4𝑘 2 =
𝑏2

2𝑏2 . Note that this is now a contradiction since 𝑎 and 𝑏 were initially assumed to be in lowest

terms; they are no longer in lowest terms since we can divide through by 2 on both sides to
𝑎
simplify the expression. Thus, there can be no such rational such that 𝑏 = √2 .15

Commensurability and Incommensurability

In mathematics, two non-zero real numbers16 a and b are said to be commensurable if

their ratio a/b is a rational number; otherwise, a and b are called incommensurable. For

7√2 7
example, the numbers 7√2 and 3√2 are commensurable because their ratio, 3 = 3, is a
√2

rational number despite the constituent numbers, 7√2 and 3√2, themselves being irrational.

√2
Contrastingly, the numbers √2 and 3 are incommensurable because their ratio, , is an irrational
3

number. Note that commensurable is not equivalent to the concept of divisibility; 7√2 and 3√2

are commensurable but are not divisible. Two numbers that are commensurable, are not

necessarily divisible. However, if two numbers are divisible, then they are commensurable as

well.

The concept of incommensurability is strongly related to the concept of irrationality. If a

number is irrational, then it is also incommensurable. If a number is incommensurable, it is

necessarily irrational.17

15
Although this is a classical proof for the irrationality of the square root of 2 and has been known since the
ancient Greeks, thanks is given to mathematician Jason Edmund Johnson for checking my proof.
16
NB: real numbers include both the rational and irrational numbers.
17
It is also interesting to note that Euclid's concept of commensurability is somewhat anticipated in in
Plato's dialogue, Meno, wherein Socrates in a classical Socratic dialogue uses a slave boy's own inherent logic to

25
Dyadic and Non-Dyadic Rational
𝑎
In mathematics, a dyadic rational is a rational number in the form of whose
𝑏

denominator is a power of two when the ratio is in coprime or minimal terms. Simply
𝑎
demonstrated, a dyadic rational is a number in the form of where a is an integer and b > 0
2𝑏

and is a natural number. Examples of dyadic rationals thus include 4/1, 3/2, 3/4, 7/8, 5/16, and so

on.

Any rational number not in the above form can be considered a non-dyadic rational (e.g.

1/3, 2/5, 7/11, etc.), except those where b = 0. Non-dyadic rationals may include a variety of
𝑎 𝑎 𝑎
subsets, such as triadic rationals 3𝑏 , quintic rationals , septic rationals 7𝑏 , and so on. Note,
5𝑏

𝑎
however, that subsets like quartic rationals are actually a subset of the dyadic rationals, since 4𝑏

𝑎
is simply . In fact, only powers of prime numbered denominators will form unique subsets
22𝑏

within the non-dyadic rationals.

Musical Definitions

This document will deal intimately with the following musical concepts:

1. Scientific, actual, and musical time, prescriptive and non-prescriptive

2. Prescriptive and non-prescriptive rhythm
3. Tempo and Speed
4. Measure and meter
5. Rhythmic hierarchy: mensuration and time signature
6. Dyadic and non-dyadic rational rhythmic hierarchies

demonstrate and solve a complex geometric problem. In essence, Plato develops a proof which is intriguingly
Euclidean, speaking to the concept of incommensurability and irrationality prior to its codification in The Elements.
It is likely, however, that such concepts as commensurability and incommensurability were well-known prior to
Euclid’s discussion of them. See Plato's Meno. Translated with annotations by George Anastaplo and Laurence
Berns. Focus Publishing: Newburyport, MA. 2004.

26
Time: Preliminary Axiom

First, it will be given that any piece of music deals intrinsically with time. Music cannot

exist without time. Music can be visually represented, and such visual representations can be

abstractly considered independent of time except when trivially considered. However, no

audiation of music can happen without time. Consequently, no matter the contents of the visual

representation of music (this includes a wave function or wave file for instances of music

without notation), this visual, which we will simply call a score, represents something to do with

time.

Time in music, however, is not necessarily understood according to any convention of

time established by seconds, minutes, or hours. While some pieces of music may be delineated

using conventional notions of time as understood by the rotation of the earth or the microwave

spectra frequency of a cesium atom, music generally can be played at any arbitrary rate

arbitrarily delineated, often in units called “beats,” “measures,” “phrases,” or “tacti” (tactus for

singular). Beats or any arbitrary time unit in music may be understood in relation to established

metrics of time (seconds for example) but not necessarily so.

Thus, we can suppose some necessary distinction between time as conventionally

understood and time as musically understood, the latter being a specialized case of the former.

Thus, we will establish three genres of time.

Time: Scientific, Actual, and Musical

It will be necessary to first delineate various definitions of time to understand rhythm in

music. Note, however, that this is not a dissertation on the philosophy of time, and so our

discussion of it will only serve our particular focus herein and will consequently not be

exhaustive.

27
 Let us then establish three different modes of time: scientific, actual, and musical. This is

by no means the gamut of temporal modes; however, this will suffice for our purposes here. We

will simply define scientific time as the common regular measurement of time according to

seconds, minutes, and hours, defined not only by the periodic rotation and revolution of the

Earth, but also by modern atomic clocks according to the hyperfine transitions in hydrogen,

cesium, or rubidium.

Next, we will use Susanne Langer’s definition of actual time, distinguishing it from the

relatively absolute measurement of scientific time, in which actual time operates. Langer also

provides a definition for musical time, but we will not consider this definition for reasons

mentioned below.

According to Langer, actual time is “composite, heterogeneous, and fragmentary.” By

“composite,” Langer means that actual time is generally understood by people according to a

“composed” sense of time organized according to various observations of change in

environmental conditions (change of days, seasons, age, decay, etc.). By “heterogeneous,”

Langer means that our sense of actual time will vary from circumstance to circumstance and is

relative from person to person. Last, Langer believes that actual time is “fragmentary” because

people are generally not aware of every moment that passes in time. Simply, actual time is

different from scientific time in that it is generally how people “feel” time, not how time truly is.

Langer’s definition of actual time will be useful; however, Langer’s belief that actual time and

musical time are intrinsically different18 is somewhat myopic. Consequently, Langer’s temporal

philosophies reach their limits of usefulness for us here.

18
Langer, S. K., Feeling and Form. London (Routledge & Kegan Paul, 1953), 109-113.

28
 Rather than consider actual and musical time as intrinsically different, we understand

musical time to exist in actual time, as was our original supposition. For a sense of musical time

to exist, we must have actual time (and consequently scientific time). Just as a square is just a

subset of rectangles, musical time is just one mode of generalized actual time.

Musical time is one mode of structuring, delineating, and presenting actual time, perhaps

allowing one to become more aware of actual time’s passage, since the act of listening to any

piece of music to the exclusion of other activities (listening to music not as a background

experience to some other activity) requires the active choice to “spend” actual time. Musical time

is a specialized from of actual time in that musical time is a controlled, artificial experience of

actual time designed by its performer and composer for a relatively confined temporal space.19

Musical time heightens and perhaps hyperbolizes the experience of actual time, allowing for the

coexistence of differentiated strata of time; the passage of these strata at various rates; the

suspension, acceleration, deceleration, reversal, or general and relatively exaggerated fluctuation

of time’s passage.

Musical Time: Prescriptive and Non-Prescriptive

The design of musical time may be prescribed a priori before the audiation of that music.

Equally, musical time may not be precisely prescribed or indicated at all if there is no designed

system to communicate the time dimension of that music apart from the given scientific and

actual time experienced regardless of musical time. For a music to have a prescribed musical

time it must have some method, either written or directly communicated, to encode the nature of

the intended musical time. If there is no method to communicate the composition of musical time

19
This obviously excludes such fringe examples as John Cage’s As Slow as Humanly Possible or other
similar pieces, which take exceptionally long periods of time to complete, ranging from days to centuries.

29
before audiation, meaning that the musical time cannot be understood before its execution in

actual time, then the music has a non-prescriptive musical time.

Prescriptive musical time may be as simple as telling someone that the sounds they

produce, once begun, should have some distinctive pulse every second and should have a

stronger, macro-pulse after every four seconds (understood in common musical jargon as 4/4

time-signature with 60 beats per minute). Obviously, in this example there are many elements to

musical time’s ordering that are left undetermined; however, there is still an element of

prescription before audiation in actual and scientific time. Contrastingly, if one were to say,

“make whatever sounds you like and use this picture of a vase of flowers as inspiration,” there is

no intrinsic language that communicates any previously defined ordering to musical time,

consequently leaving any temporal decision to the performer at the moment of audiation in

actual time; this is non-prescriptive musical time – musical time will happen with the audiation

of music, but cannot be known before audiation.

Consequently, prescriptive musical time is “scribed” in actual and scientific time and

audiated in musical time (our “pre-scribing” is realized once we enter musical time), while non-

prescriptive musical time is essentially “scribed” in musical time. This dissertation will only

consider prescriptive musical time.

Rhythm: Prescriptive and Non-Prescriptive

Prescriptive musical time, as demonstrated before, may range from vague to detailed

prescription. Thus, we can assume a spectrum of prescriptiveness from highly nebulous to

exceedingly precise.

Those musics which will be more ambiguous in their delineation of musical time will

necessarily consider time in indefinite, larger intervals with low granularity. For example, some

30
musics will indicate general musical materials to be played over a duration of scientific time (i.e.

“play this gesture for 30 seconds,” “this passage should last one minute”) or over a felt duration

of actual time (i.e. “play this passage as slow as humanly possible,” or “play this collection of

motives until the flute starts playing, then change to this.”). These vague prescriptions of musical

time provide an exceedingly blurry, low granularity of musical time; while one might have

specified the time frame of “one minute,” the elements of time within that minute are indefinite.

Those musics which will be more precise in their delineation of musical time will

necessarily delimit time with greater granularity. Rather than only specifying temporal details on

large scales or in equivocal terms, a music of high temporal granularity will delineate musical

time in smaller intervals with further specificity. To manage complex organization of musical

time, music generally will incorporate more intricate structures for encoding information to

prescribe additional temporal data points in order to precisely control the passage of musical

time.

Consequently, musics which require high temporal granularity require a panoply of tools

to demarcate musical time at ever small intervals: measures, beat groupings, beats, beat

subdivisions, etc. Those tools which help to precisely delineate musical time we will call

prescriptive rhythms. Those tools which help to only vaguely delineate musical time we will call

non-prescriptive rhythms. This dissertation will only consider prescriptive rhythms and thus is

intrinsically concerned with rhythmic musical notation with relatively high granularity.

Tempo and Speed

The relationship between arbitrary time units in music (e.g. beats) and established time

units (e.g. seconds) is called tempo (pl. tempi). Equivalently, the relationship between musical

31
time and scientific time is called tempo. Simply put, tempo is the rate of regular successions of

arbitrary units of time in music (e.g. “one beat per second,” “five notes per week,” etc.).

Traditionally, tempi could be established using descriptors without specified rate (e.g.

“Allegro,” “Fast,” “March,” etc.). Here we will exclude such descriptors from our understanding

of tempo. Rather, vague terms that attempt to describe the rate of the music will be called speed.

Tempo, contrastingly, prescribes the rate of time in music as compared to the rate of time

exterior to music. For example, the speed of the music is “Fast,” but the tempo is precisely 180

beats per minute. Thus, speed is descriptive while tempo is prescriptive.

Measure and Meter

As a principle, we will understand measure and meter as musical concepts existing in

prescriptive musical time; however, measure and meter may exist with either prescriptive or non-

prescriptive rhythms.20

A measure is a grouping of individual pulses in music. These pulses are established

through rhythm, either prescriptive or non-prescriptive. Thus, a measure is a macro-unit of

musical time containing definite delimiters of that macro-unit of time. The number of pulses in

one measure may vary over a collection of measures. However, all measures must have at least

one pulse.

A measure may be either an abstract or concrete measure for organizing musical time.

An abstract measure is a measure that is not necessarily perceptible to the person listening to the

music or may not be designed to be audible. A concrete measure clearly establishes perceptible

20
For example, pure improvisation uses rather non-prescriptive rhythmic devises: general tempo, general
pulse, etc. Consequently, improvisational musics like jazz will have macro and micro beat structures. However, pure
improvisation is, by nature, un-notated, and so beats are not generally precisely prescribed.

32
macro units of musical time. This dissertation will not concern itself with establishing whether a

measure is abstract or concrete, however it is important to recognize that not all “measures” as

prescribed are necessarily perceptible, such as measures that are significantly long (e.g. a

measure of 20/4 in a tempo of 60 1/4 beats per minute). Nevertheless, such large measures may

be prescribed, likely for compositional purposes more than auditory ones.

While measure focuses on macro units of micro delineations of musical time, meter

describes groupings of these macro units. Namely, a collection of measures of a similar length in

concatenation defines a meter.

In this dissertation we will not consider measure and meter arising from non-prescriptive

(or improvised) rhythms. Only those rhythms which are prescribed, or notated, before

performance will be considered herein.

Rhythmic Hierarchy: Mensuration and Time Signature

In all common time signatures found in Western Music, the hierarchy of rhythmic

subdivision of any time unit will always be dyadically rational. If we refer to traditional time

signature, we refer to the system of dyadic rationalism. Henceforth we will consider time

signatures to not necessarily be exclusively “traditional.”

In musics with higher granularity of rhythmic prescriptiveness, measures are not only

subdivided into time units, but these smaller time units are also subdivided into even smaller

time units. The collection of various strata of temporal subdivisions in musical time is called the

rhythmic hierarchy. A particular stratum within a hierarchy is called a subdivision. All

subdivisions have an order based on their cumulatively multiplied subdivision. For example, the

third subdivision of the traditional time signature 2/2, or the eighth note, is called the third order

33
subdivision. That time unit from which all subdivisions are derived, also called a “measure,” may

also be abstractly referred to as the fundamental time unit (FTU).

Traditionally, there are two means by which to establish a fixed rhythmic hierarchy: time

signature and mensuration.

• Time signature is either a perfectly or imperfectly regular hierarchical system wherein a

FTU is established and then the FTU’s first order subdivision is prescribed and

partitioned. From this partitioning a new macro-grouping (or measure) of these equal

subdivisions may or may not be prescribed. Beyond the first order subdivision, however,

partitioning is disallowed, otherwise this would logically contradict the original

partitioning. Note that subdivision and partitioning are totally independent operations,

meaning that any choice to partition and form a new macro-grouping of pulses less than

or greater than the FTU has no bearing on the subsequent divisibility of the first order

subdivided time units.

• Mensuration is a system to establish either a perfectly regular, imperfectly regular, or

irregular hierarchy of prescriptive rhythmic subdivisions based on an established, non-

partitionable fundamental pulse.

By partitionable, we mean that the FTU, once divided into equal parts, may be reduced or

supplemented by one or more of those parts. For example, the time signature 4/4 tells us that the

FTU should be divided into four equal parts, and then all four of those parts should be grouped to

form a measure. However, the time signature 3/4 tells us that the FTU should be divided into

four equal parts and then only three of those parts should be grouped to form a measure. 4/4 is

unpartitioned while 3/4 is partitioned. By non-partitionable we mean that the FTU, once

divided into equal parts, must maintain all those parts without exclusion or addition.

34
 By regular we mean that all further subdivisions beyond the first order subdivision

accord to the first order subdivision. Namely, if the first defined subdivision of the FTU is into n

equal parts, then all further subdivisions can be subdivided into any of the prime factorizations of

n. Perfectly regular hierarchical rhythmic systems subdivide at all levels into the same number

of equal parts. This will occur should the first subdivision have a prime factorization with only

one factor.21 The most well-known example of this is our current system of time signature,

wherein all subdivisions on all levels are into two equal parts. Imperfectly regular hierarchical

rhythmic systems subdivide according to a systematic pattern established based on the prime

factorization of the first subdivision but are not necessarily the same equally divided subdivision

at all levels. This will occur should the first subdivision have a prime factorization with more

than one factor.22 Consequently, first order subdivisions by any prime number or prime power

will be perfectly regular, while any first order subdivision by any non-prime or non-prime power

will be imperfectly regular. By irregular we mean that at some point in the sequence of

progressive subdivisions some order is not derived from the prime factorization of the prior

order. See Figure 1-01 for a clarification of this classification system.

21
In standard parlance, one might call this kind of time signature “simple,” such as in 3/4, which, without
the application of some exceptional artifice like a tuplet, cannot be subdivided into any other equal division but 2 on
any level.
22
In standard parlance, one might call this kind of time signature “compound,” such as in 6/8, which, by
application of an augmentation dot, could be subdivided on any level into either 2 or 3 equal parts.

35
Figure 1-1. Classification system of rational time signatures.

Take for example 3/4, wherein all subdivision levels are divisible by the prior level

beginning with the first (recall subdivision and partitioning are independent operations and the

manner of subdivision is dependent on the prime factorization of the original subdivision of the

fundamental pulse, here 2).

1 fundamental pulse (measure) 4 1⁄ one quarter note 8 1⁄ one eighth note 16

4 8
4⁄ quarter notes = = = = 2⁄ two sixteenth notes = =…=1
4 4 2/8 two eighth notes 8 16 16

Thus, 3/4 is perfectly regular. Similarly, 3/5 is perfectly regular.

1 fundamental pulse (measure) 5 1⁄ one 5th note 25 1⁄ one 25th note 125
5 25
5⁄ fifth notes = = = = 5⁄ = =…=1
5 5 5/25 five 25th notes 25 125 five 125th notes 125

5/6, however, is imperfectly regular. The prime factorization of 6 is 2 and 3. Thus, at the second

order subdivision, we may choose either 2 or 3 as the subdivision unit. Following this, if we

choose the route of duple subdivision, then we will construct 12 th note subdivisions. Equally, if

we had taken 3 as the second order subdivision, then we would construct 18th notes. This

continual binary option between 2 and 3 at every order of subdivision creates “regularity” in the

rhythmic hierarchy, however our location in this continually branching tree may fluidly migrate

36
from branch to branch without any collapse into a non-branching order of subdivision. Thus, we

call this imperfect regularity.23

More simply put, perfect regularity describes a hierarchy without branches, while

imperfect regularity describes a hierarchy with patterned branches. An irregular system is a

linear hierarchy without prime factorization patterning (e.g. the first order subdivision might be

into fifths, but the second then might be into sevenths, and then the third into halves). In

exceptional circumstances such irregular systems are described using mensural hierarchies

wherein each subdivision necessary in the music is delineated. We will encounter these

mensurations later.

Theorem 1: time signatures describe either patterned trees or lines of rhythmic

hierarchies, while mensurations describe only patterned or un-patterned linear

hierarchies.

In fact, mensuration is highly similar to time signature as understood here. Mensuration

is different from time signature only in that it can prescribe totally irregular patterns of

subdivision and that it cannot be partitioned. If we understand time signature as representing a

tree of possible rhythmic subdivisions with the exceptional case of perfect regular time

signatures where there is only one hierarchical path, then each branch of this tree represents one

mensuration. Consequently, a perfect regular time signature, being one branch, is itself a

23
As suggested in footnote 22, imperfect-regular time signatures are similar, but not perfectly congruent to
the ability of 6/8 in traditional time signatures to either be construed as compound duple (6/8 with two beat
groupings of three 8th notes) or as triple-duple (6/8 with three beat groups of two 8th notes, equivalent to 3/4). This
property of 6/8 however, does not carry to all levels of its rhythmic hierarchy like 5/6 (barring the use of appended
artifices like the dot or tuplet); as soon as we take the path of triple-duple in 6/8, we can no longer find a continued
path down this branch into a triple subdivision. Once we define the quarter note or eighth note (rather than the dotted
quarter note or dotted eighth note) as our order of subdivision, then we are irrevocably stuck in duple, since one
quarter note or one eighth note cannot be equally divided into three parts without some tool exterior to the
established hierarchy (namely, tuplets). Ideally, compound 6/8 would be better written as 6/6, since this would allow
for continuous branching into either triple or duple subdivisions at all levels without the use of tuplets or
augmentation dots.

37
mensuration, while an imperfect regular time signature represents a family of related

mensurations. Mensuration is only distinct from time signature in cases where the established

rhythmic hierarchy is irregular.

Theorem 2: All time signatures may be translated into a mensuration or a set of

mensurations, but not all mensurations may be translated into a time signature.

Take for a first example, the classical time signature of 4/4. First, we subdivide the FTU

into 4 equal parts without partitioning. Next, we recognize that the first order subdivision has a

prime factorization of 2. Thus, we have only one path in this rhythmic hierarchy tree at each

level: a subdivision by 2. Thus, we may say that this time signature is regular.

Demonstration 1: time signature 4/4 (non-partitioned regular time signature) =

• [Link] (duple) = perfect regular mensuration

Each progressive order of subdivision provides us with 2 times as many subdivisions as the

previous order. This time signature corresponds directly with one mensuration. Thus, we can say

that this time signature (and its corresponding mensuration) represents not only a rational

rhythmic hierarchy but also a divisible hierarchy. Specifically, this is a dyadic rational hierarchy.

Thus, a 4/4 time signature represents a non-partitioned, regular, perfect, divisible, dyadic,

rational rhythmic hierarchy.

For another example, take the time signature of 6/6. First, we subdivide the FTU into 6

equal parts without partitioning. Next, we recognize that the first order subdivision has a prime

factorization of 2 and 3. Thus, we have two paths in this rhythmic hierarchy at each level: either

subdivisions by 2 or 3. Consequently, the second order subdivision will provide us with two

possibilities – either duple or triple – while the third order subdivision will provide us with four

ultimate possibilities – triple-duple, triple-triple, duple-triple, and duple-duple. All the possible

rhythmic hierarchy trees are outlined below:

38
 Demonstration 2: time signature 6/6 (non-partitioned imperfect regular time signature) =
• [Link] (triple-duple) = imperfect regular mensuration
• [Link] (triple-triple) = perfect regular mensuration
• [Link] (duple-triple) = imperfect regular mensuration
• [Link] (duple-duple) = perfect regular mensuration

Each of these branches is one mensuration, all of which are regular but two are perfect and two

are imperfect. Notice that each of these mensurations represent a sequence of rational numbers,

each a multiple of the previous. Thus, we can say that these mensurations represent not only a

rational rhythmic hierarchy but also a divisible hierarchy. Specifically, this is a non-dyadic

rational hierarchy.

Thus, we say that a 6/6 time signature represents a non-partitioned, regular, imperfect,

divisible, non-dyadic, rational rhythmic hierarchy.

For another example, consider the time signature of 5/8. First we subdivide the FTU into

8 equal parts with partitioning by the removal of three of these equal first order subdivisions.

Here, we should recognize that since this time signature is partitioned, it will have no directly

corresponding mensuration. Next, we recognize that the first order subdivision has a prime

factorization of 2. Thus, we have only one path in this rhythmic hierarchy tree at each level: a

subdivision by 2. Thus, we may say that this time signature is perfectly regular.

Demonstration 3: time signature 5/8 (partitioned regular time signature) =

• [Link] (duple) = perfect regular mensuration

First, we must consider the partitioning and compensate for this on every level of the

rhythmic hierarchy if we wish to accurately fill the specified measure no matter what order of the

hierarchy we might be operating within at any moment. Let t be the number of subdivisions of

the FTU on any subdivision order m. Let n be the number of subdivisions in the first order

subdivision, p the prime factor by which we are subdividing within every level, and k be the

number of subdivisions added (or subtracted) from the first order subdivision. Thus, the number

39
of subdivisions within any subdivision order in a perfectly regular rhythmic hierarchy can be

determined by the formula:

Formula 1: 𝑡𝑚 = 𝑛(𝑝𝑚−1 ) + 𝑘 (𝑝𝑚−1 ) = 𝑝𝑚−1 (𝑛 + 𝑘)

It is important to note that this formula only holds true for perfectly regular hierarchies. Also

note that for the first order subdivision, 𝑝𝑚−1 will always be equal to 1 since any number to the

power of zero is 1. Thus, the value of p in the first order subdivision is trivial and may essentially

be any integer. Given the above formula we can derive the number of subdivisions at any level

with ease.

First order subdivision

𝑡1 = 2𝑚−1 (𝑛 + 𝑘 ); 𝑛 = 8, 𝑚 = 1, 𝑝 = 2, 𝑘 = −3

21−1 (8 − 3) = 20 (5) = 1(5) = 5

Thus, five subdivisions within the first order subdivision

Second order subdivision

𝑡2 = 2𝑚−1 (𝑛 + 𝑘 ); 𝑛 = 8, 𝑚 = 2, 𝑝 = 2, 𝑘 = −3

22−1 (8 − 3) = 21 (5) = 2(5) = 10

Thus, ten subdivisions within the second order subdivision

Tenth order subdivision

𝑡10 = 2𝑚−1 (𝑛 + 𝑘 ); 𝑛 = 8, 𝑚 = 10, 𝑝 = 2, 𝑘 = −3

210−1(8 − 3) = 29 (5) = 512(5) = 2560

Thus, 2560 subdivisions within the tenth order subdivision.

For another example, consider the time signature of 2/6. First, we subdivide the FTU into

6 equal parts with partitioning by the removal of 4 of these equal first order subdivisions. Next,

we recognize that the first order subdivision has a prime factorization of 2 and 3, and thus is

40
imperfectly regular; consequently, we cannot use the above formula. Rather, we must develop a

branching algorithm to solve this problem sequentially as necessary.

We have only two paths in this rhythmic hierarchy tree at each level: a subdivision by 2

or 3. Thus, we have an infinite number of branching hierarchies as the tree grows. However, let

us only consider the first three orders of subdivision.

Demonstration 4: time signature 2/6 (partitioned, imperfect regular time signature) =

• [Link] (triple-duple) = imperfect regular mensuration
• [Link] (triple-triple) = perfect regular mensuration
• [Link] (duple-triple) = imperfect regular mensuration
• [Link] (duple-duple) = perfect regular mensuration

Let us now just consider the path along the triple-duple tree, [Link]. Since p will

unpredictably vary along this path, simply selected from any of the prime factors of the first

subdivision, we must progress through the hierarchy systematically, making the selections for p

as we progress. Thus, let t be the number of subdivisions of the FTU within any subdivision

order m. Let n be the number of subdivisions in the first order subdivision, and k be the number

of subdivisions added (or subtracted) from the first order subdivision.

It is established that 𝑡0 = 1 and 𝑡1 = 𝑛 + 𝑘 always. Then 𝑡𝑚 = 𝑝𝑡𝑚−1 where p is a prime

factor of n. Thus, to know 𝑡5 we must first know 𝑡4 and so on. Let us solve the first three orders

of subdivision for 2/6 time signature on the triple-duple branch [Link].

n = 6; m = 1,2,3; p = p,3,2; k = -4

• Zeroth order: 𝑡0 = 1
• First order: 𝑡1 = 𝑛 + 𝑘 = (6 − 4) = 2
• Second order: 𝑡2 = 𝑝𝑡1 = 3(2) = 6
• Third order: 𝑡3 = 𝑝𝑡2 = 2(6) = 12

Thus, accounting for partitioning, there will be 1 FTU divided into 2 first order subdivisions (2

sixth notes), then these 2 sixth notes will be divided into 6 second order subdivisions (6 18 th-

notes), and then these 6 18th-notes will be divided into 12 third order subdivisions (8 36th notes).

41
 This particular branch of 2/6 time signature is congruent to the classically understood 6/8

time signature. However, 6/8 has nothing intrinsic to it to suggest, without a priori convention,

that the FTU of a 6/8 measure should be subdivided first into two parts. 6/8 could be subdivided

into two or three parts in its first subdivision (the latter division being equivalent to 3/4).

However, 2/6 gives its reader a clear instruction to first divide into two parts, each of which

could be subdivided into either two or three smaller parts.

As an alternative example, let us consider an irregular mensuration (recall that

mensurations cannot be partitioned). Since there is no defined pattern to an irregular

mensuration, we must specify each necessary subdivision of the fundamental unit.

Demonstration 5: mensuration [Link] (irregular)

First, this mensuration tells us that our FTU is first divided into 5 equal parts. Second, we

must divide 5 into some equal parts to get 7 equal subdivisions of the FTU; however, there is no

way to equally divide 5 such that we then can subdivide our FTU into 7 equal parts. Whereas

before, we could divide each of our 6 equal subdivisions into three equal parts to produce 18

equal subdivisions, there is no integer we can multiply 5 by to get 7 (7 is simply not a multiple of

5, or rather 7 is not divisible by 5).24

Consequently, the second order subdivision cannot be derived from the first order. Our

first order subdivision will be into 5 equal parts of the FTU and our second order subdivision

will be into 7 equal parts of the FTU, and there is no divisibility between these orders. This is

24
Note that this should not be confused with the process of subdividing each of the first five subdivisions
into seven equal parts and each of these seven equal parts into 12. This would not produce seven and twelve equal
subdivisions of the FTU, but rather 5(7) = 35 and 5(7)(12) = 420 equal subdivisions of the FTU. Such a mensuration
would look like [Link] and could possibly be represented by the time signature X/(5)(7)[(2)(3)] = X/210, since
2,3,5,7 are the various unique prime factors of the ultimate product 420. However, such a time signature is perhaps
unwieldy, though valid and imperfectly regularly patterned.

42
also the case with the third order subdivision. In typical musical terminology, such an irregular

mensuration might be understood as “5 in the time of 7 in the time of 12.”25

Thus, we note that this mensuration is not only a non-dyadic rational rhythmic hierarchy,

but it is also an indivisible hierarchy, however commensurate. So, we may say that the

mensuration of [Link] is an irregular, commensurably indivisible, non-dyadic rational rhythmic

hierarchy. This mensuration cannot be a time signature since this hierarchy is irregular.

Theorem 3: Mensurations can express incommensurate hierarchies (irrational

hierarchies) and commensurably indivisible hierarchies, whereas time signatures cannot.

Time signatures, by their nature, can only express commensurably divisible hierarchies,

which themselves contain one or more mensurations of the same quality.

Consider for one final example the “time signature” found in Conlon Nancarrow’s Study

for Player Piano No 33 (see Chapter 4 for more details on this piece). The time signature of this

piece is 2:√2, which is unquestionably an incommensurable ratio, and is consequently an

irrational number. Since such a rhythmic hierarchy implied by this ratio is intrinsically

indivisible, it cannot be a “time signature.” Rather, by our definitions, we will understand this as

a mensuration with only one linear hierarchy with two orders of subdivision, each of which is

incommensurate with the other.

While such a mensuration has classically been understood, and perhaps was understood

by Nancarrow himself, as a tempo proportion, the combined result will create an implied

rhythmic hierarchy. This alternative conceptualization is meaningful to attempt to understand. In

dissecting this rhythmic prescription, we can start by recognizing that this is simply telling us

25
Though we typically do not understand traditional time signatures as such, we could equally understand
4/4 time signatures as “2 in the time of 2 in the time of 2 and so on.” This example really does not mean much, since
this all reduces to 1 because all orders of the subdivision are similarly divisible. Such an understanding, however, is
useful in irregular mensuration.

43
that there are “2 pulses in some order subdivision in the same time of √2 pulses in another

order.” The level of these orders really does not matter, since mensurations provide all levels and

each level is simply a subdivision of the FTU.

Obviously, since this number is irrational, it can never be fully realized by any physical

machine, player piano, computer, or otherwise. However, we can consider not only the close

rational approximations of this ratio but should at least consider the aesthetic and musico-

philosophical questions posed by such a seemingly impossible compositional study.

While traditionally all time signatures are dyadic rationals, if we expand our notion of

time signature according to the above definitions then we can begin including non-dyadic

rational time signatures and even indivisible and incommensurate (irrational) time signatures and

mensurations in music. Furthermore, while we generally consider mensuration to be an

outmoded method to communicate rhythmic hierarchy, popular during the Middle Ages and

Renaissance, mensuration is an intrinsic part of all time signatures and may continue to allow for

rhythmic hierarchies that are impossible in time signatures, traditional or not. More will be said

of this in later chapters of this dissertation.

Polyrhythm and Polymeters

Simply, polyrhythm is the occurrence of two rhythmic hierarchies at the same time. To

be a true polyrhythm, these opposing hierarchies should not be wholly congruent. We can

establish degrees of similarity between polyrhythms by defining the order(s) on which their

respective trees overlap, if they overlap at all. A pure polyrhythm is a polyrhythm wherein there

is no overlap between the opposing rhythmic hierarchies. Pure polyrhythms are rare in music and

often feature a dyadic rational rhythmic hierarchy against one that is non-dyadic. This is seen in

44
notated music often by the use of tuplets in some collection of voices, which go against some

overriding meter present in other voices that are under no tuplet.

A polymeter, like meter, is the continuous concatenation of congruent measures, though

congruent measures of congruent polyrhythms. Polymeters may vary in degree or “pureness”

like polyrhythms. This degree is directly related to the degree of their combined polyrhythm as

defined above. A pure polymeter is pure if and only if its polyrhythmic structure is pure.

Musical Scope and Delimitations

Given the above definitions, we may now specify the delimitations of this study. First, we

will not consider non-prescriptive systems of musical time and rhythm. This dissertation will

focus on prescriptive systems of musical time, specifically prescriptive rhythmic hierarchies.

To this end, this dissertation will consider dyadic rational time signatures only generally as

part of a broader understanding of time signature; this includes partitioned dyadic rational time

signatures.

Specifically, this dissertation aims to explore historical examples of music utilizing non-

dyadic rational rhythmic hierarchies, both perfectly and imperfectly regular as well as partitioned

and non-partitioned. We will also examine irregular mensurations, both commensurable and

incommensurable, unavailable in time signature. Last, we will examine polyrhythms and

polymeters, in which are used non-dyadic rational rhythmic hierarchies and irregular

mensurations, both commensurate and incommensurate.

Document Layout

In proceeding with this historical and aesthetic overview of exceptional rhythmic

practices in Western music we will begin in Chapter 2 with the creation of hierarchical rhythmic

45
notation systems in medieval Europe and discuss the general development of rhythmic notation

in the Ars Antiqua and Ars Nova with the first systems of rhythmic hierarchies, both triadic and

dyadic. The ultimate goal of this chapter is to bring us to the Ars Subtilior, wherein we can

examine the first polyrhythmic and pan-rational rhythmic hierarchies. Chapter 2 will conclude

with a discussion of the notational collapse of the Ars Subtilior and attempt to assess the reasons

why this extended Franconian, pseudo-pan-rational notation failed to last beyond the beginning

of the 15th century.

Chapter 3 will discuss the rise of exceptional rhythmic practices during the 15th and 16th

century as a reaction to the concurrent proliferation of new mathematical aesthetics in Western

Europe. We will examine the proportional notation innovations of Tinctoris and Gaffurius and

trace this system’s migration to the British Isles where it would experience its greatest popularity

under John Dygon and “The Windsor School,” active at the end of the 16th century in the court of

Elizabeth I. Like Chapter 2, Chapter 3 will conclude with a discussion of the collapse of

proportional mensuration in favor of a simplified set of basic mensurations that would ultimately

evolve into our modern time signatures.

Chapter 4 will begin with a survey of the “second collapse” of rhythmic diversity during

the 17th and 18th century under the Classical paradigm of “balance, simplicity, and form.” This

survey will take us through the 19th century, wherein we will trace the rekindling of interest in

rhythmic diversity under mid and late Romantic composers and into the early 20th century.

Emerging from this rhythmic “dark age,” the second half of chapter 4 will examine four pivotal

figures of early and mid-20th century rhythmic diversification: Charles Ives, Henry Cowell,

Conlon Nancarrow, and Elliott Carter.

46
 Chapter 5 will trace the consequence of Ives’, Cowell’s, Nancarrow’s, and Carter’s

influence on late-20th century composers and the further evolution of rhythmic diversification

and complexification under “hyper-rhythmicity” and newly emergent “rhythmic pan-

rationalism.” Central to this discussion will be the works of Milton Babbitt, Brian Ferneyhough,

Michael Finnissy, Ben Johnston, Michael Gordon, Akira Nishimura, and Thomas Adès along

with the theoretical work of Paul Creston. Within this survey of late 20 th-century composers, we

will examine the problems emergent in contemporary music notation’s ability to “cope” with

mounting rhythmic complexity, particularly beyond the purview of dyadicism.

Following each of these above listed historical surveys, I will provide a “synthesis,”

wherein I will attempt to outline a fundamental theory of notation systems, generally applicable

but ultimately and specifically aimed towards a careful challenge to and modification of our

present notation system. Each synthesis will build basic Axioms, Definitions, Observations, and

Conclusions based on the historical survey that preceded it. Synthesis I, II, and III will culminate

in a “Meta-Synthesis,” wherein we will discuss, according to our own observations and those

made by music notation historians, what makes a successful modification to rhythmic notation.

Chapter 6 will present the new system of pan-rational rhythmic notation as developed

from our definitions supplied in Chapter 1; the survey of historical innovations to rhythm in

Chapters 2, 3, 4, and 5; development of a rudimentary theory of notational systems in Synthesis

I, II, and III; and fusion of the aforementioned observations and theories in the Meta-Synthesis.

In the presentation of this pan-rational system of rhythmic notation, I will address both its

advantages and shortcomings, showing how it is not a replacement for the system in present

usage but rather an extension of it. In this chapter, I will discuss some of my own compositions

that speak to this history of rhythmic heterogeneity. In presenting some of my own works that

47
use (or could use) this extension of rhythmic notation, I will demonstrate how a composer might

approach the problems of performability in complex rhythmic subdivisions and multifarious

hierarchies.

48
 CHAPTER 2
SUBTLE SIGNATURES FOR MUSIC’S TIME:
RHYTHMIC INNOVATIONS OF THE LATE MIDDLE AGES AND THE ARS SUBTILIOR

While a complete study of the development of systems of prescriptive rhythm and its

hierarchical structure in Western Music are not the purview of this document, it will be useful to

touch upon past periods of rhythmic experimentation in order to understand the problems and

limitations present in our current notational systems and address other possibilities for rhythmic

notation via past models. Through such a cursory survey of past and present rhythmic devices

used to expand de facto systems of dyadic and triadic rational hierarchies, we will be able to

more clearly devise optimal routes forward through a synthesis of past and present systems’ best

aspects. By this optimization of all available notation methods, we can potentially create a better

system by which to accomplish the ends of pan-rational rhythmic notation in music, which will

be fully outlined in Chapter 6.

This historical survey will cover three periods of “exceptional rhythmic practice,”

wherein schools of composers sought to notate music using rhythmic hierarchies outside the

traditional dyadic and triadic systems of their time. Consequently, while one might typically

locate rhythmic notational revolutions within the 13th century Notre Dame School, the Ars Nova

of Franco of Cologne and Philippe de Vitry, or the standardization of mensuration during the 16 th

century that would ultimately lead to the evolution of time signature as it is widely understood

today, we, in fact, set the locus of revolution in the “counter-revolutions” to these

“standardizations.” In examining “exceptional” rhythmic practice, we prioritize those schools

whose radical rhythms were counter to the overarching contemporary or traditional practice and

whose aesthetics and/or tools of expression did not ultimately survive due to a variety of factors,

contemporaneously and/or subsequently.

49
 As a consequence, the three chapters of this historical survey will concentrate on the late

14th and early 15th century Ars Subtilior, the late 16th century Elizabethan Avant-Garde with its

Continental antecedents, and 20th century composers inspired by the early and mid-20th century

experimentations of Charles Ives, Henry Cowell, and Conlon Nancarrow. Within each of these

parts, we will draw conclusions from each epoch’s experimentations and notational

developments to ultimately craft a new and hopefully more comprehensive, limited, and flexible

theory of rhythmic demarcation than we presently possess.

The Franconian Vision: the Ars Antiqua, the Ars Nova, and the Birth of Rhythmic
Hierarchy

In 1279, the anonymous music theorist of St. Emmeram produced a new and unaffectedly

practical guide, De Musica Mensurata, designed to comprehensively teach one to read and write

the illustrious conductus and discant rhythmic styles of the 12th and 13th century Notre Dame

School. As a book of music theory, it was in many ways unprecedented in its “leisurely” style

with “detailed, rich, and expansive”1 content, touching on all aspects of harmonically and

rhythmically modal music from that period. Retrospectively, we can see this treatise as a

culmination of a musical age, coming at the end of the Ars Antiqua and comprehensively

codifying that style to a greater length than any other extant treatise.2 Ironically, this massive

codification would be to posterity only; the following year would forecast the obsolescence of

the herculean collation of De Musica Mensurata, whose notation system would, within just a few

decades, be utterly and permanently abandoned.

1
Yudkin, Jeremy, and Anonymous of St. Emmeram, De Music Mensurata: The Anonymous Of St.
Emmeram (Bloomington: Indiana University Press, 1990), 1.
2
Ibid.. The work of the anonymous theorist of St. Emmeram comprises approximately 50,000 words, while
the treatises of Johannes of Garlandia, Lambertus, and Anonymous IV – all on the same subject - together amount to
only about 35,000 words.

50
 If we take the approximate date of 1280 as the year of composition for Franco of

Cologne’s treatise Ars Cantus Mensurabilis,3 then the turn of this decade in the late 13th century

marks a seemingly unremarkable yet truly pivotal shift in Western music notation. While

Franco’s treatise on a new system of musical notation to clearly demarcate the six rhythmic

modes of the Notre Dame School is often taught as a revolutionary and forward-thinking

invention in music, the truth of this manuscript’s inception is that it was, in its time, rather

conservative in its ambitions. Rather than proposing a new style of music (as it would ultimately

instigate in the later 14th century Ars Nova styles), Franco’s goal was to provide a clearer method

by which one could notate the music of the Ars Antiqua. Likely recognizing the overwhelmingly

numerous and frustratingly cumbersome rules outlined in treatises like De Musica Mensurata,

Franco wished to simplify the process of reading and writing music. However, the ultimate

ramification of this new, less obfuscated notation system would be far reaching once its full

implications were realized by subsequent generations of composers and theorists and would

ultimately make the Ars Antiqua as “antiquated” in the late 14th century as its present name

suggests to us in the 21st century.4

Far less determined by the contextualization of various neumes as was the Ars Antiqua

notation style, Franco’s systems of “shaped” and relatively fixed rhythmic values5 allowed for

the clear notation of notes which contained two equal parts. Prior to Franco’s system, notes were

3
See Hughes, Andrew, "Franco of Cologne," Grove Music Online, 2001. There is much debate about the
date of this treatise, ranging in popular consensus between 1260 and 1280 with good evidence to support both sides
of the time spectrum. Here, simply for some narrative context, we will follow the 1280 hypothesis.
4
The author is aware that the name “Ars Antiqua” was given to this music ex post facto as a conservative,
counter-stylistic descriptor to the later Ars Nova and not given by the contemporaneous practitioners of this music
during the 12th and 13th centuries.
5
Albeit the “perfection” or “imperfection” of the shaped values.

51
only seen as groups fitted into variations of trochaic metrical structures, many of which were

reflective of contemporary popular Latin verse forms. Such trochaic groupings implied metrical

sub-units of either two unequal parts (e.g. long + short or short + long) or three fundamental

equal parts (short + short + short), both contained in the unit of a singular accented pulse (see

Example 2-1 and 2-2 for samples of the genre). There was no clear or standardized method to

notate a singular pulse unit that would divide into sub-units of two equal values (medium +

medium). To arithmetically demonstrate the issue, consider the number 3: the Ars Antiqua

system recognized that 3 is the sum of 2+1, 1+2, and 1+1+1; however, this older system had no

clear way to show that 3 was also the sum of 1.5+1.5.6 Franco’s system, though it might not have

seen the counter-Ars Antiqua seeds it was sowing, allowed one to simply indicate that one

macro-metrical pulse could constitute two equal micro-pulses as well as two unequal or three

equal micro-pulses.

Example 2-1. Perotin, Alleluia, Nativitas Gloriosae (Montpellier Manuscript, ca. early 13th
century)7

6
The author is aware that this is a generalized oversimplification; the point aims to be clearer to those
unfamiliar with the fundamental ramification of Franco’s new notation system.
7
Perotin. (unpublished). Alleluia, Nativitas Gloriosae (Source: Bibliothèque Interuniversitaire, Section
Médecine, Montpellier, France [F-MO H 196], fol. 9v-12r) (Key's Historical and Performance Editions) (J. A. Key,
Ed.). Gainesville, FL: Jordan Alexander Key.

52
Example 2-2. Anonymous Franconian Motet (Montpellier Manuscript, ca. late 13 th century)8

Furthermore, Franco’s system had embedded within itself a hierarchy of metrical values

beyond long and short and the combination of the two. Franco’s system contained such

subdivisions at various levels, allowing one two subdivide a macro-pulse unit (called long) into

either two or three subunits (called breve), and further allowing one to subdivide these subunits

into two or three hypo-subunits (called semibreve). Such hierarchical structuring allowed for

different layers of music to move at relatively different rates (one with shorter, faster rhythmic

values and the other longer, slower values); prior to this, all voices had to move at relatively the

same rate unless acting as an indefinitely long drone. While this notation system allowed for the

clear notating of Ars Antiqua polyphony, it also allowed for the flowering of a new late 13th

century Ars Antiqua genre, now called the Franconian Motet, which had three voices moving at

three somewhat different measured rates though still resembling older genres in its overarching

8
Anonymous. (unpublished). French Double Motet upon an ostinato, ‘Cil s'entremet / Nus hom / Victimae
paschali laudes’ (Source: Bibliothèque Interuniversitaire, Section Médecine, Montpellier, France [F-MO H 196],
fol. 223v-223r) (Key's Historical and Performance Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander
Key.

53
trochaic pulse architecture (see Example 2-2). This late Ars Antiqua style presaged a century of

unprecedented notational and rhythmic development in Western music.

Despite the late Ars Antiqua’s traditionalist trochaic bent, the non-trochaic implications

of Franco’s system would inspire following generations of composers, who would begin writing

music in a style distinct from its predecessors, forming a generational schism that would endure

into the early 14th century, giving us our modern appellations of Ars Antiqua and Ars Nova.9

While Ars Antiqua genres and aesthetics would linger into the mid-14th century, as evidenced by

compositions like the Mass of Tournai, wherein one can find Ars Antiqua and Ars Nova style

compositions juxtaposed (see Example 2-3 and 2-4), this new genre would inevitably spur a

totally new style of music in the 14th century, one rife not only with contrasting musical layers

but emphasizing various trees of rhythmic hierarchies containing both the potential for dyadic

and triadic subdivision at all levels10 and wholly independent of known poetic meters. Even by

the Mass of Tournai in the mid-14th century, one can see the level to which rhythmic

stratification had developed from purely Franconian-style motets in movements like the Kyrie11

9
The divide between the “old art” and “new art” was a generational divide as common today as in the 14th
century, akin to the colloquial dichotomy of “old fuddy-duddies” versus “young hipsters” and common platitudes
and adages like “they don’t make music like they used to.”
10
It is worth recognizing that such rhythmic trees as developed in the late 13th century by the “Franconians”
th
and the 14 century by the Ars Nova generally were not indefinitely sub-divisible as our systems allow today. There
was in Franco’s system a minimum value called the “semibreve,” to which was later added the “minum” and “semi-
minum” and even later the “fusa” and “semifusa.” There was some debate as to how far one could divide a rhythmic
value. Since these systems were not based on fractional subdivisions as today (quarter, eighth, sixteenth, thirty-
second, etc.), there was no clear indefinitude to the subdivision process. Words like “minum” in themselves
suggested a fundamental limit; however, such limits were clearly subverted by the demands of composers, who, by
their musical needs, added smaller genres of notes like the “fusa.” Though with new and smaller genres, the next
possibly smaller genres was not implicit: what the “semifusa” divides into is not implied in the name itself, as is
1 1
implied with a 32nd note, which would divide into whatever number is the result of the product of × (given our
32 𝑥
modern de facto dyadic system, x = 2 and so the subdivision of a 32nd note is a 64th note).
11
For my edition and an accompanying recording of this work click here.

54
and Agnus Dei12 to the “newer” style embodied in the Gloria and Ite Missa Est,13 wherein voices

are moving at starkly differing rates in rather complex rhythmic schema.

While Franco seems to have possibly seen his system as a better method to notate

contemporary music, the possibilities inherent in this new notational system, with some

modification, allowed subsequent composers to notate and reproduce sound structures difficult or

impossible to notate under the older system. The next century, between 1280 and 1380, bears

witness to the flowering of rhythmic notation in Western Europe and the British Isles with the

widespread dissemination of Franco’s Ars Cantus Mensurabilis, which appears in part or full in

nearly every discussion of rhythm following 1300. By the early 14th century, theorists like

Jacques de Liege, in the seventh book of his Speculum musicae, and Robertus de Handlo, in his

Regule, brought the Ars Antiqua developments as seen in the Franconian motet to full fruition,

exemplified in the works of Philippe de Vitry (1291 – 1361) and some early works of Guillaume

de Machaut (1300 – 1377).

The decade of 1320 harkened the end of the transitional Franconian period and the

emergence of the Ars Nova as a distinct musical trend. In the 1320s a number of seminal works

that defined the “new” style were penned, including Johannes de Muris’ Ars novae musicae

around 1320, Philippe de Vitry’s Ars nova notanda around 1322, and Robertus de Handlo’s

Regule in 1326. Handlo’s Regule, like Anonymous of St. Emmeram’s treatise nearly 50 years

before, stands as a culminating work of a generation, being perhaps the most comprehensive

12
For my edition and an accompanying recording of this work click here.
13
For my edition and an accompanying recording of this work click here.

55
Example 2-3. Franconian style “Agnus Dei” from The Mass of Tournai (early 14th century)14

Example 2-4. Ars Nova isorhythmic motet “Se grasse” from The Mass of Tournai (early 14th
century)15

14
Anonymous. (unpublished). Mass of Tournai, Kyrie (Source: Chapitre de la Cathedrale, Tournai,
Belgium [B-Tc], 476 “Mass of Tournai.” fol. 28r.) (Key's Historical and Performance Editions) (J. A. Key, Ed.).
Gainesville, FL: Jordan Alexander Key.

56
work on music from the late Ars Antiqua;16 adding a considerable amount of material to the

Franconian theoretical corpus, Handlo’s treatise was nearly twice the size of Franco’s own

treatise.17 Despite the availability of new ideas circulating in Parisian circles during this decade,

the Regule is quite retrospective, discussing none of the more radical innovations to the

Franconian system from the previous decade that point to the “new art.” Rather, Handlo’s

treatise presents notational developments that modify Franco’s system, but do not expand it.18

Thus, in an age departing from “antiquated” styles, Handlo comprehensively codifies that

notational system which is becoming progressively obsolete or too limited in light of new note

shapes and the rhythmic possibilities they allowed.

If we move forward again another 50 years to 1375, we find ourselves in the heyday of

the Ars Nova. Two years before the death of the venerated forefather of the “new” school,

Guillaume de Machaut, in the midst of Francesco Landini’s most formative compositional years

at the age of 40, and just about five years after the birth of the composer who would become the

arguable pinnacle of the Ars Nova, Johannes Ciconia, English theorist Johannes Hanboys pens

his Summa, which takes as its departure Handlo’s Regule of 1326. One of many treatises

outlining extant and proposing new rhythmic developments for this period, Hanboys’ Summa

directly incorporates the Regule into a rather ambitious and systematic expansion of the mensural

15
Anonymous. (unpublished). Mass of Tournai, Ite Missa Est (Source: Chapitre de la Cathedrale, Tournai,
Belgium [B-Tc], 476 “Mass of Tournai.” fol. 33v.) (Key's Historical and Performance Editions) (J. A. Key, Ed.).
Gainesville, FL: Jordan Alexander Key.
16
The only other extant treatise comparable in its coverage of the late ars antiqua developments in rhythm
and form is the contemporary seventh book of the Speculum musicae of Jacques de Liege.
17
Handlo, R. D., Hanboys, J., & Lefferts, P. M. (1991). Regule and Summa: A new critical text and
translation on facing pages, with an introduction, annotations, and indices verborum and nominum et rerum.
Lincoln, NE: University of Nebraska Press. p. 9.
18
Ibid.

57
system, including an exhaustive survey of note shapes and rests.19 According to Lefferts, nearly

eighty percent of the Summa is “just a straightforward series of descriptions of the nearly

identical properties of eight [note shapes] [of the system],” outlining their numerous slight

variations and the various rhythmic implications of each. While the treatise is prefaced with text

drawn from Franco’s Ars cantus mensurabilis and is closed by a vestigial chapter on the

rhythmic modes of the Ars Antiqua, the work is primarily a monumental compendium of

rhythmic devices allowed through the addition of various note shapes and appended ligatures,

dots, stems, flags, colors, or color redactions now available to composers of the late 14 th century

and many of which were wholly undeveloped in the time of Handlo’s Regule fifty years prior.20

The progressiveness of Summa was clear to its author. Lefferts notes that Hanboys was

“emphatic about his modernity,” showing how Hanboys “speaks of today (hodie), of the new (de

nova modo), of what the ‘moderni’ do and what the ‘antique’ did.”21 Leffert’s assessment of

Hanboys’ Summa is poignant and relevant enough to cite at length: Leffert writes that Hanboys’

notational theories are

…essentially a French ars nova posture, a modernity of the mid-fourteenth century, with
preoccupations that include the extension of note forms by the additions of new figures
instead of signs; the indivisibility of the minima and the concept of gradus [namely, the
hierarchies of rhythmic values]; the importance of the duplex longa and even bigger
values; rests for all these new figures; alterations and imperfection; diminishment by
remote values; puncti perfectionis that are, in effect, puncti additionis; binary
mensuration [as a legitimate and independent rhythmic design from ternary mensural
hierarchies]; mensuration signs [to distinguish the various possibilities between ternary
and binary mensurations]; and syncopation in binary and ternary mensurations.22

19
Ibid., p. 38. According to Lefferts, nearly eighty percent of the Summa is “just a straightforward series of
descriptions of the nearly identical properties of eight notes [of the system].”
20
The resulting presentation is the longest of all the English treatises on notation.
21
Ibid., p. 64.
22
Ibid.

58
Nearly a century since Franco’s composition of Ars Cantus Mensurabilis (whether 1260 or

1280), the system which he originally proposed had unrecognizably burgeoned with new shapes,

colors, augmentations, diminutions, and mensurations without much consensus across the

continent as to what was standard practice beyond the fundamental Franconian premise.

What is important to glean from Leffert’s assessment is not only this proliferation of the

Franconian system and rhythmic notation devices but the equalization of the status of dyadic and

triadic mensural hierarchies, the formulation of various discrete signifiers of these hierarchies

(mensuration signs), and the expansion of these hierarchical trees allowing for even greater

rhythmic stratification within a piece. The equalization of these mensurations, the possibility of

notating all rhythmic complexes in dyadic and triadic rhythmic subdivisions, and the capability

to move fluidly between the two mensurations or have them working concurrently in different

parts opened the possibility of the intimate interplay between dyadic and triadic hierarchies

(what we will herein classify as hexadic rational rhythmic hierarchies – e.g. X/6 time-signatures

and the like) by the last decades of the century. Such proto-pan-rational rhythmic designs would

ultimately find their greatest Medieval fulfillment in the late Ars Nova movement called Ars

subtilior (or “the more subtle art”) of the late 14th and early 15th century.

The Ars Subtilior and the First Experiments in Pan-Rational Rhythmic Notation

Hanboys’ Summa was only the first in many treatises from the late 14th century that

would codify and expand the genres of note shapes available to the composer. The systems of

late Ars Nova and Ars subtilior note shapes became so complex and multifarious between

England, France, Spain, Germany, and Italy23 that one did and still can find it difficult to

23
with England, France, and Italy holding primacy in musical development during this period.

59
understand pieces from one region or manuscript to another, often requiring one to understand a

whole new sets of note shapes or contradictory rules for note shapes shared between two

cultures. There was little consensus in the meaning and varieties of advanced note shapes within

the late 14th century, and, often, composers of rhythmically complex styles like the Ars subtilior

would invent new notes shapes as they needed for new pieces, shapes that theorists might or

might not codify at a later date depending on the ultimate prominence of the shape in question,

which would often be dependent on the ultimate popularity of the composer or works using such

shapes.

This notational complexity, or “subtilitas,” now available to the composer either through

conventionally known or whimsically invented rhythmic symbols, or “figurae,” was engendered

by the continual probing of rhythmic possibilities within the accepted dyadic and triadic

mensurations of the late 14th century. Due to the constantly diverging notational traditions of the

French, Italians, and English, the interpretation of notated music in the late 14 th century could be

“as confusing to the musician of that day as they are to the modern scholar,” as notes Medieval

scholar Philip Schruer.24 Oftentimes, music was as much a concrete art of the ear as a concrete

and abstract art of the eye and intellect; Schruer’s assessment of aesthetics at the end of the 14 th

century is worth quoting at length, for it points us to the presence and problem of “fine

distinctions” as were popular during this period:

Subtlety and intentional obscurity, while having a similar effect in the eyes of the
uninitiated observer, are diametrically opposed in intent, for there is a yawning chasm
between the subtlety of making fine distinctions and the willful obscuring of
interrelationships. The fourteenth century was an era of many fine distinctions.
Scholasticism has reached its zenith, and this unification of language, ideas, and
terminology brought about by a common university education allowed a shared rigor to
be applied to all seven of the liberal arts. Detailed inquiries through the art of logic into

24
Schreur, Philip, trans. Tractatus Figurarum = Treatise on Noteshapes: A New Critical Text and
Translation. Edited by Thomas Mathiesen. Greek and Latin Music Theory. Lincoln: Univ. of Nebraska Press, 1989.
xi

60
 the perception of universals, or intentio secunda, were matched by increasing
sophistication in the definition of musical concepts through the universal principles of
mathematics.25

Via the work of mathematicians like Leonardo of Pisa in the 13 th century and the general

reemergence in Western Europe of mathematics as a serious study beyond mere commercial

arithmetic, as well as the introduction of Hindu-Arabic numerals to the West via the

aforementioned mathematician’s work, the continent slowly emerged from the Dark Ages with

mathematics as a renewed tool that seriously impacted the arts between the late 13th and late 15th

century.

Not since Boethius had mathematics been so aptly applied to music to invent new modes

of expression with the art. For nearly a millennium the intervals of the modes as well as the

consonances and dissonances those intervals created when concurrently combined had been

understood mathematically, as evidenced in nearly every musical treatise in Europe since

Boethius. However, rhythm had never seriously been understood or employed in a mathematical

fashion until the Ars Nova, by which time mathematics was emergent enough to offer the tools to

better articulate rhythm abstractly rather than only concretely through paralleling various

metrical structures in language as heard in poetry. While the Ars subtilior was viewed by critical

contemporaries and has been assessed by some modern scholars as a “misshapen flower

blooming from the muddied and decaying quagmire of Medieval thought before the refreshing

and renewing vigor of the Renaissance,” the music itself being “inconsistent and willfully

complex,”26 it needs to be understood that this music was “born of a time alive with the vigorous

25
Ibid., 1
26
Willi Apel, The Notation of Polyphonic Music: 900-1600, 5th ed. (Cambridge, Massachusetts: The
Medieval Academy of America, 1953), p. 403.

61
exploration of rhythmic and harmonic possibilities,” much of which was based on “reasoned

extensions of traditional mathematical and musical principles.” 27 Not but a century prior to the

beginning of this complexification of rhythmic design in the mid-14th century, Thomas Aquinas

proclaimed, “Logicus et mathematicus considerant tantus res secundum principia formalia.” 28

Moreover, it was such principia formalia, elucidated to Europe generally by Leonardo of Pisa

and to musicians particularly by Johnnes de Muris in his Notitia artis musicae,29 and the

development and implementation of them by 14th century music theorists and composers

respectively that led to the “more subtle art.”

However, it seems as though mathematics was still in its nascency among most

composers and music theorists, who could only imagine, in retrospect, the most trivial

mathematical tools to represent rhythm. While obscurity is not inherently the aim of

mathematics, obscuritas was one aim of medieval religious exegesis and scholasticism, and so

the clarity of mathematics and the obfuscation of scholasticism intermingled in music of the Ars

subtilior. While at once there is a new and clearer method to demarcate various macro rhythmic

structures, there is a proliferation of seemingly arbitrary and inconsistent symbols to obscure, on

a micro level, the fundamental clarity inherent in the basic system.

One surviving treatise wherein the complexity of rhythmic demarcation during the late

Ars Nova and Ars subtilior comes into full relief and which attempts to remedy the problematic

obfuscation of the late Ars Nova notational system is the anonymously composed Tractatus

27
Schreur, Tractatus Figurarum. 1-2
28
“The logician and the mathematicians view things only according to their formal principles” (St. Thomas
Aquinas, “De potential,” ed. R. P. Pauli [and] M. Pession, in Questiones disputatae, 2 vols., 8th ed. by P. Bazzi
[Taurini: Marietti, 1949], 2:160b; quoted in ibid., 2:59-60).
29
See Johannis de Muris Notitia artis musicae et compendium musicae practicae, Petrus [sic] de Sancto
Dionysio Tractatus de musica, ed. Ulrich Michels, Corpus scriptorum de musica, no. 17 (n.p.: American Institute of
Musicology, 1972).

62
Figurarum.30 Tractatus proposed a notational system combining elements of the French and

Italian systems, attempting to create a standard set of figurae that would also allow for the

simultaneous combination of any two of the four prolations of the French mensural system. 31

The growing rift between the French and Italian systems and the regular invention of new note

shapes during this period was the primary problem that this treatise’s proposed synthesis

attempted to address. For an example of this mounting problem of complexification, consider the

Codex Chantilly, one of the major sources for the Ars subtilior tradition, wherein there are at

least 16 special note shapes used among a variety of compositions beside the standard note

shapes of the Ars Nova; furthermore, these 16 do not include some exceptional shapes found

only in single compositions from this source.32 Neither is this list of 16 shapes comprehensive

beyond Chantilly, for even more shape varieties exist in sources across the continent from this

period, some found in compositions, some in theory, some French, some Italian, some English,

some possessing unique qualities, and some redundant to others in differing sources. Compared

to the basic Franconian system of a century prior, the overwhelming complexity and obscurity

engendered by the “more subtle” movement of rhythmic diversity in the late 14 th century was

and still is a serious problem for interpretation, for while there was consistency in the derivation

of some shapes, there was also inconsistency, redundancy, and arbitrariness concurrently.

30
The authorship of this treatise has been under great debate among scholars over the past half century.
While historically and popularly this text has been attributed to Philippus de Caserta (c. 1350 – c. 1420), this is now
widely contested, and so most recent scholarship admits that there is no good attribution for this treatise. Thus,
herein we consider it anonymously composed, though it was likely authored sometime in the last decades of the 14th
century during the height of the Ars subtilior since the text deals to a great extent with note shapes from that style
and references works from that period, particularly from the Codex Chantilly, which is known for its abundantly
“subtle” aesthetic.
31
Schreur, Tractatus Figurarum. p. xi
32
Ibid., p. 150.

63
 Such problems of obfuscation, clearly evident to at least some theorists, were also not lost

on composers and performers of the period, who were naturally the most effected by the steadily

growing complexity of disparate yet intermingling systems between the French, Italian, and

English schools. One poignant example comes from one Ars subtilior composer himself, Guido,

who, in one of his two extant compositions, decries the cumbersomeness of the system with

which (ironically) he is writing. In Guido’s unicum ballade “Or voit tout en aventure,” attributed

mononymously in the Chantilly Codex, he sets the below text, which bemoans the “new figures”

(figurae) that “displease everyone” and require one to “compose against nature” and “destroy

that which was well made,” praising old masters like de Vitry and criticizing contemporary

composers like Marchetto “the imposter,” whose music “has no measure,” “never allows for

perfection,” “is too presumptuous,” “drags good manner,” is an “outrage,” and is simply “not

well done.”

Now everything is left to chance,

Because it is thus necessary for me
To compose with the new figures
Which displease everyone.
It is completely opposite to the good art
Which is perfect, certainly this is not done well.

We compose against nature

And thereby destroy that which is well made,
For which Philippe [de Vitry],
Who is no longer alive gave us good example.
We abandon all his works
In favor of Marchetto the imposter.
Certainly, it is not well done.

The art of Marchetto has no measure

And never allows for perfection,
It is too presumptuous
To follow and to design
There figures and to drag good manner
Out of this outrage.

64
 Certainly, it is not well done.33

As suggested before, the irony of this song, only available to those who can see the manuscript

music (for the modern notation of this piece is rather straightforward and the resultant sound,

whether Medieval or modern, is not that bizarre relative to music of the Ars subtilior) is that the

song is written, perhaps unnecessarily, using the very note shapes about which the song’s text

complains. Clearly, while some composers reveled in the various possibilities of a system that

allowed capricious invention, some were not as compelled.34

33
First strophe of the ballade Or voit tout en aventure attributed in Chantilly, Bibliotheque du Chateau, Ms.
564 to “Guido.” Translated by Kees Boeke and Jill Feldman in Boeke, Kees and Jill Feldman. “Or voit tout en
aventure.” Text translations in accompanying booklet, Codex Chantilly II: Tetraktys. 2008-2011. Olive Music KTC
1905, 2011. Compact disc.
34
Another poignant example of late 14th century music complaining about the Ars subtilior includes the
anonymous “Tout clarte.” The reference therein is much subtler than Guido’s criticism and couched in much
metaphor: the poetic speaker describes a world of reversals in which everything clear is obscure and all joy is
sadness. This is because he is unable to see “the figure,” meaning both the lady whom he loves but also the written
note shapes that the composer uses to write his music down. One further example of composer critique of the
present state of music is Francesco Landini’s three-texted madrigal “Musica Son - Già Furon - Ciascun Vuol,“ the
text and translation of which is provided below. While Landini’s criticism is less clearly directed towards composers
of the Ars subtilior, he is criticizing the nobility for becoming corrupted by music that is not “down to earth.”
During this time, many courts and papal chapels were hiring and commissioning composers of the Ars subtilior, so it
is possible that this “corrupted,” “imperfect,” and unsweet music that Landini “laments” needs “to come down to
earth” is, in fact, the “popular” songs of the more subtle art of “arrange[d] musical notes.” For more information see
Key, Jordan Alexander. (2017, May 25). “A Medieval Composer's Thoughts on Art & Pop Culture: ‘Popular’ (?)
Music from 14th Century Italy: Part 3 - Francesco Landini (c. 1325/35 - 1397), Madrigal, ‘Musica Son - Già Furon -
Ciascun Vuol.’”

CANTUS: Musica son che me dolgo piangendo / Veder gli effecti mie dolcie profecti / Lasciar per frottol' i
vagh' intellecti. / Perche 'ngnorantia 'n vici ongnun' chostuma / Lasciasi 'l buon' e pigliasi la schiuma.

I am Music, who weeping regret to see / intelligent people desert my sweet / and perfect effects for popular
songs; / because ignorance and vice abound / good is deserted and the worst is seized.

CONTRATENOR: Gia' furon le dolcecce mie pregiate / Da chavalier', baroni e gran singnori, / Or sono 'n
bastardita 'n gienti chori. / Ma i' musicha sol non mi lament / Ch'ancor l' altre virtu lascia te sento.

Formerly my sweetnesses were prized / by knights, barons, and great lords. / Now gentle hearts are
corrupted. / But I, Music, do not lament alone, / for I see even the other virtues deserted.

TENOR: Ciascun vouli narrar musical note / Et compor madria, chaccie, ballate, / Tenend' ongnun' in la
su autenti charte. / Chi vuol d'una virtù venire in loda / Conviengli prima giugner' alla proda.

Everyone wants to arrange musical notes, / compose madrigals, caccias, and ballates, / each holding his
own to be perfect. / He who would be praised for a virtue / must first come down to earth.

65
 However, when composers found their new rhythmic freedom compelling and could

fluently speak in this complex system, works of stunning rhythmic complexity, nearly

unparalleled until the 20th century, could emerge. One example that demonstrates the capabilities

of these new note shapes is “Se Genevre” by Johannes Cuvelier. Just in the first fourth of the

work, given in Example 2-5, one can witness a drastic change from the early Ars Nova

compositions given previously. Now, mensuration signs along with new note shapes, stems,

flags, ligatures, coloration, and particoloration allow for the simultaneous interplay of triadic and

dyadic rhythmic hierarchies (poly-meter, poly-rhythm, and poly-mensuralism), branching

hexadic rhythmic hierarchies (tuplets and nested tuplets with powers of three and two), various

partitioning and re-partitioning of fundamental time units (regular changes in the macro pulse by

perfection or imperfection), as well as the precise and nuanced use of rests as part of the

rhythmic structuring.

With the cultivation of “subtle” expectations in the stylistic interpretation of music from

this period, composers of the late 14th and early 15th century began imagining unique and

exceptional rhythmic complexes and could also begin employing extra-musical devices, not

necessarily allowable by the notational system, to accomplish such extraordinary ends. 35 It is

35
By the cultivation of stylistic expectations in musical interpretation, I mean to suggest that in any period
or genre of music there is a collective understanding of “style,” such that, given instructions to “perform music,” the
performer understands what is meant generally by “music.” For example, if one commanded Johann Sebastian Bach
to perform music and he was inclined to do his best to fulfill this request, he would perform music akin to what we
would call “baroque,” and not spontaneously invent Jazz or Rock and Roll, or play distantly past musical styles like
Medieval organum. A professional court or chapel composer from the late 14th century, given some instruction to
play music, would likely play something familiar sounding to them and their audience. Furthermore, given
directions from a known composer, the performer, being familiar with the style of the composer, would likely
perform something akin to the aesthetic of that composer. For example, if Pierre Boulez wrote a piece of music
demanding a performer more or less improvise some music, the performer would likely improvise something
aesthetically akin to the music of Pierre Boulez and not improvise in the style of Mozart or Bob Dylan (though of
course anything is possible). Thus, should a composer of the Ars subtilior give extra-musical instructions to generate
a musical piece, the performers would likely attempt to create a music akin to the Ars subtilior, or at least the Ars
Nova generally, and would not create something more aesthetically aligned to Perotin two centuries prior.

66
Example 2-5. Opening portion of Cuvelier’s Se Genevre from the Chantilly Codex.61

61
Cuvelier, Johannes. (2020). Se Genevre (Chantilly, Bibliotheque du Chateau, MS. 564, f. 41v) (Key's Historical and Performance Editions) (J. A. Key,
Ed.). Gainesville, FL: Jordan Alexander Key.

67
during this period that the first musical “canons” 62 begin to appear in sources, which further

allowed composers to communicate complex musical structures to performers when notation

could not easily communicate an idea.

As an example of this practice, let us examine one of the most rhythmically distinctive

pieces that survives from the late 14th century Ars Nova and Ars Subtilior: Johannes Ciconia’s63

Le ray au soleyl,64 which employed canon to a rather unique end. The canon Le ray au soleyl is a

striking unicum work, singular in style and form within Ciconia’s oeuvre. In the Lucca

Manuscript only one musical part is notated; however, Ciconia gives a Latin inscription at the

end of the melody, indicating that all three voices should be derived from this single line by

62
By “canon” I mean the fundamental definition of the word: law. While canon denotes very particular
kinds of musical “laws” in modern usage as a musical form, this word originally denoted any musical law, which
once established and followed, allowed the performer to generate the piece in full from some given musical germ.
While an anachronistic analogy, musical canons were, in essence, the first algorithmic music. Over time, particular
algorithms simply became more popular and common than others and so became the de facto “law” when no
specific law was given.
63
Johannes Ciconia (c. 1370 – 1412) was born in Liège sometime around 1370. Most of his biography is
uncertain and much is still being discovered with current archival research by musicologists such as Jason Stoessel,
John Nádas, Agostino Ziino, and Bonnie Blackburn, along with many others. Ciconia was by training likely a
Franco-Flemish composer of the Burgundian Kingdoms but was most compositionally fruitful during his adulthood
in Italy. Compared to his contemporaries, more music by him survives, all of which demonstrates a remarkable
stylistic variety and virtuosity. His relatively ample output which as survived the ages along with his comprehensive
understanding of composition at the end of the 14th century has solidified his position as one of the most prominent
composers along with Guillaume DuFay and Antione Binchois during the transition from late medieval Trecento
and Ars Nova styles into what musicologists identify as early Italian Renaissance polyphony. He wrote 11 Mass
sections (mostly Glorias and Credos), 11 sacred motets, and 20 French and Italian secular pieces, including French
virelais, Italian ballata, and Italian-styled madrigals. Of his 11 motets, four are in the French isorhythmic style. The
stylistic influence of Philippe de Vitry (1291 – 1361), Guillaume Machaut (1300 - 1370), and composers of the Ars
Subtilior, particularly Philippus de Caserta (ff. mid to late 14th century), is clear throughout his ouvure.
64
Most of Ciconia’s music, particularly from his Paduan and Pvian periods, survives in the singular
fragmentary Lucca manuscript (see Nádas and Ziino), probably transcribed in Padua during Cicona’s latter life. New
leaves belonging to the Mancini and Lucca Codex strengthen the attribution of unascribed pieces, Ciconia’s
authorship of these pieces having previously been treated as dubious (see Bent and Hallmark, eds., PMFC, xxiv,
1985). Among these newly confirmed pieces are his Ave vergene, Chi vole amar, Gli atti col dançar, Le ray au
soleyl and Poy che morir. See Nádas, John L., and Agostino Ziino, eds. The Lucca Codex: Codice Mancini: Lucca,
Archivio Di Stato, Ms. 184, Perugia, Biblioteca Comunale "Augusta" Ms. 3065. Lucca: Libreria Musicale Italiana,
1990. Also see Stoessel, Jason. "Con Lagreme Bagnandome El Viso: Mourning and Music in Late Medieval Padua."
Plainsong and Medieval Music 24, no. 01 (2015): 71-89.

68
rhythmically manipulating and superimposing the melody with itself. Below, the melody from

folio LXXXIII is excerpted (see Example 2-6). The Latin text of the riddle reads:

Canon: Dum tria percurris quatuor va[let].

Tertius unum subque diapa[son]
sed facit alba moras.”65

Which translates as:

Law: While three traversed four counts,

the third, one - an octave below
but it makes white delays.

While such a composition is immensely clever to devise under any system of

counterpoint, the rarified and intellectual nature of this musical process has continuously

obscured its understanding. There is still debate as to how one should properly interpret the

riddle and construct the canon. Various seemingly correct solutions have yielded completely

different pieces. At least three are commonly accepted in critical editions of the Mancini/Lucca

Codex. While each differs slightly in its solution, all agree that this piece is a prolation canon. 66

Fascinatingly, the durational proportions between the three voices in Ciconia’s prolation canon

are [Link], requiring continuous use of indivisible poly-mensuralism (dyadic and triadic

subdivisions of a FTU within the same order: 1:4 and 1:3; see Example 2-7).

65
A few of the missing letters that seem to have been cut off by the books binding have been supplied
through inference. See Newes, Virginia. "Mensural Virtuosity in Non-Fugal Canons." In Canons and Canonic
Techniques, 14th - 16th Centuries: Theory, Practice, and Reception History; Proceedings of the International
Conference, Leuven, 4-6 October 2005, edited by Katelijne Schiltz and Bonnie J. Blackburn, 19-46. Leuven:
Peeters, 2007. 19-46.
66
A prolation canon (also known as a mensuration canon or proportional canon, though each can suggest
something slightly different) is a subgenre of temporal-canon, in which the main melody is accompanied by mimics
of itself usually begun simultaneously but performed at different rates. For example, the main melody’s
accompanying mimics would be sung at double and half the temporal rate as the original, all beginning at the same
time rather than at staggered moments.

69
Example 2-6. Johannes Ciconia, Le ray au soleyl67 with canonic inscription.

Furthermore, Ciconia seems to have constructed this singular melody in such a way that when

sung in these proportions, there are few rhythmic coincidences between the voices throughout

the piece (see Example 2-8 for a rhythmic mapping of the whole work wherein whole barlines

indicate a full 3-voiced rhythmic coincidence). While Ciconia could perhaps notate this out fully,

he could not have notated this using the same economy as in the Lucca Codex, and the common

conventions of mensuration during this period would not have allowed for such a construction

conventionally.68

This style of piece (a continuous use of simultaneously indivisible mensurations, e.g.

[Link] against 1:3), as far as extant sources show, is unique until the early 16 th century.69 Newel

describes the prevalent tendency towards musical clarity during the early fifteenth century,

67
Biblioteca Comunale Augusta [I-PEc], MS 3065 “Mancini Codex/Lucca Codex,” no. 9 fol. LXXXIII.
68
Such musico-mathematical thinking is clearly consequent of the mathematization of rhythm over the
century leading to Ciconia’s flourishing during the late Ars Nova.
69
Here, we are referring to the prolation and mensuration canons of Pierre de la Rue and Johannes Mittner.

70
Example 2-7: Johannes Ciconia’s Le ray au soleyl from the Lucca Codex.70

70
Ciconia, Johannes. (2018). Le ray au soleyl (Mancini Codex/Lucca Codex, I-PEc MS 3065 no. 9 fol. LXXXIIIr) (Key's Historical and Performance
Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

71
Example 2-8: Rhythmic mapping of Ciconia’s Le ray au soleyl from the Lucca Codex.71

71
Key, J. (2016, September 24). Mathematical Medieval Music: "Popular" (?) Music from 14th Century Italy: Part 1 - Johannes Ciconia (c. 1370 -
1412), Canon "Le ray au soleyl" [Web log post]. Retrieved from [Link]
14th-Century-Italy-Part-1---Johannes-Ciconia-c-1370---1412-Canon-Le-ray-au-soleyl-Source-Mancini-CodexLucca-Codex-I-PEc-MS-3065-no-9-fol-LXXXIII.

72
which gradually supplanted the rhythmic, melodic, and harmonic complexities of the Ars

subtilior. This tendency towards rhythmic simplification is reflected in the proportion canons

that have survived from this period. Newel observes that “in none of them do we find ratios

between the given and canonic voices, such as the 4:3 of [Le ray au soleyl], that produce

irreconcilably conflicting rhythms at the minim level. Most belong to the simple augmentation

type, employing the ratio 1:2,”72 such as that found in DuFay’s Bien veignes vous, amoureuse

liesse. What seems to have allowed Ciconia the ability to semiologically articulate such a

musical structure is not note shapes only but the insight to append the extant system not with

new signs but with simple generative directives.

A final work we should examine from the Ars subtilior will deliver to us the most

obfuscating and extensive use of the note-shape system of this period: the duo Angelorum psalat

tripudium by the mononymous Rodericus.73 This work is so obfuscating, in fact, that modern

scholars have yet to come to a definitive interpretation of the work,74 which itself contains many

invented note shapes appearing nowhere else in the extant literature. On its surface, the work

seems to be an exploration (perhaps pedagogic and catalogic) of new note forms, containing

many shapes found in Guido’s small oeuvre as well as the works of Senleches. Just considering

72
Newes, Virginia. "Mensural Virtuosity," 27.
73
The unicum source for this piece names the composer as “S Uciredor,” which is clearly a reversed
spelling of “Rodericus;” the suggestion of reversal here comes from the text of the music itself. By all studied
accounts, no composer by that name appears in musical records; however, a tentatively proposed identity is
Johannes Rogerii de Watignies, who was born in Cambrai and is traceable from 1378 as a canon in Reims, a
member of the chapel of Burgundian Duke Philip the Bold in 1391, and in the service of Pope Benedict XIII in
Avignon. For a full account of the interpretation history see Young, Crawford. (2008). Antiphone of the Angels:
'angelorum psalat tripudium'. Recercare, 20(1-2), 5-23.
74
The modern history of “Angelorum psalat tripudium” is a story in itself; for more information on the
modern history of this work’s interpretation see Young, C. (2008). Antiphone of the Angels; Stoessel, Jason. (1999).
Symbolic Innovation: The Notation of Jacob de Senleches. Acta Musicologica, 71(2), 136-164; Josephson, Nors S.
“Rodericus, ‘Angelorum Psalat’”, Musica disciplina, XXI, 1971, pp. 113-126; and French secular music.
Manuscript Chantilly, Musee Conde 564, ed. Gordon K. Greene, texts edited by Terence Scully, 2 vols., Les
Pemparts – Monaco, L’oiseau-lyre, 1982 (Polyphonic music of the fourteenth century, 18-19), vol. II, p. 93.

73
the superius voice, one finds an overwhelming potpourri of note colors and shapes within the

mensuration perfectum maior (see Example 2-9 below for a modern transcription of its opening).

Consider the categorized list of note shapes in Figure 2-1 below.

Black note shapes:

• maxima, breve
• semibreve
• semibreve caudata
• minim
• dragma
• flagged dragma
• virgulated semibreve caudata.
Red note shapes:
• red breve
• red semibreve
• red semibreve with dot of perfection
• red minima
• red virgulated semibreve caudata
• red virgulated fusa.
Hollow note shapes:
• hollow-black dragmas
• hollow-red semibreves
• hollow-red minims (always preceded by a reverse-C sign)
Particolored note shapes:
• particolored breves
• particolored semibreves
• particolored semibreve ligatures75

Figure 2-1. Notes shapes in the superius voice of Rodericus’ “Angelorum psalat tripudium”

The work itself is a poetic intermingling of mathematics, religious exegesis, and social

commentary; the text of Angelorum is a metaphorical treatment of the meaning of opposites via

75
Three note shapes, black and red virgulated semibreves caudatae and red fusa virgulata, occur uniquely
in this composition. See Young, C. (2008). Antiphone of the Angels.

74
Example 2-9. The opening of Rodericus’ (S Uciredor’s) “Angelorum psalat tripudium.”76

76
Rodericus. (unpublished). Angelorum psalat tripudium (Chantilly, Bibliotheque du Chateau, MS. 564, f. 48v) (Key's Historical and Performance
Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

75
Example 2-10. Rodericus’ “Angelorum psalat tripudium” from Chantilly Codex (f. 48v).77

77
Rodericus. Angelorum psalat tripudium. Chantilly, Bibliotheque du Chateau, MS. 564, f. 48v.

76
poetics and math, through the interplay of musical (e.g. note shapes, colors, and values and their

consequent formal and rhythmic proportions) and poetic (metaphorical and literary allusion)

‘perfection’ and ‘imperfection.’ Young explains that Angelorum “tells a story centered around

perfection becoming imperfection, referencing the Fall of the Rebel Angel”78 with the symbols

of perfection and imperfection reflected in the numbers 3 and 2, respectively.79 Young elaborates

that the proportion of 3:2 “provides the keystone of the piece on multi-levels: different levels of

rhythmic organization, harmonic organization, and symbolically, spiritual-moral organization.”80

However, it is possible, like in Guido’s Or voit tout en aventure, that this work does not

take itself as seriously as its apparent sophistication (or “presumptuousness,” as Guido might

critique) would lead us to believe. Young notes that this work might not be pedagogic or

catalogic as many previous scholars have believed81 but is rather a “tongue-in-cheek comment on

the extreme complexity of advanced ars subtilior music, ribbing those… who take it very

seriously.”82 The capricious invention of note shapes in this work and the perhaps overwhelming

78
Young, Crawford. “A Collector’s Court.” Essay in accompanying booklet, Corps Femenin: L’avant-
garde de Jean Duc de Berry. 2010. Media SRL 551, 2011. Compact disc. 12.
79
During this time, the number 3 was often considered symbolically representative of the Christian Holy
Trinity. Consequently, music also adopted threes as symbolic of perfection or the divine (albeit the tritone); for
example, rhythmic groupings of threes were called “perfections” and mensurations of triadic hierarchies were called
“perfect.” To contrast this, groupings of twos in rhythm were considered “imperfect,” since they did not possess the
completeness of three. Thus, the juxtaposition of things involving two and three in music can often be interpreted as
symbolic of the earthly and divine, the corrupt and the pure, Satan and God, or any such classical dichotomy.
80
Ibid.
81
Some have even thought this work might be a piece of “initiation” for young composers into the school
of the Ars subtilior composers: if one can unravel, with satisfaction, the notation and symbolism of this work, then
they are worthy to be considered a master composer of the “more subtle art.” This hypothesis, however, has no real
substantiation, though it is entertaining to consider the anxiety such a test would engender in a young initiate when a
perfect solution might very well be impossible or impossibly obfuscated (and likely intentionally so).
82
Young, C. (2008). Antiphone of the Angels. p. 13. Even the opening sonority and the final cadence (a
perfect twelfth) are representative of 3:2 proportionality, a 12th being a perfect consonance with the ratio 3:2, which
is unique as a final cadence in any two-part work from this period.

77
incorporation of so many note shapes are perhaps an attempt to be absurd and show the problems

of such a whimsical practice that is “left to chance,” as Guido would claim. Indeed, modern

attempts at interpreting this work have been left to some chance, scholars of this period not

knowing exactly what some shapes mean and not being sure that Rodericus is even using known

note shapes in their conventional fashion.

Conclusions to Chapter 2

These three works given above embody the fullest implications of Franconian Notation

and its development throughout the 14th century during the Ars Nova and Subtilior.83 But what

are we to make of the numerous problems encountered through this brief survey of late Medieval

rhythmic exceptionalism that disappeared as suddenly and subtly as it had emerged? By the end

of the Middle Ages and the first decades of the 15th century, musical taste shifted away from

obfuscation; the stunning rhythmic advances of the Ars subtilior were forsaken before

contemporary theorists and historians like Johannes de Muris or Jacques de Liège could fully

document such developments. Rather we are left only with pamphlets and incomplete guides in

comparison to the comprehensive tome of Anonymous of St. Emmeram. While Hanboys’

Summa provides us with a comprehensive overview of Ars Nova development up to about 1375,

the 40 years following, during which the Ars subtilior truly flourished, begets no surviving

comparably comprehensive textbook for the developments of that nearly half century of

experimentation. All we are left with are relative trifles and a few short practical guides, such as

the Tractatus figuararum.

83
Other poignant examples of exceptional rhythmic practice within the Ars subtilior that demonstrate early
use of pan-rational rhythmic structures and the problems inherent in Ars Nova notation I recommend exploring
include: Goscalch, ballade “en nul estat;” Caserta, ballade “Du val prilleus;” Suzoy, ballade “Pictagoras Jabol;”
Johannes Ciconia, virelais “su un’ Fontayne;” Matheus de Perusio, ballade “le greygnour bien;” Zachara de Teramo,
“sumite karissimi;” and Senleches, virelai “la harpe de melodie.”

78
 Furthermore, the system proposed by the Tractatus figurarum was never actually

accepted into the mainstream of 14th century theory or composition, evidenced by its new note

shapes’ nonappearance in any surviving music. So even such practical guides were not as

practical as their authors might have hoped for. Given little contemporary consensus over the

advanced notations of the Ars subtilior, even the various guides available to use are mostly

useless to interpret the subtler works from this period. However, despite its lack of practical use,

Tractatus figurarum’s interest to musicians of the late 14th century is suggested by the number of

surviving copies.84 While ultimately theoretical, the principles used in the creation of the

Tractatus figurarum reveal much about notational theory at the end of the 14th century, which

can be important to consider when assessing the limits of our present systems of rhythmic

notation.

While their note shapes were never employed in a French-Italian Ars subtilior notational

synthesis, the concerns expressed by the author of the Tractatus, evidenced in the three works

above, reflects the subtly dire situation within the ever-mounting notational subtlety in the

surviving works of the Ars subtilior style. Likely recognizing an untenable expansion of a system

and a growing confusion, if not frustration, among interpreters of these late 14 th century works,

Tractatus hoped to offer a remedy to this situation by providing a new simplification of an

overcomplex system just as Franco of Cologne had offered to the Ars Antiqua a century before.

It is meaningful to entertain the possibility that had such a standardization as Tractatus

endeavored to craft occurred, the history of musical notation in the West might have been

significantly different. It was likely due to such a gross lack of standardization by the beginning

of the 15th century that the Ars Nova/subtilior system collapsed under its own weight of

84
Schreur, Tractatus Figurarum. p. 3

79
cumbersome interweaving and contradictory semiotics to return, by the beginning of the Musical

Renaissance (sometime between 1420-1440) to a system very akin to Hanboys’ expansion of

Franconian Notation before the Ars subtilior. Had some sense of cohesiveness and consistency

been forced upon the “subtle art,” then it is possible that many aspects of this system would have

carried forward, even with a general change of aesthetics away from Scholasticism and

obscuritas and towards Humanism and claritas during the Renaissance. However, this

standardization did not happen, and so there was no similar relatively wide-spread popularity of

equally “subtle music” as during the late 14 th century until the 20th century.

Indeed, medieval musicologist Jason Stoessel is apt to note that “prior to the twentieth

century, no greater diversity of notational [rhythmic] signs are found… than in the surviving

versions of works by Jacob de Senleches, his contemporaries, and his immediate successors.”85

The use of special and newly invented figures in Senleches En attendant Esperance and La

harpe de melodie86 is rivaled consistently only by Guido and is exceeded only by Rodericus’

Angelorum psalat. However, despite their sophistication and renown by the beginning of the 15th

century, composers like Senleches, Guido, Ciconia and others were totally forgotten half a

century after their death, their style and methods of communication becoming utterly opaque to

subsequent generations, who could no longer read many of their forbearer’s strange signs.

Broadly speaking, the generation following the Ars subtilior composers liberally pruned what the

late Ars Nova had developed, keeping only the most common and necessary parts as were

already in wide use by the death of Machaut in 1377. While some of the note shapes and names

changed, while a few special characters were added or simplified, most of what the Ars subtilior

85
Stoessel, Jason. Symbolic Innovation. 136-137.
86
For a complete pedagogical edition of this work with the manuscript sources click here.

80
had invented disappeared from music until it would be reinvented in different fashions in the late

Renaissance and again in the early 20th century.

What can we learn from this “collapse” or abandonment of a sophisticated notation

system? In attempting to reassess our own methods of rhythmic diversification in the early 21 st

century, what could we learn from the Ars subtilior about our own route forward in the polishing,

clarifying, and systemizing of our own subtle rhythmic arts? Ultimately the Ars subtilior offers

us some broad conclusions to consider in crafting any musical notational system. Let us conclude

our survey of the earliest rhythmic experimentations by establishing these conclusions.

81
 Synthesis I

From the historical observations made in Chapter 2, we can distill some rudimentary

principals upon which we can begin constructing a theory of notation systems. From this theory,

we can ultimately come to a logical (and perhaps longer lasting) expansion and/or revision of our

current system of musical notation to the end of rhythmic diversification.

Definitions

DEFINITION S1.01A: A notational system (henceforth called a “system”) is understood to

be a limited collection of abstract semiotic (potentially) generative elements, which under
a defined generative process generate other derivative elements.

DEFINITION S1.01B: If a system’s elements, beyond the limited generative set, is un-
systematically produced (namely, without some generative process), the system is
unsystematic. It is understood that if a system becomes unsystematic, then it is no longer
a system and is rather a semiotic “collection,” which is unsustainable (see Axiom S1.01).

DEFINITION S1.02: An element (generative or derivative) is a pointer to a concrete

phenomenon (called an “event”), which the system stores in abstraction to be later
interpreted and translated back into concrete form by a processor.

DEFINITION S1.03: A processor is the interpreter of a system. All processors have a limit
to their processing power, which is a processor’s ability to interpret and translate a
system.

DEFINITION S1.04: By “interpret and translate” a system, we mean the method by which a
processor reads the system (the generative and derivative elements along with the
generative processes), matches the system to its events, and then translates the given
system from abstract semiotics back into their corresponding concrete phenomena.

From these basic definitions, we can establish an axiom and lemma, by which we may

begin to make meaningful observations about the nature of any abstract system of notation

(musical or otherwise). First, let us establish the axioms and lemmas, from which the

observations will follow.

82
Axioms and Lemmas

AXIOM S1.01: If a system becomes unsystematic (namely a “collection”), it will

inevitably collapse or contract to a previous point of systematization.

LEMMA S1.01: Since derivative elements within a system of generative elements are
logically derived via some generative process applied to the original limited set of
generative elements, the additional derivate elements conform to the system, being
systematically generated from the system itself.

Observations

OBSERVATION S1.01: Note that while generally there need be no particular upper bound
on the number of generative elements or their derivatives in the set, it should be
recognized that in “practical circumstances,” (namely, systems applied to unaided human
processing) a relatively small number of generative elements should be employed for any
system so that the system remains reasonably coherent.

OBSERVATION S1.02: Some derivative elements will become trivial given a process that
allows infinite derivation. Though we need not enumerate such derivations ad infinitium,
and while nearly all such infinite derivations will be inapplicable to human-processed
systems, we may still recognize the potentiality and existence of such trivial elements.

Based on these observations of our rudimentary theory of notation, we can note several

important conclusions, which will have bearing on our ultimate “meta-synthesis” and crafting of

a new/emended system of musical notation.

Conclusions

CONCLUSION S1.01: The success of a system is dependent on the set of generative

elements and the process by which derivative elements are generated. Such generative
and derivative elements

a. need to have some logical and consistent origin for ease of interpretation and
translation, and consequently

b. cannot be based on seemingly arbitrary and endlessly proliferating forms lest the
processor fail to interpret and translate due to a dwindling processing capability.

CONCLUSION S1.02: If one wishes to append a new generative element to an extant

system, such a new element needs to either

83
 a. be derivative and follow, via some process, from the elements already present (as
per our first conclusion), or

b. be justifiably and wholly different from all other forms, not generatable from
some process and semiologically distinguishable, being itself uniquely
distinguished by its intended interpretation and ultimate concrete translation.

CONCLUSION S1.03: Recognizing the limited nature of human perception as a discrete

processor,87 as noted in Observation S1.01, the elements of the system must not only be
logically derivative from each other but must also have some reasonable limit to variation
or invention, disregarding trivial ad infinitium derivation as in Observation S1.02.

CONCLUSION S1.04: A system needs to be limited (as in Conclusion S1.03) but flexible,
able to be fluently interpreted broadly while also capable of communicating nuance when
and where needed. If a system is too limited and allows for no flexibility, flexibility could
become unsystematically grafted onto the system when the system has excluded or
cannot allow for some unforeseen necessary element. Such a grafting will possibly cause
the system to become unsystematic, which, by Axiom S1.01, will cause the collapse of
the system or the ultimate redaction of such an emendation to re-stabilize the system.88
Thus, a system must either be amendable within reasonable limits by some logical
process, or, better yet, have the inbuilt capability to carry multiple
interpretations/translations per pointer, which, given some limiting context, cannot be
misinterpreted.89

87
Here, we do not consider the notation of music for computerized performers or some theoretical artificial
(or artificially enhanced) intelligence, which need have no necessary bounds on semiotic processing so long as the
data to interpret is within the computer’s memory and processor capacity, and which obviously does not require
music be notated using “symbols” as conventionally done in music notation, but could just as coherently “read”
music using any variety of computer coding language.
88
As will be discussed later in this document, this is likely the cause for the inability of many of the
notational additions to our modern system in the 20th-centuryto never be fully adopted into common usage. These
additions are often grafted on as needed per piece and never carefully considered in the context of the extant
elements of the system. Clearly our present system has many lacks for the requirements of the modern composer,
but should we hope to keep our system or maintain coherence among mounting notational complexification, either
our system needs to be changed from its fundaments (though not necessarily overhauled) or such elements need to
be carefully appended to or cleverly generated from the given generative elements, which themselves need better
definitions.
89
For a simple modern musical example, while a quarter note (abstract semiological element, arguably
generative and not derivative) always looks like a quarter note and will be abstractly interpreted classically as such
(though differently in the system as outlined in Chapter 1), the rhythmic translation into concrete sound and the
consequential auditory perception by a listener of a quarter note will change in regards to meter depending on the
“limiting context” of the time signature as classically understood (in 2/4 time signature, a quarter note is “one beat”
and half of the measure, while in 9/8 it is “2/3’s a beat” and 2/9’s of the measure). Given a standard performance of
a piece in 2/4, no one will likely hear the piece as in 9/8 with each quarter-note pulse being 2/3 of a beat and 2/9’s of
the macro-pulse structure (i.e. measure).

84
CONCLUSION S1.05: Given Conclusions S1.03 and S1.04, it seems advisable to have a
system with hierarchical limiters that allow for the clear pointer pivoting within an
element rather than create wholly new elements for similar types of pointers. Such
pivoting itself will need regulating according to the processor, which we assume is
human.

CONCLUSION S1.06: Building a new system from a known system or modifying a known
system to accord to new needs through the careful amending and appending of generative
elements seems advisable, though is not necessary. Again, this conclusion is necessitated
by Conclusion S1.03.

85
 CHAPTER 3
THE DREAM OF TINCTORIS:
PROPORTIONALITY IN THE RENAISSANCE AND ADVANCED STUDIES IN
RHYTHMIC COUNTERPOINT1

Muris and the Proportions from which “a song will be made”

While 14th and 15th century theorists never restricted their use of complex ratios when

describing tuning and intervallic systems,2 why was the set of ratios so limited when describing

rhythmic constructs during the 14th- and early 15th century? Anna Berger suggests that this

particular small subset is not due to their seeming correlation to the most important harmonic

intervals (since this does not explain the occurrence of 9:4), nor is it due to the possibility that

these were the easiest proportions to perform. Rather, all these proportions are all the proportions

that can be symbolically represented by some combination of mensuration signs when an equal

breve is assumed.3 Berger contests that the fact that “the earliest theorists mentioned only those

proportions that can be indicated by means of mensuration signs, and that they listed all

proportions that can be thus indicated” cannot be dismissed as a “mere” coincidence.4

Consequently, Berger then proposes that proportion signs were initially introduced not just to

override minim equivalence in the French notational system, but also to provide an

“arithmetically precise notation” that could “circumvent the ambiguity” that arose from

combining different mensuration signs.5

1
Note that all examples for this Chapter are complete works, and, as such, are found in Appendix A. These
complete musical editions are transcriptions by Jordan Key.
2
Often taking the systems of ratios as given by Boethius.
3
Berger, Anna Martha. Mensuration and Proportion Signs: Origins and Evolution. Oxford: Clarendon
Press, 1993. 178.
4
Ibid.

86
 While the use of the concept of proportion to pedagogically discuss the mensural system

as given by Franco of Cologne can be traced back to Franco himself,6 the concept of proportion

as a notational device can be traced back to the writings of Johannes de Muris, who, in the

second book of his Notitia artis musicae, written in 1321, advocated the possibility of the

proportional division of the breve into anywhere from two to nine semibreves of equal duration,7

proclaiming that “A song will be made of 2, 3, 4, 5, 6, 7, 8, 9 equal semibreves of the same

shape.”8 While simply a list of counting numbers greater than one and less than ten, this list,

when considered to its fullest implications, is rather startling in light of nearly 100% of all music

from all of Western music history prior to 1900. By this, I mean to direct us to consider how

many “songs” (a word really connoting “real” or “practical” music for “performance” rather than

“speculative” or “theoretical” music designed as a pedagogical tool) from the Medieval,

Renaissance, Baroque, Classical, and Romantic periods have any rhythmic subdivision “made

of… 5 [or] 7,” equal parts (semibreves or otherwise)? The examples (for there are some), if one

can conjure any, are so infrequent and rare as to essentially count for nothing in the broader

context of all other Western music. The question that then becomes unavoidable, and yet one that

5
Ibid.
6
Alberto Gallo described how Franco de Cologne, Marchettus de Padua, and Johannes de Muris all used
the term proportio when comparing different note-values with each other. See Gallo, Alberto, "Die Notationslehre
Im 14. Und 15. Jahrhundert." In Die Mittelalterliche Lehre Von Der Mehrstimmigkeit, by Hans Heinrich
Eggebrecht, 334-36. Darmstadt: Wissenschaftliche Buchgesellschaft, 1984.
7
Johannes de Muris, Notitia artis musicae, ed. Michels, 106-8, and Gallo, ‘Die Notationslehre’, 272-8.
8
'Fiet igitur cantus ex 2, 3, 4, 5, 6, 7, 8, 9 seinibrevibus equalibus eiusdem figurae’; Notitia, ed. Michels,
104. Of course, an additional source of the proportional concept was the Italian notational system, which permitted
sesquitertia and sesquialtera proportions; however, this was accomplished via the combination of different
mensuration signs. See Gallo, ‘Die Notationslehre’, 304-27. and Long, Musical Tastes, 54-97. It is meaningful to
note, however, that the Italian notational system could not admit the 9:4 proportion, as demonstrated by Berger,
since “the imperfect tempus had to be one-third shorter than the perfect one, there being no provision for equality in
the breve between perfect and imperfect time.” See Berger, Mensuration. 178-179 and Gallo, ‘Die Notationslehre’,
312-314

87
is not well addressed, is “why?” Why do “songs” in hierarchies of 5 and 7 essentially not exist in

Western music before the 20th century?

So, despite clear inspiration to use proportions from the writings of Johannes de Muris

and the Italian notation system at the beginning of the 15th century generally, composers of the

14th and 15th century (as well as pretty much all of Western music history) apparently opted to

only select certain divisions from Muris’ system (namely the triadic and dyadic divisions as well

as the product of the two: hexadic), utterly ignoring the potential of pentadic and heptadic

divisions.

Interestingly, a likely answer to this exclusion has been proposed by Berger, and this

answer is totally reliant upon the origins of the signs of the traditional mensural system. While

we perhaps take the notation of fractions and ratio using numbers for granted, the history of

number notation and arithmetic in the West made the prescription of ratios like 5:3 or 7:4

difficult for all but the most mathematically literate in the 13th and 14th century. While the wide

dissemination of Hindu-Arabic numeral arithmetic texts in the 15th century made the notation of

proportions self-evident, the prominence of Latin mathematics in the late Middle Ages made the

notation of such higher prime number ratios relatively difficult,9 limiting composers to use only

those proportions which could be indicated using circle/semicircle signs (which were derived

from short-hand forms of basic Roman fractions), note-shapes, and coloration. Granted, note-

shapes and colors could expand the possibilities of the traditional mensuration signs, such

invented note-shapes allowing for pentadic and heptadic hierarchies were never used. In fact, the

anonymous author of the Tractatus figurarum described exactly the struggle of notating the

combination of various mensural divisions under equal breves, developing a complete set of

9
Berger, Mensuration. 179-180.

88
note-shapes that could allow for all Muris’ proportions;10 however, the new note-shapes of

Tractatus figurarum, as mentioned previously, were never employed consistently or

comprehensively in any known source.11

However, given the possibilities opened by note-shapes and colors, the potential of

diversity has its own potential for collapse, as outlined in the prior section. Due to no clear

consensus or logic in the derivation of note shape or color, no system of notating pentadic or

heptadic mensuration arose out of the ambiguity of arbitrary shape and color. In surveying the

growing dialogue on this ambiguity in mensural demarcation at the outset of the 15th century, it

is clear that theorists were troubled by having no notational device to indicate proportions

unambiguously. The author of the Berkeley manuscript, along with Johannes Ciconia, Ramos,

and Burtius, all attempted to use the signs themselves in combination, setting one beside or

above another, to indicate various proportions implied from the proportion inherent in the signs

themselves, the upper denoting prolation and the lower tempus, since the signs were inherently

denoting proportional relationships.

There was seemingly no impetus for theorists or composers to simply use unconstrained

numerals until the mid-15th century. Considering the growing confusion and ambiguity to

indicate proportions, it is unsurprising that Hindu-Arabic fractions were welcomed with

enthusiasm by Prosdocimus de Beldemandis12 and Ugolino, the former saying:

10
Schreur, Treatise on Noteshapes, ch. 2.4-6. For a detailed discussion of the passage see 41.
11
See Kurt von Fischer, Studien zur italienischen Musik des Trecento und des fruehen Quattrocento
(Publikationen der Schweizerischen Musikforschenden Gesellschaft, ser. 2, no. 5; Berne, 1956), 118-121, and Gallo,
‘Die Notationslehre , 336-339.
12
Prosdocimus, and Franco Alberto Gallo. Expositiones tractatus practice cantus mensurabilis magistri
Johannis de Muris. 1966. 141-2.

89
 We prefer the use of figures, because through them the demonstration of proportions can
be shown more clearly. For there is no shortcoming in [fractions], but in the
[mensuration signs] there can occur error and ambiguity.13

Prosdocimo and the 15th Century Advent of Hindu-Arabic Fractions in Music

During the 15th century, we witness the advent of theoretically pan-rational proportions to

demarcate rhythmic hierarchical structures in Western music. While proportions applied to

harmonic intervals had formed the pedagogical foundation of music treatises since antiquity and

throughout the Middle Ages, their potential use to communicate rhythmic design was totally

overlooked throughout the early history of musical notation until the early-15th century.

According to Anna Berger, the earliest theoretical explanation of numerical proportions as

applied to rhythm, namely fractions wherein some numeratorial number of notes equal some

different demoniatorial number of notes of the same type, was proposed in 1408 by Prosdocimus

de Beldemandis in his Tractatus practice de cantus mensurabilis.14 Shortly thereafter, we begin

to see numerical proportions used in music, found perhaps earliest in Anthonello de Caserta’s

Amour m’a le cuer mis in ModA, compiled in Bologna between 1410 and 1411. Margaret Bent

established that these proportions came into existence to override the minim equivalence of the

French mensural system, yet beyond this specialized use, such proportions seem to have no other

use, speculative or practical, and the proportions that do appear are clearly limited and derivative

of common proportions in intervallic theory.

Practically all late-14th and early-15th century theorists outline the same sets of

proportions when discussing the possible use of proportion in teaching, describing, or notating

13
“... nobis plus placet cifrarum positio qua proportionum clarior ostenditur demonstratio. In eis namque
nulla deceptio, in hic autem ambiguitas cadcre potest et error”; Ugolino, and Albert Seay. Declaratio musicae
disciplinae. Rome: American Institute of Musicology, 1962, ii. 211.
14
Berger, Mensuration, 164.

90
rhythm or mensuration: 2:1, 3:1, 4:1, 3:2, 4:3, 9:8, 9:4, and 8:3. However, even among this small

list of ratios, even a smaller subset was more typical of most sources. Prosdocimus de

Beldemandis and Ugolino of Orvieto listed the following five proportions: 2:1, 3:1, 3:2, 4:3, and

9:4.15 Among contemporary sources, the rarer of the above proportions was 4:1 and 8:3, and 9:4,

the first of which seem to be rare not because of any indivisibility but perhaps due to seeming

redundancy with a duple subdivision of 2:1.16

With the advent of numerical proportions as an avenue for rhythmic theory and notation,

15th century mensural treatises begin making clear distinctions between mensuration itself and

proportions. DeFord notes that some sources only treat one of these topics while others treat both

but assign them separate chapters, seeming to recognize theories of mensuration as a practical

matter while theories of proportion occupied some “ambiguous middle ground between

speculation and practice.”17 Proportions in music did acquire practical weight when finally

applied to rhythm during the 15th century, but the “speculative baggage” of proportions from its

long history, extending back at least to Boethius, was unavoidable.18 Prosdocimo’s own treatise

on musical proportions, Brevis summula proportionum quantum ad musicum pertinet (1409),19

along with many other contemporary sources on proportions are indistinguishable from writings

on pure mathematics, whether the titular word “music” suggested otherwise.

15
Prosdocimus, Expositiones, 218-19; Ugolino, Declaratio musicae disciplinae, ii. 210-11. Ugolino used
the same proportions in his compositions.
16
See the table in Berger, Mensuration. 166.
17
DeFord, Ruth I. Tactus, Mensuration and Rhythm in Renaissance Music. 2018. 19.
18
Ibid.
19
Prosdocimus, Brevis summula proportionum quantum ad musicam pertinet, ed. and trans. Jan Herlinger,
Greek and Latin Music Theory 4 (Lincoln: University of Nebraska Press, 1987). However, the earliest discussion of
practical mensural proportions appears not in Prosdocimo’s treatise on musical proportions, but in his Tractatus
practice de musica mensurabili. See Prosdocimo, Expositiones, “De signis mensure,” 218-219.

91
 Together with Ugolino, Prosdocimo preferred fractions wherein numerators designated

the new number of figures that fill some time unit while the denominator designated the number

of the same figure that filled the metric unit before the proportional alteration.20 This use of

ratios of two numbers had significant power in its clarity, which is, as Ugolino proclaimed

above, why Prosdocimo and Ugolino likely promoted this paradigm. However, despite its

unambiguity and ease to write, the use of two numbers in proportion did not become popular

following Prosdocimo, though it would be continually addressed in theoretical texts throughout

the century.

Rather, later 15th century theorists and composers, especially the latter, preferred single

numerals to stand-in for proportions, indicating that a whole number of figures replaces one - 2

for duple proportion, 3 for triple, and 4 for quadruple – as though writing one more number was

far too cumbersome. While not including the seemingly redundant 1 in these ratios appears

logical and reasonable since any number against 1 is simply itself, these singular numbers were

nevertheless ambiguous. For example, the number 2 could represent either mensural diminution

per medium or duple proportion;21 the number 3 in practice could mean triple proportion (3:1) or

sesquialtera (3:2).22

20
Prosdocimus, Expositiones, “De signis mensure,” 218; Ugolino, Declaratio musicae disciplinae, book 3,
ch. 5 (“De signis modum, tempus et prolationem distinguentibus”), 210. For example, 2/1 meant that two figures
take the time typically occupied by one of the same figure, or 3/2 meant that three figures would take the time
otherwise occupied by two, etc.
21
To circumvent this particular ambiguity, theorists and composers began distinguishing between these two
potential interpretations by using a mensuration sign with “2” for diminution and a “2” alone for duple proportion.
22
In practice, “3” usually represented sesquialtera in proportional contexts, but often the meaning must be
inferred based on context, which might not be obvious without some significant consideration prior to performance.
The lack of consensus in the use of “3” and what it precisely denoted is reminiscent of the problematic variety
during the Ars subtilior, wherein there were many possible ways to accomplish the same rhythmic/mensural ends,
since Sesquialtera could also be notated with coloration. Thus, the use of both devices within a composition, source,
or composer’s oeuvre begs modern questions as to whether something slightly different was intended between
sesquialtera via “3” and coloration. This multiplicity to similar ends is actually addressed in numerous (possibly
pedagogical) works from the period; in particular, one can consider the bicinium by Johannes Stockem, “Ave Maris

92
 The complex web of mensural paths to similar ends via the use of fractional mensuration

signs, new mensuration signs, coloration, and now numbers and proportions, ultimately only

worked to further confuse the subject, despite Prosdocimo and Ugolino’s hope for new-found

clarity. Since the practice was only nominally adopted, the chimera of mensural notation only

became arguably more monstrous with arguably no more utility than before.

Johannes Stockem (c. 1445 – 1487 or 1501), composer and theorist whose music was

clearly a result of the chimeric mensural agglomeration from the early 15th century, demonstrates

the complexity and perhaps unnecessary cumbersomeness of this web of needlessly limited

mensural possibilities interconnected through seemingly redundant notational paths. In his

mensural bicinium23 Ave maris stella (see Example 3-1; note that all examples in this chapter are

presented as complete historical editions in Appendix A of this dissertation), Stockem uses all

mensural demarcative methods – sign, sign-sign, sign-number, coloration, number, and

proportion – to continuously move the music between different rhythmic structures, seemingly

recognizing that many of these methods accomplish congruent ends. Towards the conclusion of

the expansive bicinium, he even demonstrates the power of proportion to allow rhythmic

hierarchies between dyadic, triadic, and hexadic subdivisions, showing two examples of pentadic

mensurations.

Stella,” which seems to be an encyclopedic attempt at showing one all the notational paths to a handful of common,
contemporary rhythmic/mensural structures.
23
A “mensural bicinium” (pl. mensural bicina) is a new name I have given to a genre of music heretofore
largely ignored by early music scholarship. Such pieces are arguably pedagogical in nature (likely the reason for the
lack of attention modern given to these works), though well-developed enough to be considered music works in
themselves rather than mere theoretical exercises. Thus, they occupy an interesting middle-ground between
pedagogical and practical music. These pieces usually use mensuration in a multifarious manner, drawing upon
many different mensurations and methods of mensuration notation to accomplish a piece of relative rhythmic
virtuosity. The use of mensuration in such simple framework and in such a comprehensive fashion might suggest
that these pieces could and did function in some pedagogical fashion, teaching a variety of mensurations and their
combinatorial results at once, demonstrating both common and relatively exceptional practices, possibly using all
standard demarcative methods – sign, color, number, and proportion – even if the results seem needlessly redundant.

93
 Thus, while Prosdocimo’s ratios merely became absorbed into the potpourri of

possibilities, rather than becoming the paradigm, the potential to use such proportions to allow

music a departure from dyadic, triadic, and hexadic mensurations was not entirely lost on

composers and theorists, though the use of proportions to such ends was still rare. Indeed,

following Prosdocimo’s 1409 text on proportions, the Turin manuscript (Biblioteca Nazionale

Universitaria, MS [Link].9), completed in the second decade of the 15 th century, demonstrates that

the introduction of the fraction resulted in the use of proportions not inherent in the traditional

mensural system. The collection includes several pieces with fractions using the numbers five

and seven, unheard-of in earlier repertoires, evening during the Ars subtilior.24 Now that Muris’

century old proclamation to use 5 and 7 had finally come to fruition, the pertinent question

becomes not why composers did not use 5 and 7 at all, but rather why did it take composers and

theorists such a long time to apply the relatively simple construct of fractions to indicate

rhythmic proportions unavailable in standard mensural semiotics? Furthermore, why did

composers continue to use the cumbersome and ambiguous mensural chimera of the early 15 th

century well into the early 16th century if they could have accomplished all ends imaginable,

traditional and unconventional, with ratios exclusively? The history of mathematical notation in

Europe during the 15th century continues to have bearing on an answer to these questions.

Though computing with Hindu-Arabic numerals had been thoroughly described and

shown to be far more efficient and clearer than the same computations with Roman numerals and

made accessible to the average Italian merchant (let alone music theorist) of the 13 th century by

Leonardo de Pisa (Leonardo Fibonacci) in his Liber abaci of 1202,25 computing with these

24
See e.g. the ballade Puisque ami, which makes use of a 5:2 (m. 58) and a 7:3 (m. 67) proportion. See also
Richard H. Hoppin, ’The Cypriot-French Repertory of the Manuscript Torino, Biblioteca Nazionale, [Link].9’» Musica
discipline, 9 (19S7), 79-125

94
“new” numbers was only gradually accepted into practice, economic or otherwise, in Western

Europe. Ultimately, it would take nearly two centuries before the use of Hindu-Arabic numerals

would have any popularity and be somewhat commonly taught in schools. Before the wide-

spread use of Leonardo’s “new” system,26 fractions were either “written” out (i.e. with Latin

words27) or indicated by a small collection of special, non-numerical shorthand symbols

developed by the Romans to indicate common fractions.28 Interestingly and significantly, there

were such symbols for only the following subdivisions of the whole (or unica, as called in this

system): 1/2, 1/3, 1/4, 1/6, 1/12, 1/24.

Remark that all of these fractions are either dyadic, triadic, or hexadic; in this shorthand

system, fractions involving the prime numbers greater than 3, like 5 and 7, would be rounded to

the closest integer or else constructed artificially, and often only approximately, by the

summation of some set of the above given fractions.29 Also significant to note is that all the

traditional mensuration signs are derivative from these Roman fraction shorthands. Thus, not

only was the Latin fashion of fractional notation extremely limited and prone to approximation to

very simple fractional sums or multiples of 1/2 or 1/3, but the system of rhythmic subdivision

and its notation was wholly based on the shorthand of this limited system. Consequently, not

only were composers and theorists constrained by the adoption of such shorthands, but also

25
Smith, History of Mathematics, i. 215-216.
26
derivative from the Arabic system Leonardo encountered in his time in Egypt, which itself was derivative
from an older Indian system
27
A system, if one bothers to learn it, is itself rather confusing, cumbersome, and inconsistent, totally
removed from the Roman system from two millennia prior.
28
See Menninger, Number Words and Number Symbols, 158-162 and 305-307.
29
For example, 5/6 would have been shown by adding the signs for 1/2 and 1/3. However, numbers with
higher primes in the denominator (such as 3/7) could not be precisely given.

95
discouraged from seeking different proportions by the cumbersomeness and inconsistency of the

other, less popular, more erudite method of fraction notation (i.e. the orthographic form).

Ultimately, the introduction of Hindu-Arabic numbers and fractions was immeasurably

revolutionary to all Western mathematics, let alone mercantilism and the arts, thereafter.

However, as far as music practice and theory was concerned, fractions could be applied to

mensural proportions only once these numerals and their surrounding arithmetic semiology was

accepted in general and broadly practiced and taught; this broad acceptance was a gradual

process that was incomplete even by the late-14th and early-15th centuries. In fact, Berger

remarks that Prosdocimus de Beldemandis should be considered “progressive” in using Hindu-

Arabic numerals as early as 1408, at a time when “the Medici account-books were still

predominantly using Roman numerals.”30 Indeed, we should recognize that Prosdocimus’

progressiveness was likely due him being a mathematician.31 Even as late as 1482, musical

theoretician Ramos still called these numbers “figures of the Indians,” suggesting that he did not

think of them as ordinary figures, implying the still present “normal” status of Roman

numerals.32 Even in 1532, composer and theorist Martin Agricola33 still used Roman numerals to

indicate proportions in his monumental theoretical tract Musica figuralis Deudsch, written

largely in German “most simply and understandably” nonetheless.34

30
Berger, Mensuration. 182.
31
On Prosdocimus as a mathematician see Smith, History of Mathematics, i. 246-247 and ii. 502.
32
Ramos de Pareja, Bartolomeo, and Johannes Wolf. Musica practica, Bartolomei Rami de Pareja,
Bononiae Impressa opere et industria ac expensis magistri Baltasaris de Hiriberia MCCCCLXXXII. [Wiesbaden]:
[Breitkopf u. Härtel], 1969. 82.
33
not to be confused with the earlier composer Alexander Agricola from the mid to late 15 th century.
34
Agricola, Martin. Musica figuralis deudsch (1532): Musica instrumentalis deudsch (1529), Musica
choralis deudsch (1533), Rudimenta musices (1539). Hildesheim: George Olms, 2011. Quotation taken from the title
leaf from Musica figuralis Deudsch.

96
 Thus, one can conclude that, despite fractions’ introduction to provide musicians with

precise and clear notational demarcation for mensuration and rhythmic structure, musicians and

composers (along with many theorists) took almost a century to adopt proportions without

exclusion and in an arithmetically correct fashion, forgetting the fraction’s initial musical use as

a substitute for already given and rather limited sets of mensuration signs. As we will come to

see, the towering 15th century theorist-reformers Tinctoris and Gaffurius would ultimately be the

ones to initiate this change thoroughly (though never completely), emancipating the fraction

from traditional mensuration signs.

What then was the germ that allowed the 15th century to see the flowering of proportion

in music as well as all the arts? What allows Tinctoris and Gaffurius to succeed where

Prodocimo met with far less success? What, if anything, would well-educated 15th century

musicians, like Tinctoris and Gaffurius, have learned with respect to mathematics?35 Since we

are dealing with elementary constructs in mathematics – numbers and proportions – it seems

appropriate to seek answers in 15th century maths.

The Rise of Fractions in the 15th Century: The Road to Tinctoris

Naturally, one might suppose that if Martin Agricola hoped to be “simple and

understandable” in his presentation of musical rudiments to his new-found printed-media

consumer in the early-16th century, he might have used the numerical system that had become

common in commerce by that time. Clearly even by the 16th century, music still had some

catching up to do with other spheres of European thought (academic and practical), despite

centuries old and continuously resurfacing encouragements and admonitions from music

35
For a more detailed discussion of this problem, see Berger, Anna Maria. 1990. "Musical Proportions and
Arithmetic in the Late Middle Ages and Renaissance". Musica Disciplina. 44: 89-118.

97
authorities to use Hindu-Arabic numerals and their proportional notations to indicate mensural

structures without ambiguity. Clearly, tradition was (and always is) difficult to forsake.

The Hindu-Arabic numerals and fractions of them, along with all arithmetic operations,

did gain wide-spread popularity during the 15th century, despite what music’s general state might

suggest, a state more in line with the concurrent elementary education of abacus arithmetics and

the historical weight of Euclid and Boethius, both of whom dealt with “the cumbersome Greek

ratio-system which testified to the ancient difficulty with fractions,” in the words of mathematics

historian D. E. Smith.36 Generally, there were three kinds of common37 mathematics textbooks

from this period:38

1. Theoretical books based on Euclid and Boethius, which dealt with figurate numbers39 as

well as the Classical (and as previously noted, “cumbersome”) Greek ratio system.

2. More traditionally oriented commercial arithmetical textbooks that taught the long-held

tradition of abacus arithmetics, which used Roman rather than Hindu-Arabic numerals

and wherein one made calculations on a counter (abacus) with the help of pebbles

(calculi).

3. More forward-thinking commercial textbooks called algorisms (or algorismi), which

used Hindu-Arabic numerals and were consequently targeted at a new audience of

36
Smith, David Eugene. Rara Arithmetica: A Catalogue of the Arithmetics Written Before the Year MDCI,
Part 2. [Place of publication not identified]: Nabu Press, 2012., 4.
37
Here we are not considering texts for specialized mathematicians, all of which were exclusive to the
scholarly elite of the field and so irrelevant to wider, popular mathematical trends at the time.
38
For a detailed discussion of the various mathematical textbooks used during the 14th, 15th, and 16th
century, see Smith, Rara arithmetica, 4-7.
39
A figurate number, a concept popular in Classical Greek mathematics as well as during the European
Renaissance (also known as figural numbers), is a number that can be represented by a regular geometrical
arrangement of equally spaced points. Should such an arrangement form a regular polygon, the number is called a
“polygonal number.” Such numbers were immensely popular with the Pythagoreans and Neo-Pythagoreans.

98
 merchant and other commercial classes who used numbers regularly in their work and

who could benefit from a less cumbersome system of record and account keeping as well

as commercial calculation.

While in all countries outside Italy, the traditional Roman arithmetics still competed with

algorisms at the outset of the 15th century, in Italy the algorisms slowly gained preeminence by

the late-15th century, likely due to the burgeoning international trade centers in Italy and the need

for fast and clearer methods for the calculation of sums and products as well as the conversions

of endlessly proliferate currencies. By contrast, titles of even 16th century German arithmetic

texts still used expressions like “on the lines and field,”40 referring to the lines of the abacus on

which the counters, like seeds in a field, were cast.41 While by the middle of the 16th century

Northern Europe would also witness the predominance of algorisms, the 15th century saw just the

beginnings of this revolution in mercantile pockets of Italy.

So, what texts were musicians, composers, and music theorists most familiar with? The

answer is likely self-evident: Boethius for music, and traditional abacus arithmetics from grade

school. Indeed, it is well-known that Euclid and Boethius were both required textbooks in all

university and cathedral schools in Europe at this time. While Boethius was clearly admired by

musical scholars, Boethius was realistically not influential within mathematical scholarship at

this time; even Euclid did not hold the preeminence in the Middle Ages and Renaissance among

mathematicians that he once held in the Classical world, most mathematicians were more

interested in the recent import of Algebra from the Middle East at this time rather than well-

40
In 16th century German, “auff der Linien und Felder“
41
Menninger, Number Words, 349-353

99
understood geometrical proofs.42 Claude Palisca notes that even Boethius’ De institutione musica

was of little significance to the scholarly or practical musician at this time since the

contemporary “preoccupation with polyphony and the problems of measured music” was in no

place addressed by Boethius; consequently, this led to De institutione musica’s “neglect in the

training of musicians.”43 Despite the declining applicability of Boethius or Euclid to

contemporary mathematical and practical issues of arithmetic, centers of higher learning like the

University of Paris, in its statutes of education throughout the 14th, 15th, and 16th centuries,

required that admitted students listen44 to ‘aliquos libros mathematicos’.45 Despite this

requirement to listen to mathematical texts and engage with that material, the conservativeness of

these treatises is testament to the continued use of Roman numerals up to and throughout the 15th

century, only slowly falling out of fashion across Europe by the end of the 16th century.

Despite the Classical, old-fashioned, theoretical predilection of universities, algorisms

and texts on abacus arithmetic were unquestionably the most popular mathematical books of the

Renaissance, merely due to their commercial utility, optimized or not. Swetz notes that the first

50 years of the printing press in Europe (essentially the second half of the 15 th century)

witnessed a relatively equivalent number of publications between theoretical and commercial

42
See Hankel, Hermann. Zur geschichte der mathematik in alterthum und mittelalter (classic reprint).
[Place of publication not identified]: FORGOTTEN Books, 2016. 353ff.; Cantor, Moritz. Vorlesungen über
Geschichte der Mathematik. Bd. 2 Bd. 2. Vaduz: Sändig Reprint, 2005, ii. 125-280; Juškevič, Adolf Pavlovič, Hans
Wussing, and Viktor Ziegler. Geschichte der Mathematik im Mittelalter. Basel: Pfalz-Verlag, 1966. Part 4.
43
Preface to Boethius, Anicius Manlius Severinus, Claude V. Palisca, and Calvin M. Bower. Fundamentals
of Music. Music theory translation series. New Haven: Yale university press, 1989. p. xiii.
44
Since many books had few copies at this time, most books were read aloud and “listened to” by students
in large forums rather than “checked out” from some library and read individually, as might be done today. Books
were still rare and valuable, especially ones dealing with specialized subjects unappealing to the commercial printer
of the early 16th century, and so books were frequently inaccessible for individualized use by students outside a
library, wherein books were often chained to the shelves that held them.
45
“some books of mathematics.“ See Hankel, Zur Geschichte der Mathematik, 355.

100
mathematical texts: 30 books on practical arithmetic and 26 Boethian-style treatises.46 However,

as Berger notes, after about 1500, commercial arithmetic books took clear precedence. 47 In

contrast to theoretical mathematics, which was exclusive to university students, nearly everyone

with a rudimentary education in Europe by the end of the 16th century learned how to compute

with Hindu-Arabic numerals. Berger notes that even by the end of the 15th century, nearly “half

the commercial textbooks published… were written in the vernacular: seven in Italian, four in

German, and one in French.”48 Being written in vernacular made these texts even more popularly

successful and influential compared to their Latin, theoretical counterparts.

Whether algorismi or abacus-based, arithmetic was taught in schools of all kinds.

According to Swetz, boys taught at private and municipal lay secondary schools were taught

commercial mathematics from the age of ten to fourteen.49 Menninger demonstrates that

Cathedral schools also likely offered instruction in commercial arithmetic.50 Similarly, German

Lutheran reformer Philip Melanchthon recommended, at the beginning of the 16 th century, that

all university students at Wittenberg learn the “new” commercial mathematics, showing the

initial commercial migration of Hindu-Arabic numerals to the north of Europe.51 Despite still

using Roman numerals, Martin Agricola’s Musica figuralis Deudsch, written for boys at Latin

schools at Hamburg, Magdeburg, and Lubeck, made extensive use of new arithmetic techniques

46
Swetz, Frank J. Capitalism and Arithmetic: The New Math of the 15th Century ; Including the Full Text
of the Treviso Arithmetic of 1478. La Salle, Ill: Open Court, 1989. 33-34.
47
Berger, Mensuration. 199-200.
48
Berger, Mensuration. 200.
49
Swetz, Capitalism and Arithmetic, 10-24.
50
See in particular Menninger, Number Words, 433-436.
51
Menninger, Number Words, 434.

101
derived from commercial mathematics.52 Berger notes that knowledge of commercial arithmetic

was “not only useful in business circles but also among ecclesiastics, who had to take turns in

balancing the budget of their cathedrals or monasteries.”53 Art historian Michael Baxandall

eloquently summarizes the 15th century, educated European’s affinity for this “new”

mathematics of commercialism:

These people did not know more mathematics than we do: most of them knew less than
most of us. But they knew their specialized area absolutely, used it in important matters
more often than we do, played games and told jokes with it, bought luxurious books about
it, and prided themselves on their prowess in it; it was a relatively much larger part of
their formal intellectual equipment.54

Such “new mathematics” treatises included detailed instructions and methods on multiplying and

dividing fractions. Two popular examples include Johannes de Lineriis’ treatise Algorismus de

tninutiis and even Prosdocimus’ own treatise Algorismus de integris sive pratica arismetrice de

integris (written in 1410 and printed in Padua in 1483 and again in Venice in 1540), which is

entirely devoted to operations with fractions. Undoubtably, Tinctoris, who introduced the most

systematic and logically consistent proportional mensural system with the idea that rhythmic

proportions should be cumulative (a process reliant on product operations of fractions), must

have been exposed to the new mathematics either in Italy (very likely, given fraction’s and

Hindu-Arabic numeral’s growing popularity in Naples) or earlier during his studies at the

University of Orleans (possible, but less likely). Anna Berger aptly summarizes the drastically

52
Agricola, Musica figuralis Deudsch, sig. Azr.
53
Berger, Mensuration. 201.
54
Baxandall, Michael. Painting and Experience in Fifteenth Century Italy. Oxford, UK: Oxford University
Press, 1988. 101. The ‘specialized area’ Baxandall refers to was commercial arithmetic. One example of the
“games” people played with mathematics was Rithmomachia, a game that dealt with a “battle” strategy of numbers
and their proportions, equally as popular throughout the Late Middle Ages and through the Renaissance as a similar
yet less mathematical strategy game, Chess.

102
differing situations for preeminent composer-theorists at the beginning of the 15th century and by

the end of that same century:

On the basis of [Tinctoris’] acquaintance with [new commercial arithmetic], he must

have concluded that a figure written over another one looked like a fraction and should
be treated as such, that is, that they should be multiplied when occurring successively.
On the other hand, Anthonello de Caserta probably still thought [at the beginning of the
15th century] in terms of Boethian number ratios, possibly was not familiar with the new
mathematics, and might not have known how to multiply fractions.55

Now that we can see the complex intermingling of mathematics (both academic and

popular) with broader European culture (particularly music) during the late-14th, 15th, and 16th

centuries, the context of the numerical proportion’s introduction to music becomes clear.

However, this history begs a further question, for it is indeed puzzling that composers appear to

have written music that made more extensive use of elaborate proportions during the late 14th

and early 15th century, a time of relative illiteracy regarding “new” commercial mathematics

during which theorists and composers alike were searching for better and clearer notation

methods to indicate rhythmic structure unambiguously. However, once the fraction became

associated with proportions and Hindu-Arabic numerals made the communication of these

fractions limitlessly variable without any ambiguity, starting with Prosdocimo and continuing

forward throughout the 15th century, composers lost nearly all interest in writing music with

elaborate rhythmic proportions, relegating the subject utterly to theory. A thorough answer to

this question remains to be given, though it likely lies in the realm of aesthetics as applied to

Humanism rather than the diminishing influence arcane Scholasticism.

What is of even further interest for speculation is why, despite a clear practical disinterest

from composers in the potential of proportional notation of rhythm during the 15th and 16th

55
Berger, Mensuration. 204.

103
century, did theorists persist in developing and discussing theories of rhythmic proportion? What

explains the voluminous proportional works of Tinctoris and Gaffurius, when demand for any

such theory seems to be absent. Naturally, we today might suggest that music theory need serve

no practical purpose, and indeed this was even true as far back as the Classical world of ancient

Greece. However, Berger presses this particular question about the motives of these music

theories when she observes the “pains” such theorists went to create these “complicated theories,

which they must have known neither they nor anybody else would ever apply to music.”56 What

was the motivation for such seemingly Herculean theoretical tasks that bore arguably no

relevancy on practical music?

While the lack of interest from composers is not well understood, Berger suggests that

the theoretical fascination with the subject of proportional music, demonstrated by a handful of

radical and prominent theorists, parallels the fascination that the new mathematics held for the

educated public of the late 15th century. Berger claims that “[i]t seems natural that music

theorists would apply the mathematical techniques that they observed in everyday life to music,”

and since the theory of harmonic proportions was well understood, theorists looking for new

fertile grounds of research turned their attention to rhythmic proportions.57 The scholarly

exchange between speculative music and mathematics during this time is well demonstrated by

the collegial exchange between late 15th century mathematician Luca Pacioli and contemporary

music theorist Gaffurius. Architectural historian John Onians describes in detail how Pacioli’s

theory of architecture may have been influenced by Gaffurius’ description of the modes.58

56
Ibid., 208
57
Ibid., 209.
58
Onians, John. Bearers of Meaning: The Classical Orders in Antiquity, the Middle Ages, and the
Renaissance. Princeton, N.J.: Princeton University Press, 1990. 222-223.

104
Berger, in turn, suggests that Gaffurius’ theory of rhythmic proportions was possibly influenced

by Pacioli’s Summa de arithmetica geometria proportion et proportionalita of 1494, which

includes substantial lessons in commercial mathematics.59 Berger notes that Gaffurius likely

knew Pacioli, since they taught at the same time in Milan. Furthermore, Gaffurius owned a copy

of Pacioli’s Summa de arithmetica, and Gaffurius’ Practica musicae, the fourth book of which is

completely devoted to proportions, was penned only two years after the Summa.60 Consequently,

Berger reasonably concludes that “we can thus draw a direct line between the most important

late-15th century commercial mathematician and the most important music theorist.”61

Mathematical historian David Eugene Smith, addresses an even more poignant and less

hypothetical example with the music theorist Glareanus, who was also a mathematician

concerned with optimizing methods of solving commercial arithmetic, and in a book from 1538

devised a better system of arranging the numbers for the Rule of Three.62

Thus, this history of scholastic and commercial arithmetic during the end of the Dark

Ages – the popularity of applied arithmetic of traders, accountants and craftsmen against the

backdrop of a stultifying university-based Medieval Boethian-oriented speculative mathematics

based on Neopythagorean interests from a millennia prior to the Renaissance – provides a

relevant cultural context for 14th, 15th, and 16th century developments of rhythmic proportions,

59
Berger, Mensuration. 209.
60
Ibid.. Interestingly, Pacioli was likely not the only influence on Gaffurius. Clement Miller has suggested
that the Tractus practicabilium proportionum was formulated in Naples between 1478 and 1480, likely under
Tinctoris’ influence, since Gaffurius and Tinctoris had much correspondence during this time. See Miller, Clement
A. 1970. "Early Gaffuriana: New Answers to Old Questions". The Musical Quarterly. 56, no. 3: 385.
61
Ibid.
62
Smith, History of Mathematics, ii. 489. The “Rule of Three” was an early method of conceptualizing and
solving fractions cross multiplication problems wherein one sought to solve for a missing value in a series of two (or
possibly more) equivalent ratios.

105
but leaves us with many unanswered questions, which seem, in light of this history, important

questions to ask.

Whether useful to the composer generally or not, such proportional music speculations

did lay a groundwork for a handful of composers, some famous and some obscure, who would

explore the implications of a theoretically infinitely diverse system of rhythmic demarcation

throughout the 15th and 16th century, regardless of its broader popularity. The insight Tinctoris

will gain from this new commercial mathematics were ultimately contingent upon the Hindu-

Arabic model imported by Leonardo of Pisa at the beginning of the 13 th century and slowly

adopted throughout European discourse for the next three hundred years. Starting in the late 13th

century after the Franconian revolution, theorists realized the potential to divide the breve into

integer subdivisions between two and nine equal parts, an insight codified in the early-14th

century by Johannes de Muris in 1321. However, composers of the 14th century selected only

dyadic and triadic (along with hexadic) divisions, utterly ignoring possibilities of pentadic or

heptadic subdivision suggested by Muris, likely due to the long-standing choice to communicate

mensural proportions through a set of adopted shorthands for common Roman fractions, none of

which contained pentadic or heptadic rationales. Thus, due to a particular notational choice of

shorthands and a lack of a good system of arithmetic, composers perhaps did not even consider

rhythmic subdivisions beyond two and three, lacking the basic conceptual tools to conceive of

such relatively simple musical structures.

Thus, the wide-spread dissemination of Fibonacci’s commercial mathematics based on

the Hindu-Arabic numeral system was revolutionary for music (let alone everything else) in

Europe, now granting an obvious tool to indicate any rational rhythmic proportion. Furthermore,

this mathematical innovation also allowed for a less ambiguous notation of the standard

106
proportions already in use. Ultimately, some new proportions came into infrequent use during

the early- and mid-15th century. However, such examples are vanishingly rare, and did indeed

practically vanish from musical practice during the 15th century, relegated almost exclusively to

theoretical speculation, barring those that conformed with the traditional mensurations.

This theoretical speculation on the potential of proportional music beyond traditional

dyadic, triadic, and hexadic mensuration would reach its first major culmination with Johannes

Tinctoris during late-15th century, who would essentially emancipate the fractions from

mensuration signs and use these ratios in a logically arithmetic fashion, officially formulating the

tool to allow for the infinite variety of rational rhythmic proportions, even beyond pentadic and

heptadic rationals, and essentially placing rhythm on equal status with harmonics for the first

time. We will soon see how this insight from Tinctoris and its continued development by

Gaffurius would point to even more radical departures from rhythmic tradition during the late-

16th century in England.

The Schools of Proportional Mensuration: Hothby, Tinctoris, and Gaffurius

While a broad general practice of mensuration was in use during the 15th century, there

was also much disagreement among theorists and composers about the more nuanced issues of

mensuration.63 Though the basic four mensurations’64 use and interpretation were widely agreed

upon by late-14th and early-15th century theorists, the notation of exceptional mensural structures,

beyond this traditional small subset, was highly variable and unclear, and thus needed new

methods to indicate nuanced rhythms. Despite the problems of complexity evident during the

63
For a brief synopsis of the differences in standard mensural practice during the 15 th century see DeFord,
Tactus. 114-135. See also Berger, Mensuration.
64
Namely, perfect tempus with perfect prolation (circle-dot), perfect tempus with imperfect prolation
(circle), imperfect tempus with perfect prolation (C-dot), and imperfect tempus with imperfect prolation (C).

107
late-14th century, the simplified notational systems of the early and mid-15th century continually

struggled with consensus and clarity. While generally, theorists advocated a variety of mensural

signs derivative, yet often arbitrarily so, of the standard four signs to signify the proportions

commonly used in 15th century music, some theorists, perhaps radically, suggested the use of

numerical proportions to clarify meaning, recognizing the problematic arbitrariness and

obfuscation of signs.65

Despite a collapse to a relatively simpler system of mensural notation following the late-

14th and early-15th century, the theoretical sources that discuss temporal musical demarcation in

15th and 16th century music are diverse, ranging between rudimentary instructional manuals to

sophisticated philosophical and speculative works targeting purely theoretic and humanistic

scholars rather than practical composers and performers. Some such texts are loosely organized

or cursory treatments of the subject, clearly intended only for personal or private instruction,

while others are comprehensively structured formal tomes. By the second half of the century,

some treatises begin further reformations of the system, while many still only transmit

conventional information hardly modified from traditional Franconian precepts. Though perhaps

less amorphous than the experiments of the Ars subtilior, mensural notation was not a static

system during the 15th and 16th century but was still fluctuating and evolving throughout the

Renaissance.

However, as we will overview herein, by the 1520s composers largely discontinued the

use of the more complex and esoteric aspects of the mensural system inherited from the 15 th

century, barring occasional and mostly erudite demonstrations of theory.66 Despite the

65
Deford, Tactus, 134.
66
Ibid., 14.

108
abandonment of 15th century diversity in mensural possibilities by most composers, theorists

certainly continued academic discussion and debate of mensural nuance until the late-16th

century since the music of earlier generations remained relatively present in repertoire until the

second half of the 16th century.67

Due to the paring back of mensural complexity during the early-15th century, there was

relatively no appreciable differentiation between regional mensural theories. In fact, the

foundations of nearly all 15th century treatises on mensuration were nearly wholly derivative

from the 14th century Ars practica mensurabilis cantus (aka Libellus cantus mensurabilis),

attributed to Johannes de Muris;68 therein, Muris established the prototype of theoretical and

pedagogical writings on mensuration throughout the 15th and 16th centuries, systematically

unfolding the principles of mensural notation developed and outlined in his earlier writings and

in the seminal texts by Franco de Cologne and Philippe de Vitry.69

67
Ibid.. As Deford aptly notes, “[t]he relation between theory and practice is quite different when theorists
focus on a historical repertoire than it is when they discuss the music of their own time.,” and so it is not surprising
to witness theorists in the mid-16th century still debating the use of mensuration in a repertoire that is relatively
outdated.
68
Muris, Johannes de, and Christian Berktold. Ars practica mensurabilis cantus secundum Iohannem de
Muris: die Recensio maior des sogenannten "Libellus practice cantus mensurabilis. München: Verlag der
Bayerischen Akadamie der Wissenschaften, 1999. In fact, as noted by Deford, most of the surviving sources of the
14th century treatise come to us from the 15th century (more than 50 in all), with the earliest source from c. 1375 in
the Berkeley Manuscript. See DeFord, Tactus. p. 15. See also Ellsworth, Oliver B. The Berkeley manuscript:
University of California Music Library, MS. 744 (olim Phillipps 4450) : a new critical text and translation on facing
pages. Lincoln: University of Nebraska Press, 1984. 148-83. Several early 15th century writings on mensuration are
commentaries of Music Libellus. The most extensive are Prosdocimus de’ Beldomandi’s Expositiones tractatus
practice cantus mensurabilis magistri Johannis de Muris in Prosdocimus, and Alberto Gallo. Prosdocimi de
Beldemandis opera. Bologna: [Arti grafiche Tamari], 1966. Other examples include Book 3 of the Ugolino of
Orvieto’s 1430’s text Declaratio musicae disciplinae, II: 167-266, which builds on the writings of Prosdocimo, as
well as Prosdocimo’s two works on mensuration: Tractatus practice de musica mensurabili (1408), Coussemaker,
Charles Edmond Henri de. Scriptorum De Musica Medii Aevi. Hildesheim, W. Germany: G. Olms, 1987, III: 200-
28., which is heavily dependent on Muris, and Tractatus practice cantus mensurabilis ad modum italicorum (1412),
ed. Claudio Sartori, in Sartori, Claudio in Una redazione inedita del "Tractatus practice cantus mensurabilis ad
modum italicorum". Firenze: L.S. Olschki, 1936. 35-71, which deals with 14th century Italian mensural practices.
69
The most important earlier writings on mensural theory are Franco of Cologne’s Ars cantus mensurabilis,
ed. Gilbert Reaney and Franco. Franconis, [Rome]: American Institute of Musicology, 1974.; Philippe de Vitry’s
Ars nova, ed. Gilbert Reaney, André Gilles, and Jean Maillard, Stuttgart, W. Germany: American Institute of

109
 However, just as in the last decades of the 14th century following the systemization of the

Franconian system during the early- and mid-14th century, the last three decades of the 15th

century witnessed a new divergence between national styles and theories of mensural music. By

the 16th century, theoretical writings varied greatly between German and Italian sources with the

French, Spanish, and English70 contributing their own nuanced versions of either the German or

Italian theories.71 Among this growing variety, we can mark three general phenotypes of practice

among theorists and composers on the Continent and Isles:

1. those who advanced and maintained the traditional “signed” mensural system

(utilizing seemingly arbitrary symbols to designate hierarchical rhythmic

subdivision);

2. those who developed a pure (or nearly pure) system of numerically designed

mensurations (employing exclusively or primarily numbers, often in ratios, to

designate hierarchical rhythmic subdivision);

3. those who used both systems about equally and interchangeably.

Of these three notational phenotypes, the first was the most common, followed closely by the

third; however, the second category – those rarely employing mensuration signs – are uncommon

and often exclusively theorists with little to no “practical” music.72 The most exceptional

Musicology, 1986.; and Johannes de Muris’s Notitia arrtis musicae, ed. Ulrich Michels, Corpus scriptorium de
music 17 ([Rome]: American Institute of Musicology, 1972), 47-107.
70
Deford, Tactus, 15. Deford notes that “only four books… on mensura[tion]… were printed in England in
the sixteenth century,” all of them in interestingly in the final two decades of the century. However, as we will
examine herein, we will find that these texts offer some interesting perspectives on the development of rational time
notation and its practical ramifications, practices which were arguably only speculative in continental sources for
their entire history.
71
Ibid., 14
72
Even those who championed such a system often found themselves compromising their practical music,
using traditional symbols when appropriate and typical, and only using numerical ratios in exceptional
circumstances that resided outside the prerogative of any established signs. Such compromising was likely for the

110
example from this third phenotype is perhaps Johannes Tinctoris (c. 1435 – 1511), a renowned

and prolific Dutch theorist as well as equally noteworthy composer.

Herein, we will occupy ourselves with the rarer second phenotype along with examples

from the third that bend towards the second, since they possess the primary tool during the 15th

and 16th century to depart from the exclusively dyadic and triadic hierarchies in Franconian

mensurations. This distinct and exceptional strand of mensural theory emerged between c. 1460

and c. 1490, a time during which there was much attempted reform of mensural notation, much

of which never quite established itself.

John Hothby and the Beginnings of Proportional Mensural Schools

The first musician during this period to propose numerically rational notations of

mensural hierarchies was Englishman, theorist, and composer John Hothby, along with Italian

Guilielmus Monachus and Spaniard Bartolomeo Ramis de Pareja, all of whom were concurrently

active in Italy. Hothby, being the first in this numerical reformation movement, fits more

precisely into the third phenotype, not wholly suggesting a departure from tradition; along with

various other modifications and reforms that do not concern us herein,73 his system utilizes

various combinations of circles, semicircles, dots, and numbers to differentiate levels of

mensuration.74

sake of the classical training of the performer, who would, of course, be more accustomed to seeing a mensural sign
rather than its numerically rational equivalent. Consequently, there are few, to arguably no, composers of the second
phenotype, only composers who perhaps wished to pursue this end but did not for practical reasons. There are,
however, plenty of examples of individual works that within themselves suggest the second phenotype.
73
For example, Hothby’s system of notating modus mensuration level, a hyper-level of mensuration not
communicable via standard circle/semicircle mensuration notation.
74
Earlier theorists used a variety of systems for notating mensuration, but none, barring the occasional use
of 3 or 3/2 to suggest sesquialtera, suggested the exclusive use of numbers to indicate rhythmic or mensural
proportions. See Berger, Mensuration, 13-28.

111
 Hothby’s theories of rhythmic prescription are derivative of a larger trend during the mid-

15th century wherein theorists began to combine numbers with the traditional mensuration signs,

called modus cum tempore, to indicate macro rhythmic structures, called modus. Given such

combinatorial signs, the circular symbol represented perfect or imperfect minor modus, while a

number like 2 or 3 following the circle represented tempus, or how the modus should be

proportionally subdivided. The four common signs of this type are O3 (perfect modus/perfect

tempus), O2 (perfect modus/imperfect tempus), C3 (imperfect modus/perfect tempus), and C2

(imperfect modus/imperfect tempus).75 John Hothby was the first theorist to discuss these signs in

a systematic fashion, expanding the combinatorial principle to generate signs with three

elements: circular signs that represent major modus and minor modus, numbers that represent

tempus, and dots inside the signs to indicate prolation.76 Ultimately, despite the numerous

possibilities outlined by Hothby, the only signs that had any practical significance are those

above: O3, O2, C3, and C2.77

In addition to these combinatorial signs, John Hothby’s Regulae cantus mensurati78

allowed for the use of numbers and ratios of numbers in isolation, listing the usual proportions -

2:1, 3:1, 4:1, 3:2, 4:3, 9:8, 9:4, and 8:3; however, in his motet Ora pro nobis from the Faenza

Codex, he enlarged the rhythmic possibilities by including the ratios 5:2, 5:4, and 7:4 (see

75
Deford notes that these combinatorial signs may have been invented by DuFay, who “used them in a
large collection of Mass Propers that he composed for Cambrai in the 1440s.” See Deford, Tactus. 136. See also
Alejandro Enrique Planchart, “Guillaume De Fay’s Benefices and His Relationship to the Court of Burgundy,” Early
Music History 8 (1988), 167. For an overview of the uses of O2 in the fifteenth century, see Sean Gallagher,
Johannes Regis, “Epitome musical” (Turnhout: Brepols, 2010), 104-114.
76
Hothby, John, Gilbert Reaney, and Thomas Walsingham. Opera omnia de musica mensurabili. [Rome]:
American Institute of Musicology, 1983. 21, 53-57. Hothby also proposes an alternative system in which mensural
levels are represented by concentric circles, rather than combinatorial signs. See Berger, Mensuration and
Proportion Signs, 148-63, for a detailed discussion of modus cum tempore signs.
77
Deford, Tactus. 136.
78
Hothby, Opera omnia, 54.

112
Example 3-2), giving practical use to Muris’ suggestion to use 5 and 7 as rhythmic subdivisions

as well.79

Hothby’s theoretical writings are preserved in several manuscripts with overlapping

content, though the greatest source of both his theory and music in the 15 th century appendages

to the Faenza Codex. Both Gailielmus Monachus and Ramis follow Hothby’s system in their

respective theoretical treatises: De preceptis artis musice (c. 1480-90)80 and Musica practica

(1482, print).81 Ultimately, Hothby’s modifications to the Franconian/Muris system of notation

would perhaps be the most long-lasting and influential, albeit only in particular parts, likely due

to their heavily reliance on established tradition.82

Tinctoris and Gaffurius: Revisionists of Proportional Music

If Hothby and his followers established the potential of a fully rational system of

rhythmic demarcation though maintained a firm hold on to tradition, it was Johannes Tinctoris

that first realized the economy, clarity, and possible variety inherent in that potential. In his

Proportionale musices, Tinctoris was the first to give equal theoretical consideration to harmonic

79
Of further interest is Jocobus de Regio (fl mid-15th century), a contemporary of Hothby. De Regio was a
theorist and a Carmelite, and is known only by a brief treatise on proportions, copied into the Faenza Codex (I-
FZc 117) by Carmelite Johannes Bonadies at the convent in Reggio nell' Emilia on 14 September 1474. De Regio’s
treatise is found alongside Hothby’s similar treatise on proportions and his proportional work Ora Pro Nobis. With
Hothby’s work, de Regio’s treatise is one of the earliest to apply the theory of proportions to mensural music. This
totally unstudied and untranslated treatise also contains some interesting two-part examples using exceptional
proportions. No modern edition of this work or its music exists. For a facsimile edition of the Codex Faenza see
Carapetyan, Armen. An Early Fifteenth-Century Italian Source of Keyboard Music: The Codex Faenza, Biblioteca
Comunale, 117. [Place of publication not identified]: American Institute of Musicology, 1961.
80
Monachus, Guilielmus and Albert Seay. De preceptis artis musicae. [Place of publication not identified]:
American Institute of Musicology, 1965.
81
Ramis de Pareja, Musica practica. In the second half of the 15th century, Guilielmus Monachus still
described the same set of proportions as outlined above: 2:1, 3:1, 4:1, 3:2, 4:3, 9:8, 9:4, and 8:3. However,
Monachus also adds to this list the inverted versions of all of these fractions. See Monachus, De preceptis artis
musicae, 19-44.
82
Hothby’s influence, via Monachus and Ramis was transmitted to the sixteenth century through Ramis’
student Giovanni Spataro. See DeFord, Tactus. 16-17.

113
and rhythmic proportions. He described twenty-five different rhythmic proportions and their

inversions. Tinctoris was the first theorist to attempt a comprehensive reform of the mensural

system, a reform which was further developed and perfected a generation later by Franchinus

Gaffurius (1451 –1522).83

Working in Naples throughout the 1470s and 80s, Tinctoris was the most prominent and

original Renaissance musician to propose a clear yet distinctly multifarious mensural system that

could potentially provide infinite variety to musical temporal structuring. Between 1472 and

1477, Tinctoris produced the majority of his theoretical oeuvre, which included numerous short

treatises on basic notational principles, a dictionary of musical terms (Terminorum musicae

diffinitorium), a substantial book on counterpoint (Liber de arte contrapuncti), and, most

interestingly, his seminal text on musical proportions (Proportionale musices).84

Tinctoris objects to many of the traditional conventions of mensural denotation, including

the reforms and additions of Hothby, rather approaching rhythmic notation with rigorous logic

and “strict standards of rationality,” according to DeFord.85 Though Tinctoris’ system of numeric

rhythmic hierarchical demarcation had little effect on actual composition during the late 15th or

16th century (even in his own oeuvre), these ideas were influential on subsequent theorists who

shared the sense that mensural music needed a clearer, more logical and manifold systemization

83
Gaffurius, Franchinus, and Clement Albin Miller. Practica Musicae. [S. l.]: American Institute of
Musicology, 1968, bk. 4. See also Clement A. Miller’s detailed discussion of Gaffurius’ proportions in ‘Gaffurius’s
“Practica musicae": Origin and Contents’, Musica disciplina, 22 (1968), 105-28.
84
All of Tinctoris’ writings, except Terminorum musicae diffinitorium ([Treviso, c. 1495]; repr. New York:
Broude Bros., 1966), are in Johannes Tinctoris, Opera theoretica, ed. Albert Seay, 3 vols. In 2, Corpus scriptorum
de musica 22 ([Rome]: American Institute of Musicology, 1975-78). A new, online critical catalog edition of
Tinctoris’ complete theatrical works in side-by-side translation, along with related studies, is available at
[Link]
85
DeFord, Tactus. 18.

114
of rhythmic demarcation, though most of these theories never spread beyond speculation and

pedagogical erudition.

Contrary to Hothby’s system of combinatorial signs that allow for the use of singular

numbers in conjunction with traditional mensural signs, Tinctoris argued vigorously against such

use of numbers or modification of traditional signs to signify concrete proportions, proposing

that only ratios of numbers should be allowed to serve such functions. 86 While Tinctoris suggests

the abandonment of traditional signs altogether, since all such signs can be easily replaced with

unobfuscated numbers, he does allow for the typical use of mensuration signs, which can operate

independently of numerical proportions, the latter simply causing a modification of the former

when used. Consequently, under Tinctoris’ proportional system, proportions cannot change

mensuration without a change of mensuration sign.87

For Tinctoris, the logic of his system was unassailable, for a proportion is a relationship

between two numbers, and consequently a singular number cannot clearly represent any such

proportionality; furthermore, mensuration is distinct from proportion and thus neither should be

conflated by use of combinatorial signs.88 Despite his thoroughgoing efforts and reasonable

86
Tinctoris, Proportionale musices, book 3, ch. 2 (“Qualiter proportiones signandae sint”), 45.
87
e.g. if a triple proportion alters the semibreve level of subdivision so that three semibreves replace the
time of one under C, the breves in C are still imperfect unless the proportion sign is accompanied by an additional
sign of perfect tempus; consequently, proportions can only alter one level of mensuration in the moment, typically
the semibreve, but one must always consider the context of the proportion within the music itself.
88
The combinatorial sign C2 presents us with a good example of the ambiguity Tinctoris hoped to solve.
Some theorists during this time still explained C2 in the older fashion wherein it did not specify modus (before
Hothby); other, more modern theorists, included imperfect modus as part of C2 denotation. While the difference
between the two is arguably negligible, imperfect modus only tells us about the nature of longs and cannot tell us
about the regular mensural groupings of the breves. Furthermore, longs are by default imperfect, and so the signs
seem unnecessary, unless distinguishing from a mensuration where the longs are perfect, such as O2. Ultimately, the
system was messy, ambiguous, and full of unnecessary redundancy, all of which Tinctoris hoped to do away with.
The confidence Tinctoris had for his reforms and the agitation caused to him by the theoretical sloppiness of his
contemporaries’ use and understanding of mensuration and proportion signs was such that he critiqued the masters
of his day, even addressing the “inexcusable errors” in Ockeghem’s notation. Tinctoris comments in his
Proportionale musices on Ockeghem’s implementation of the sign O3 in L’autre d’antan: “Thus is refuted the
inexcusable error of Ockeghem, who signed his bucolic song L’autre d’antan, which is composed in equal numbers

115
logic, Tinctoris’ system did not replace Hothby’s combinatorial mensural-proportion chimera,

many composers continuing to notate proportions with individual numbers, use figures

combining signs and numbers, and treat mensuration and proportion as generally the same

phenomenon throughout the 15th century and beyond.89

While Tinctoris’ proportional reforms had very little impact on music broadly, most

traditional mensurations maintaining predominance with little use of numerical proportion

outside those in common use, a small genre of music did emerge during this period, a genre

which employed proportions extensively to create bicinia and tricinia that explore the rich

variety inherent in this new system. While many two-parted pedagogical examples exist that

demonstrate to greater and lesser extents the various possibilities of purely proportional music,

these “mensural inventions,” as I will anachronistically call them, seem to function as practical

music in addition to whatever pedagogical function they potentially could serve. These mensural

inventions seem to function practically based on their sources, the general purpose of those

sources, the other pieces within those sources, and the compass of the inventions themselves (as

compared to typical pedagogical examples).

in all parts, not only with a sign of proportion, but one that is interpreted on its own, and badly, as triple by some
and as sequialtera by others… When the unskilled have the preceding song, that is L’autre d’antan, or another that
is similarly signed, they say immediately, ‘let us sing – it is sesquialtera.’ O childish ignorance to attribute a
proportion of inequality to equality! Nor do I believe the composer wished to say that, although according to some
he signed in this way, but that his song should be sung like an excited sesquialtera. To accomplish this, a stroke
drawn through the middle of a circle of each part would have sufficed. For it is proper to it [the stroke] to signify
acceleration of the mensura, whether the tempus is perfect or imperfect, as appears in innumerable compositions of
his, of which the form in both [types of tempus] is [cut-circle and cut-C].” See Tinctoris, Proportionale musices,
book 1, ch. 3 (“Divisio proportionum”), 14-15. “Ex quo confunditur inexcusabilis error Okeghem, qui suum carmen
bucolicum L’autre dantan ab omni parte numeris aequalibus compositum nedum signo proportionis, sed illo qui a
quibusdam triplae, ab aliis sesquialterae per se et male attribuitur signavit … Dum vero carmen praemissum, scilicet
L’autre dantan aut aliud similiter signatum habent imperiti dicunt repente canamus sesquialtera est. O puerilis
ignorantia acqualitatis proportionem inaequalitatis asserere~ Nec existimo compositorem, quamvis ita secundum
aliquos signaverit, ita dici voluisse, sed ut carmen suum concitae instar sesquialterae cantaretur. Ad quod
efficiendum virgula per medium circuli cuiusque partis traducta sufficiebat. Nam proprium est ei mensurae
accelerationem significare sive tempus perfectum sive imperfectum sit, ut in infinitis etiam suis compostionibus
apparet, cuius in utroque forma talis est: [cut-circle and cut-C]."
89
Deford, Tactus. 125.

116
 I have identified two major sources for these rhythmically experimental pieces: The

Segovia Codex90 and an unnamed Venetian music treatise from 1506 penned by Johannes

Materanensis,91 though numerous other examples can be found sporadically from various 15th

and 16th century sources. Segovia’s odd rhythmic works appear towards the back of the

manuscript. We will examine works by Adam, Tinctoris, and Obrecht; these three bicinia from

Segovia represent the most extensive and adventurous use of Tinctoris’ proportional notation

within the manuscript. From the Materanensis Manuscript we have already examined one work

that uses Hothby’s combinatorial system – the Ave maris stella by Johannes Stockem – but we

will examine here one of the anonymous untexted works that use Tinctoris’ purely proportional

mensural notation.92

Adam’s De tous biens playne (see Example 3-3) demonstrates a rudimentary use of

Tinctoris’ system, breaking the requirement that all notated proportions be between at least two

numbers; however, Adam does utilize an interesting clarification, which Tinctoris himself does

encourage as an alternative to numerical ratios: orthographic representation. Despite Adam’s

90
Archivo Capitular de la Catedral, Segovia, Spain (E-SE), Ms. s. s. “Segovia Codex; Cancionero musical
de Segovia; Cancionero del Alcázar”
91
Biblioteca Comunale Augusta, Perugia, Italy (I-PEc), MS 1013 (M.36). Scribal inscription found on f.
45v. This collection, along with containing numerous treatises on music also possess many polyphonic pieces apart
from the theoretical works. Along with the numerous mensural inventions, there are two Agnus Deis, two hymns,
one French secular piece, one piece with the Latin motto "Difficiles alios,” 4 canons (added later), and 61 textless
pieces, which have been identified from concordant sources and include seven Mass Ordinary sections, four Mass
Proper sections, four motets, eight French secular pieces, and three Italian secular pieces. Well-known composers
represented in this source include Alexander Agricola (one piece), Ghiselin (one), Busnois (one), Dufay (three),
Faugues (three), Roellrin (one), Stokhem (one), and Tinctoris (25) along with 31 anonymous works. Among many
of the unidentified and anonymous textless works are the rhythmically intriguing mensural inventions, all of which
have been totally uncatalogued, unstudied, and un-transcribed; DIAMM does not even list these untexted
anonymous inventions in the inventory of this manuscript, despite their location outside theoretical works and
amidst the larger known works and composers listed above.
92
More extensive work needs to be taken up on the untexted pieces from this manuscript, all of which have
gone totally ignored, likely due to their anonymous and untexted nature, attracting little attention from a field that
has generally been more interested in large, clearly functional works by famous composers. I wonder what bizarre
creatures wait in this collection for a musicologist to reveal to us someday. This one example herein is all I can offer
given the limited scope of this thesis.

117
perhaps vague use of singular numbers, he does provide the written-out form of his intended

ratios. It is worth noting that while Adam’s work is primarily modulating between various

permutations of hexadic hierarchies he does employ one use of pentadic in m. 9.

Tinctoris’ bicinium on Le Souveniri (see Example 3-4) is more developed and

multifarious than Adam’s. Tinctoris more rigorously applies his notational schema, albeit the use

of a singular 3 in m. 8 and the combinatorial mensuration O2 in m. 14. However, Tinctoris’ use

of 3 and O2 does align with contemporary notational conventions, both signs well understood

and widely used. Worth noting is Tinctoris’ use of a heptadic hierarchy in m. 13; however,

everywhere else he has used some permutation of hexadic mensuration if not simply dyadic or

triadic.

Obrecht’s Regina caeli (see Example 3-5) is the most extensively developed of the three

selected bicinia from the Segovia Manuscript, as well as the most elaborate use of proportional

mensural notation from this source. While most of the cumulative proportions result in similar

rhythmic complexes, the piece does present twelve different proportions, seven of which are

three-termed ratios and one of which is a four-termed ratio. Furthermore, Obrecht has also

supplied orthographic representations of many of these ratios, though many of the orthographies

need to be interpreted in light of the ratios given and preceding.93 Despite some of the ambiguity

consequent in Obrecht’s perhaps unnecessarily detailed numerical ratios, the orthographic

93
The interpretation of this work is somewhat uncertain, since Obrecht’s use of ratios is not precisely what
Tinctoris would have intended. It seems Obrecht is perhaps providing more information than is necessary, though
information that is not incorrectly given and which can be reduced to a simple ratio as Tinctoris would have
intended. Such single ratios are given by Obrecht as the orthographic representations; however, much is left to
speculate about all the extra information given in the numerical ratios. For example, [Link] is given alongside
quinduple sesquitertia. The ratio quinduple sesquitertia seems likely referential to 9:3 generally, since sesquitertia is
4:3 and the quinduple modification tells us to add 5 to the first number of sesquitertia ratio, which makes 9:3;
however, 12:9 is itself the third multiple of sesquitertia 4:3 and 16:12 is the square of sesquitertia. It is unsure what
to make of such four-termed ratios. Ultimately, given expectations of counterpoint and some basic observations of
fractional reduction and the orthographic ratios given, the transcribed result seems to likely be correct. However, we
are still left to wonder what Obrecht’s intent might have been in providing such ratios as he did.

118
representations of his proportions are rather clear and the results, when considered with the

resultant counterpoint, seem to be reasonable. This work uses pentadic hierarchies twice in two

disparate locations (unique to this piece within this collection) – pentadic at the end of the third

system on page one and fourth system on page 2 – apart from the various permutations of dyadic,

triadic, and hexadic hierarchies; furthermore, this piece provides the greatest musical

development of its rhythmic materials overall.

Beside the Segovia Manuscript’s bicinia, the anonymous duos in the Materanensis

Manuscript are the most extensive source, barring purely theoretical treatises, of rhythmically

experimental works utilizing Tinctoris’ system. The anonymous Le serviteur (see Example 3-6)

is but one example of numerous relatively extensive treatments of proportional writing from this

source, the pieces within which use dyadic, triadic, pentadic, hexadic, and heptadic hierarchies

throughout, all couched within a tidy presentation of Tinctoris’ rational notation.

Ultimately, however, Tinctoris’ system was largely useless, since nearly all music written

during this time could be understood using the mensuration signs established prior to and up

through John Hothby. As demonstrated by Stockem, any music written using dyadic, triadic, or

hexadic hierarchies could be accomplished using some combination of coloration, diminution, or

mensuration sign; only when one wished to invoke other prime hierarchies or multiples of those

other prime hierarchies not well expressed in traditional mensurations was Tinctoris’ purely

rational system clear and useful. However, a thoroughgoing implementation of this system with

all its vast possibilities was still on the horizon.

Following a generation after Tinctoris’ proposed numerical modifications, Franchinus

Gaffurius was also a successful theorist and composer of the late-15th and early-16th century,

holding the prestigious post of maestro di cappella at the cathedral of Milan for most of his life.

119
Along with an extensive body of compositional work, Gaffurius was an erudite Humanist

scholar, writing expansively on speculative and practical music theory. Throughout the 16th

century, Gaffurius’ Practica musice (1496),94 which was greatly indebted to Tinctoris’ various

treatises including Proportionale musices,95 was regarded by theorists and composers alike as the

most authoritative and complete source of information on mensural music, practical and

speculative. Indeed, as we will see, it is Gaffurius we must thank for our conception of time

signature, herein and conventionally, and the conception of a “fundamental time unit” for

proportional derivation. Gaffurius’ theory of rhythmic notation would take the scaffolding

Tinctoris erected and add the final facade of clarity and consistency.

The glaring issue with Tinctoris’ notational theory, one which is evidenced by the

overabundant information given in Obrecht’s Regina caeli, is that the performer cannot know

with certainty what notes are being set in proportion. For example, does 9:4 mean there are 9

semibreves in the time of 4 previously in the same voice or simultaneously in an accompanying

voice; or does 9:4 compare semibreves or does it compare breves, minims, semiminism or some

heterogeneous combination? In contrast to Tinctoris, Gaffurius does discuss the issue of note-

value comparisons in a proportion, giving a clear system with 194 examples that contain only 5

exceptions to his rule, an exception to which he gives an easy reply:

94
Gaffurius, Practica musice.
95
In fact, Gaffurius spent 1478 to 1480 in Naples where he discussed music theory directly and regularly
with Tinctoris, writing a smaller manuscript treatise (Musices practicabilis libellum) as a result of these dialogues.
This treatise would later be revised and become Book 2 of Practica musice. Consequently, Gaffurius’ views on
mensural demarcation agree largely with Tinctoris’; however, Gaffurius’ requirements are, somewhat like Hothby’s
compromises with tradition, less rigid, making allowances for some traditional practices that Tinctoris rejected. See
Franchino Gaffurio, Musices practicabilis libellum (Harvard University, Houghton Library, Ms. Mus 142). This
preliminary treatise also appeared later, but before Practica musice, in an Italian edition by Gaffurius’ student
Francesco Caza. See Francesco Caza, Tractato vulagre de canto figurato (Milan: G. P. de Lomazzo, 1492; repr.
Berlin: M. Breslauer, 1922). See also DeFord, Tactus. 18-19.

120
 Meanwhile, most agreeable singer, do not be alarmed if you find an incomplete number
of notes in the arrangement of some proportion, for generally what we do not show in
semibreves you can examine in minims.96

This seemingly trivial fact that Gaffurius made such a clear point to prescribe the semibreve as

the compared note-value allows us to conclude that Gaffurius, standing on the metaphorical

shoulders of Tinctoris, formulated a significant step away from purely mensural notation,

approaching our modern system of time signatures, wherein fractions, via time signatures, are

always referential to some fundamental note-value, by convention the whole-note, which

interestingly is the equivalent of the semibreve.

In the 14th and early-15th century, proportions, when used either numerically or

symbolically, replaced mensuration signs. During this earlier period, micro-level non-mensurally

determined rhythmic proportions were indicated through other means such as coloration, Italian

note-shapes, additional mensuration signs, and fractions. Even though Johannes de Muris in

1321 codified a division of the breve into anywhere from two to nine equal parts, theorists and

composers used only dyadic and triadic proportions representable by combinations of

mensuration signs assuming breve equivalence,97 other prime proportional factorizations unused

simply because musicians lacked appropriate indicative signs. The introduction of the numerical

fraction showing rhythmic proportions consequently presented a pivotal innovation, permitting

96
‘Quiescas interim iocundissime cantor si in alicuius proportionis dispositione imperfectum reperieris
notularum numerum, plerumque cnim quern non disponitnus in semibrevibus consyderabis in minimis’; Practica
musicae, bk. 4, ch. 6; trans. Miller, 194. Simply put, Gaffurius explains that in the majority of cases the semibreve is
the value of proportional transformation; however, in very rare cases, if the semibreve does not seem work with the
given proportion, the minim level should be considered alternatively. All five exceptions Gaffurius gives are
sesquialtera proportions. It is also worth noting that unlike many of his predecessors, Gaffurius uses larger fraction
multiples, like 9:3 or 10:2, in clear and distinct ways as compared with their reduced forms 3:1 and 5:1. When
Gaffurius writes 10:2, he means precisely this and not 5:1, calling for ten semibreves in the cantus in the time of two
in the tenor. See Practica musicae, bk. 4. ch. 3.
97
O:C or C-dot:C representing 3:2, backwards-C:O or backwards-C:C-dot representing 4:3; O-dot:C
representing 9:4; O-dot:cut-C representing 9:8; C:cut-C representing 2:1; cut-C:C-dot representing 8:3

121
proportion prescriptions not normally inherent in the traditional mensural system. However, until

the late-15th century, fractions were employed as mensuration signs, not cumulative and

determining the mensuration themselves. It took until the late 15 th century for Johannes Tinctoris

and Franchinus Gaffurius to emancipate proportions from mensuration signs and use fractions in

an arithmetically logical fashion.98

Tinctoris and Gaffurius, despite generally being known now and then as the most

prominent theorists of their time, codifying and commenting on much of standard contemporary

musical practice, were not describing mainstream musical theory in their reform of proportional

rhythmic notation, which attempted to simplify the complexity and confusion of the traditional

mensural system. Despite lacking mainstream appeal in the late-15th century, their proportional

reforms were relatively successful in the 16th century.

Towards the Elizabethan Avant-garde: Mensuration in the 16th Century

More so than either the 14th or 15th century, mensural theory was drastically simplified in

the 16th century. While within the 16th century we will ultimately witness the most complex use

of proportional notation yet discussed, rhythmic diversity would totally disappear during the 16th

century, not to reemerge until the early-20th century. The gap between mensural theory and

practice widened in the 16th century, so much so that most theoretical discussion of mensuration

and rhythmic proportions was wholly speculative or only relevant in so much as someone was

performing music from the 15th century. While the principles of mensural notation fully

98
It is interesting to note that another reason these signs became gradually less popular was the invention of
printed music and the capitalist forces behind that innovation: composers and publishers wanted to reach larger
audiences and market their publications for broader consumption. This broader, less-affluent sponsorship of the
growing middleclass hobbyist musician, learning and playing music in the growing leisure time now afforded
unprecedentedly to many Europeans, was naturally not as well-educated as professional singers sponsored by
wealthy churches and courts. The result of appealing to more “common” musician leisure practitioners was a
necessary simplification of the mensural system. See Berger, Mensuration, 230.

122
developed by the end of the 15th century, changes in musical taste and the development of what

we might call “the music industry,” via printed music, led to a migration further away from

speculative music towards more practical aesthetics, so much so that by about 1520, the only

signs commonly used were cut-C and signs of sequialtera or triple proportion, other 15th century

signs remaining permanently obsolete.99

Unquestionably, Italian music theory during the second half of the 16 th century

dominated Europe; central to the Italian school was Gioseffo Zarlino and his Le istitutioni

harmoniche.100 Zarlino still stands as one of the most influential music theorists of all time.

Interestingly and important to note, Zarlino had little interest in mensuration and theories of

proportional rhythm, even carelessly allowing inconsistencies in his discussions of mensural

issues.101 While the few observations he makes on the subject are of considerable value in

regards to standard rhythmic practice, his terse treatment of rhythm and mensuration and his

complete disregard of diverse and exceptional proportionalities only point to the growing trend

towards the vastly simplified rhythms of the Common Practice Period.102

This trend of simplification continued so that by the middle of the 16th century, many

theorists began to teach that proportions, barring the common tripla and sesquialtera, had no

practical or legitimate function, since the purpose of musical notation was “to represent audible

99
It is interesting to note that music theorist Spataro in a letter to Giovanni del Lago from January 4, 1529
observed and lamented this growing obsolescence of mensural diversity. See A Correspondence of Renaissance
Musicians, 336.
100
Gioseffo Zarlino, Le istitutioni harmoniche (Venice, 1558; repr. New York: Broude Bros., 1965).
101
DeFord, Tactus. 28-29.
102
For more information on Zarlio’s discussion of rhythm as well as his other theories of harmony see also
Giovanni Maria Artusi and Orazio Tigrini simplified version of Zarlino’s theory. See Giovanni Maria Artusi, L’arte
del contraponto (Venice: Giacomo Vincenzi & Riciardo Amadino, 1586; repr. Bologna: Forni, 1980); Orazio
Tigrini, Il compendio della musica (venice: ricciardo Amadino, 1588; repr. New York: Broude, 1966).

123
sounds, not mathematical abstractions.”103 Glarean, another giant of music theory during the 16th

century, was one of the first to ardently express this attitude. To this end Glarean writes:

Art should be considered as art is. Furthermore, the circumstance itself now shows that it
is superfluous to observe so many proportions… because there is more labor in learning
such proportions than sweetness or grace in signing of them.104

Glarean makes it clear in his monumental Dodecachordon that he believed musical proportions

functioned more as a display of erudition and ostentation than artistry. Perhaps unfortunately, we

will continue to see such close-mindedness permeate all major music theory discourses across

the European continent and British Isles throughout the 16th century, persisting into the early-17th

century wherein all discussion of exceptional rhythmic subdivision and hierarchical structure

would totally die in musical discourse until the 20th century.

The theorists Coclico and Vincentino105 agree with Glarean’s indictment of Tinctorian

proportional theory wholeheartedly in their respective texts: Compendium musices and the

enormously famous L’antica musica ridotta alla moderna prattica.106 North German music

theorist and composer Joachim Burmeister also declines to discuss proportions, only offering a

brief discussion of triple mensuration in all voices, because, as Burmeister himself puts it, “it is

the height of insanity to complicate a song that can be written in simple notes with obscure

103
Deford, Tactus. 178.
104
Glarean, Heinrich, and Clement A. Miller. Dodecachordon. [Dallas, Tex.]: American Institute of
Musicology, 1965. book 3. ch. 12 (“De proportionibus musicis”), 227. “Ars, ut ars est, tradi debet. At res ipsa nunc
clamat, superfluum esse tot proportionum observationes… ut in quibus maior labor in addiscendo quam suavitas
gratiave in cantando esse constet.”

Most famous today as the “father of microtonal music,” his theories of temperament and designs for
105

keyboards with 24, 36, and 48 keys to the octave are still fascinating and inspiring musicologists, theorists, and
composers alike even in the 21st century.
106
Coclico, Compendium musices, part II, “De prolationibus usitatis,” fol. Gijr; Vincentino, L’antica
musica ridotta alla moderna prattica, book 4, ch. 31 (“Delle proportioni musicali, che a questi tempi da Prattici
della musica son’ usate”), fol. 87v.

124
signs.”107 The German composer-theorist Praetorius explains proportions with binary

mensurations only reluctantly in his Syntagma musicum, explaining that, while some composers

still use them, such proportions are of no use, and claiming that musicians are often “confused,

obstructed, and altogether confounded” by them.108 Deford aptly summarizes this period of

rhythmic simplification and diverse proportional abandonment, pointing out that “the reversal of

attitudes since the time of Gaffurio, for whom mastery of proportions was the height of musical

erudition, [was] complete.”109

Nevertheless, some theorists continued to discuss old signs, colors, and proportions

throughout the 16th century, both to aid the interpretation of still performed music from the

Josquin generation and to maintain traditional components of music pedagogy regardless of

utility. Ultimately the fashion of erudite, if not arcane, theory faded by the beginning of the 17 th

century, but not before one final and bizarre reimagining of Tinctorian proportional notation in

the Elizabethan court at the end of the 16th century before the unchallenged ascendancy of our

modern dyadic rational system of rhythmic demarcation.

During the 16th century, theories of proportional music retained the intermingling of

practical and speculative elements that the theory possessed in the 15th century. While some

theorists detail the mathematical foundations the theory, describing far more ratios than ever

used in practice, most theorists limited their discussion only to practical proportions; towards the

latter half of the century, some dispense with the topic altogether.

107
“Summa dementia est, cantionem quae simplicibus notis scribi potestm obscuris signis perplexare.”
Burmeister, Musica autoschediastike, Accessio III, section II, “De antiphonis,” fol. Aa2v.
108
“Nicht allein die Jugend in den Schulen, sondern auch offtmahls geuebt Musici Vocales &
Instrumentales in Capellen perturbirt, remorirt, auch wol gar confundirt werden.” Praetorius, Syntagma musicum,
vol. III, part II, ch. 7 (“De tactu, seu notarum mensura; (italis battuta) & signis“), 54.
109
Deford, Tactus. 179.

125
 The proportions most often shown to be useful in practice were the same as had always

been since the early-14th century: duple (2:1), triple (3:1), quadruple (4:1), sesquialtera (3:2),

and sesquitertia (4:3). However, the basic reforms of Tinctoris and Gaffurius had persisted in

that the symbols that theorists proposed to represent these ratios were most often ratios of two

numbers, which left little chance of confusion. Though, of course, some composers and scribes

still represented common proportions by the Hothbian chimeric combination of different

mensuration signs or special symbols.110

16th century English theoretical treatises that touch on mensuration are far more limited in

number than their continental counterparts though their treatment of proportions and

mensurations is far more extensive by comparison. This handful of manuscripts includes John

Dygon’s commentary on Gaffurian proportional theory,111 along with four printed texts from the

last two decades of the century during the reign of Elizabeth I: William Bathe’s A Brief

Introduction to the True Art of Music,112 Bathe’s later A Brief Introduction to the Skill of Song,113

the anonymous Pathway to Musicke,114 and Thomas Morley’s A Plaine and Easie Introduction to

Practicall Musicke,115 which is, by far, the most substantial and celebrated of the five. These

British sources depend heavily on continental theories of mensuration, but do demonstrate

interesting differences, detailing a distinctively English notational practice, a practice which was

110
Ibid., 169.
111
Dygon, John, Theodor Dumitrescu, and Franchinus Gaffurius. John Dygon's Proportiones practicabiles
secundum Gaffurium = Practical proportions according to Gaffurius. Urbana: University of Illinois Press, 2006.
112
Dygon, John, Theodor Dumitrescu, and Franchinus Gaffurius. John Dygon's Proportiones practicabiles
secundum Gaffurium = Practical proportions according to Gaffurius. Urbana: University of Illinois Press, 2006.
113
William Bathe, A Briefe Introduction to the Skill of Song (London: Thomas East, [1596?]; repr.
Kilkenny, Ireland: Boethius Press, 1982).
114
Anonymous, The Pathway to Musicke (London: [by J. Danter] for William Barley, 1596).
115
Morley, Thomas. A Plaine and Easie Introduction to Practicall Musicke. Farnborough: Gregg, 1971.

126
far more concerned with proportional possibilities in rhythm than contemporary continental

composers and theorists generally.116

Dygon to Baldwin: The Gaffurian Legacy in Britain and the Elizabethan Rhythmic Avant-
garde

Early English music theory’s survival to the present is notably remarkable; sources from

the Tudor period are sparse as a consequence of many purges of religious institutions – medieval

storehouses of knowledge - during this period. For investigators, such paucity of surviving

sources makes tracing any insular systems underlying English polyphony during the early-16th

century difficult.117 Three surviving documents stand as the primary sources of music theory

from this period, one of which, unfortunately, consists largely of copies from one of the other

two and thus has little independent worth.118 Consequently, as musicologist Theodor Dumitrescu

notes, the study of English theory from the late Renaissance is “an endeavor requiring a

considerable measure of imaginative reconstruction,” a reconstruction which is primarily based

on sources from proximal periods.119 To date, only a fraction of early-Tudor theoretical material

is published in modern editions or translations.120

116
DeFord, Tactus. 31-32.
117
Roger Bray, “Music and Musicians in Tudor England: Sources, Composition Theory and Performance,”
in The Blackwell History of Music in Britain: The Sixteenth Century, ed. Roger Bray (Oxford: Blackwell, 1995), 3-
6.
118
Cambridge, Trinity College (GB-Ctc), 0.3.38; London, British Library (GB-Lbl), Additional 10336; and
London, Lambeth Palace (GB-Llp), 466, which itself contains mostly a copy of London Additional 10336.
119
Dygon, Proportiones practicabiles, 1.
120
See Ronald Woodley, John Tucke: A Case Study in Early Tudor Music Theory (Oxford: Clarendon
Press; New York: Oxford University Press, 1993). Woodley’s investigation of this notebook includes translations
and transcriptions from GB-Lbl Add. 10336, many parts of which were also recopied into GB-Llp 466. See also
Dygon, Proportiones practicabiles. Dumitrescu’s volume offers a mensural transcription and modern translation of
Trinity College’s manuscript 0.3.38 and Cambridge (GB-Ctc), both of which are companion English theory books
from the first half of the 16th century. This two-volume theory text contains closely related texts on arithmetic
proportions and their application in rhythmic notation of polyphony.

127
 While rhythmic complexity and mensural music was generally on the decline on the

Continent during the 16th century, England witnessed during the final decades of that century the

flowering of the most complex rhythmic music yet notated in European sources. The rhythmic

complexity of this music was a direct result of the dissemination of proportional rhythmic

theories from the Continent to the Isles, primarily through the work of John Dygon (fl. 1497 –

1538, aka Wyldebore) during the early-16th century and, to a lesser extent, Thomas Morley

(1557/58 – 1602) later in that century; however, there is a clear link in the 16th century notational

theories of the English to the works of John Hothby from the mid-15th century as well, giving

British rhythmic notational theory both the proportional, albeit chimeric, impetus of Hothby

along with the purely rational system of Gaffurius via Tinctoris. In fact, we will come to see

English composer’s use of three rhythmic notational systems simultaneously: chimeric Hothbian-

like symbols in the manner of Dygon’s theory, Tincto-Gaffurian numeric proportions, and

orthographic proportions.

This period of proportional music’s flowering in England we will call the “Elizabethan

rhythmic avant-garde,” for, without question, the time from the reign of King Henry VIII

through the reign of Queen Elizabeth I witnessed the most complex rhythmic music ever notated

in Europe to that point and up until the 20th century. While there are many composers and pieces

well studied from this period – Byrd, Tallis, Bull, Dowland, Taverner, and Tomkins to name a

few composers – we can recognize that any period of popular trends possesses within itself,

whether known or unknown, those migrating between such trends and other aesthetic spheres.

The experimentations within this period, the arguable school of composers conceiving these

rhythmic oddities, and the general genre that has arguably arisen from the coalescence of such

rhythmically exceptional works across English sources from the second half of the 16 th century

128
has interestingly gone nearly unstudied in modern scholarship; often such musical peculiarities

are disregarded as mere pedagogical exercises,121 and so there is little present scholarship on this

121
For example, every modern musicologist that discusses The Baldwin Commonplace Book either fails to
mention the rather large assortment of rhythmically bizarre works in this tome’s second half or simply passes over
them with little more than an aside to some ‘pedagogical pieces’ that likely were used only didactically. The only
modern musician – not musicologist – that takes any time to recognize the astounding character of these exceptional
pieces is Kees Boeke, a Dutch recorder player and composer, in his performance edition of selections from the
Baldwin Commonplace Book. Dumitrescu says myopically of these pieces, which one should be able to clearly see –
if one is familiar with the literature – have a significant bearing on his own work on Dygon’s treatise, that they are
“clearly included and grouped together as examples of unusual and extreme proportional writing, perhaps with some
didactic intent;” nothing more of these phenomenal works is said, though they are likely a consequence of Dygon’s
theory, even if a modification of it (see Dygon and Dumitrescu, Proportiones practicabiles secundum Gaffurium. p.
53). The facsimile edition of the Baldwin Commonplace Book says of the second section of the manuscript that it
“consists primarily of didactic pieces for two voices, copied as separate parts on facing pages, [concluding] with
Byrd's three-voice mass…. A number of pieces are exercises for teaching proportions, while others provide practice
with the hexachord” (see Owens, Jessie Ann, and John Baldwin. London, British Library, R.M. 24.d.2. New York:
Garland Pub, 1987. p. vi). Again, those works which perhaps distinguish Baldwin’s book the most, pieces
unparalleled in rhythmic complexity and development of that complexity until the 20 th century, pieces that if
didactic are some of the longest and most demanding pedagogical examples ever handed to a student for study
(something I obviously think unlikely), are passed over.
The case of Baldwin’s collections of strange rhythmic works is not unique in modern scholarship. In every
instance I have found, every musicologist overlooks rhythmically atypical pieces and disregards them as didactic,
even if the source containing them and the pieces surrounding them would suggest otherwise. See for another
example DIAMM’s assessment of Biblioteca Comunale, Augusta Perugia, Italy [I-PEc], MS 1013 [M.36], an
assessment which does not even bother to list in its inventory or discuss in its description nearly all of the atypical
mensural bicinia contained therein, again ignoring them on the basis of these pieces being ‘pedagogical’ exercises
only – why the potential that something was written as an etude should matter to the point of selected ignorance in
any case, I am unsure. Shall we ignore Chopin’s etudes as well, since they might, in part, have some pedagogical
purpose? I suggest that this disregard for the atypical is rather myopic, especially in light of previous shortsighted
failures in the world of early musicology to disregard works, composers, and styles that seem to break the mold of
what we believe appropriate for the period (I think here of August Ambrose’s famous assessment of Alexander
Agricola in the 19th century, an assessment which damned Agricola to absolute obscurity until the late 20th century
and even arguably to the present despite Agricola’s renown during his life as “second only to Josquin,” according to
the similarly preeminent Johannes Tinctoris).
Shall it only be the composers and performers of early music that can see the artistic intent and merit of
these clearly non-didactic works (or, at least, works whose pedagogical intent is in addition to loftier goals)? If these
works, many of which are quite substantial in scope, be only didactic, then we must begin to assume many early
works lauded as masterpieces as also merely for teaching purposes, either due to their physical proximity in
manuscripts to “pedagogical works,” location in texts that also devote part of their substance to theoretical matters,
or their “impractical and speculative” complexity. Shall we assume the ars subtilior was a generation only
concerned with rhythmic pedagogy? Does Brian Ferneyhough only write music for the erudition of students
studying complex rhythm? Are Conlon Nancarrow’s piano rolls only meant as “experiments” without artistic merit?
Obviously, the answer to all these rhetorical questions is ‘no.’
Though, I do need to pay deference to the field of musicology, from which we gain so much, and duly
recognize such attitudes are not only the prerogative of musicologists; many fields fail to see exceptionalisms as
revolutionary. Here I think upon the history of the Antikythera Mechanism, originally disregarded as yet another
religious trinket from Ancient Greece and consequently consigned to molder unexamined on museum shelves until
British physicist and science historian Derek John de Solla Price recognized that this “religious object,” given its
obviously apparent gear-like parts, was likely a mechanical computer. It is now well-known, due to a more open-
minded and cross-disciplinary reevaluation, that the Antikythera Mechanism was, unquestionably, a mechanical
computer used to calculate astronomical phenomenon, a revelation which has utterly changed our understanding of
Greek science in the Classical World. Yet again, it took someone outside a field, complacent with its typical

129
subject other than the music itself, the most extensive and varied English collection of which

comes from the Baldwin Commonplace Book.122 It is during this period we will find the most

extensive use on non-dyadic, -triadic, and –hexadic hierarchies: pieces which are themselves

entirely in five or other prime groupings, use proportions never before used in any extant

European source, and use collections of proportions that are themselves unprecedentedly

multifarious.123

Given the scarcity of theoretical sources from this place and time, it is difficult to say

with total certainty what inspired composers at the end of the century to create rhythmically

exceptional works. We cannot know unquestionably the impetus for these experiments, should

we assume they did not simply spontaneously emerge in the minds of Elizabethan composers. If

we attempt to draw some line between “Tincto-Gaffurian” proportional theory on the continent

from the late-15th and early-16th century to the practice of proportional music in Brittain during

the 16th century, the best link we have is John Dygon’s commentary on Gaffurian proportional

theory, Proportiones practicabiles secundum Gaffurium, penned likely at Oxford sometime

during the first half of the 16th century.

Gaffurius offers a significantly more thorough introduction to proportional music than

Dygon, the former systemically and comprehensively proceeding through proportional species in

a logical and increasingly complex order (multiplex and submultiplex, superparticulare and

submultiplex superparticulare, and subsuperparticulare, superpartiens and subsuperpartiens,

expectations, to see the exceptionalism in something atypical and thoroughly bring the significance of that
exceptionalism to light. Such examples as these abound; we might learn from them.
122
Baldwin, John. Commonplace Book: Anthology of Miscellaneous Music Including Textless Madrigals
and Motets, Masses, Songs, Canons and Excercises by English and Italian Composers. 1591.

There are even examples of pieces using new kinds of coloration, such as John Ashwell’s Missa Ave
123

Maria. See Woodley, Ronald. John Tucke.

130
multiplex superparticular, and multiplex superpartiens and submultiplex superpartiens) while

also touching upon the abstract mathematical properties of the same; Dygon’s definitions of

these proportional genre are quite brief by comparison, presenting only the essential attributes of

each genus and giving a more limited set of musical examples, though still a great number

relative to other sources on the same subject.124 Of Dygon’s departures from Gaffurius’ treatise

of proportional music, the most significant are the large collection of newly composed examples

in GB-Ctc 0.3.38 that both replace and augment Gaffurius’ own, and the new system of mensural

demarcation in the second part of the treatise. Of the 82 examples in Dygon’s manuscript, only

two were copied from Practica musice, and even these were given some modification.125

Dygon’s treatise offers some interesting extensions of Gaffurius’ speculative music;

while Dygon not only provides many elaborations and extensions of Gaffurius’ own examples,

Dygon also provides newly constructed two-part examples along with seven three-part examples,

some of which contain three differing proportions simultaneously (something not done by

Gaffurius). Those three-part examples that include all parts in differing proportions include those

on folio 4r with 7:1 against 3:2 against 6:1, folio 4v with 3:2 against 9:2 against 4:1, and folio 6r

with 3:2 against 9:2 against 5:2. Among his two-parted examples, Dygon has exceptional

proportional relationships ranging from 7:6 (fol. 3v), 10:6 (5r), 9:7 (5v), 7:4 (5v), 8:5 (5v), 10:7

(6r), 9:5 (6r), 11:7 (6v), 13:9 (6v), 11:6 (6v), 13:7 (7r), 11:5 (7v), 13:4 (8r), 17:4 (8v), 13:5 (10r),

17:5 (10v), and 19:4 (12r) among others and various multiples of these given here.

Furthermore, there exists in the first treatise a type of example wholly absent in

Gaffurius’s treatise, a type that appears to be Dygon’s own invention: three-part mensural

124
Dygon, Proportiones practicabiles, 33.
125
Ibid., 38.

131
tricinia. Of the 59 compositions in Dygon’s first treatise, seven are three-voice works featuring

one Tenor and two Discantus parts. In these examples, Dygon uses two genres of multiplex

proportions in two voices to create, through their combination, the more complex proportions of

later genre (see Example 3-7 for three such tricinia examples from Dygon’s treatise).126 Even in

proportionally complex works from other English sources and elsewhere on the Continent, the

simultaneous employment of multiple proportions is exceptional; there are far fewer mensural

tricinia, especially those utilizing different proportions on all levels, than mensural bicinia of

even the most complex variety. However, as Dumitrescu notes, if these exceptional three-part

demonstrations in Dygon’s treatise were “intended to influence practical composition,” they

“appear not to have had great effect” in broader musical practice.127 Despite Dumitrescu’s

disparaging remarks, Dygon’s treatise does seem to have had some influence, or at least

correspondence, with some English musical practice from this century, given the sampling of

126
For example, in the first such case on folio 3r, the first Discantus sings under quintupla while the second
Discantus sings quadruple; the resultant rhythmic complex is sesquiquarta (5:4) of the superparticulare genus.
Here, it is interesting to note that Dygon is providing a simpler method to execute more difficult aspects of
proportional theory.
127
Ibid., p. 43. Again, to make a cross-disciplinary comparison to the point of edification, though the
Ancient Greeks had very few mechanical computers, the presence of at least one is nevertheless earth-shattering.
Though only a few works might show Dygon’s influence, the presence of even a few works of extreme rhythmic
complexity is, and should be considered, significant, even if those works are not ultimately influential themselves.
For a variety of reasons, though probably tied somewhat to the rise of Rome and their lack of interest in abstract
sciences by comparison to the Greeks, mechanical computers did not persist in Ancient Greece, but the fact that they
did not persist does not make their appearance something to ignore. In fact, their disappearance and lack of further
development begs questions of the age, questions which might have relevance to anti-intellectual movements
throughout history. It even makes one wonder what the world would be like had such devices persisted and evolved.
Wilhelm Sckikard (b. 1592) designed the first mechanical computer in 1623, followed shortly thereafter by Blaise
Pascal’s arithmetic machine in the 1640s; three-hundred years later in 1946, the first electronic, digital computer is
built. What if Greece had not been politically, economically, and scientifically desolated by Rome? What if Rome
had not killed Archimedes in 212 BCE? What if the Antikythera Mechanism had been developed further beyond the
mid-2nd century BCE? Might we have flown to the moon in 150 CE? Though it sounds absurd, Greece was on the
same trajectory in the 2nd century BCE Europe was on the 17th century CE. In many ways, the Renaissance in
Europe during the 16th century was a return to the period of Greece directly before that civilization’s decline in the
2nd century; Europeans restarted that same curious and scientific engine lost in the Western world for the past two
millennia. Following that course forward, Europe industrialized, nuclearized, computerized, and journeyed into
space in the mid-20th century. Rather than ignore strange and atypical inventions in history, it would be better to ask
why they failed and what things might be like if they had succeeded.

132
complex three-parted mensural tricinia in the Baldwin Commonplace Book, which we will

address shortly.

In addition to presenting the Tincto-Gaffurian proportional system to Britain through his

abbreviated reference list of proportional musical examples, Dygon also offers many such

examples alongside a new and unusual system of proportional notation that arithmetically

reinterprets the traditional mensural signs to re-represent their combinatorial use as stand-in for

numerical proportions. Dygon’s new proportional notation system essentially translates

mensuration signs (C, O, C-dot, O-dot, O-slash) into numerical values, which are then treated

arithmetically in combination and compared to other sets of similarly arithmetically combined

signs to collectively represent proportions not representable by the traditional use of the signs.

While no straightforward explanation of the system is outlined by Dygon, each combinatorial

sign is given with an example and a proportion, and thus the derivation of the combinations of

signs from the proportions is simple enough to deduce.128

While this system might seem redundant with the Tincto-Gaffurian numerical system as

well as needlessly arcane, Dygon’s notational method did have one additional capability:

different groups of signs could signify the same numerical proportion and consequently could

then be used to differentiate subdivisions on different hierarchical levels of the mensuration.

Dygon’s new system, though more cumbersome, allowed the composer to encode more

information in one symbol than the Hindu-Arabic rational system. Signs may be applied to

128
Dygon’s arithmetic system is relatively straightforward: each mensuration sign receives a numerical
value, dependent on its perfection or imperfection on the semibreve-breve and minim-semibreve levels. The values
of grouped signs are summed to a group value. For example, in the treatise there are two methods of notating dupla
sesquialtera (f. 16v): to notate the proportion at the semibreve level, Dygon gives either 5:2 or OC : C. To
deconstruct this combinatorial sign, we need to note that there are 3 semibreves per 1 breve in O, and there are 2
semibreves per 1 breve under in C. By converting the signs to numerical values, we obtain (O+C : C) = (3 + 2 : 2) =
(5:2). Similarly, at the minim level, the grouping is C-dot C : C. Thus, we note that in C-dot there are 3 minims per 1
semibreve, while in C there are 2 minims per 1 semibreve. Consequently, we have (C-dot + C : C) = (3 + 2 : 2) =
(5:2). See Dygon, Proportiones practicabiles, 50.

133
clearly signal a change at either the semibreve or minim level, whereas the numerical designation

typically only signaled changes on the semibreve level, with the minim level usually only

changeable either by written prescription or by context.129

In fact, Dygon was not unique in formulating such a notation theory. One other source

from the mid-16th century offers, according to Dumitrescu, “some corroboration for the system in

[Dygon’s second treatise],” and furthermore, is possibly a source for Dygon’s notational

theory.130 The Notebook of John Tucke (GB-Lbl Add. 10336), a somewhat atypical British

musical source from the early-16th century, contains many miscellaneously assembled musical

items and ideas along with various texts dealing with proportions in music, including not only

the “classical” treatise by John Hothby (Proportiones secundum ioannem otteby) but various

writings on purely arithmetical discussions of proportion, notes on the proportions represented

by unusual notational colors, and various, as well as (on f. 100r-v) a particularly relevant,

discussion on similar usages of combinatorial mensuration signs similar to those devised by

Dygon.131

While Dygon’s system in GB-Ctc 0.3.38 and John Tucke’s in GB-Lbl Add 10336132

(albeit in a more fragmentary fashion) both demonstrate a clear link to continental proportional

theories and expand on those theories in a uniquely British manner, the notational method in both

129
In which case, the composer helplessly relies on the careful and accurate interpretation of the performer
(or musicologist at a much later date) to realize the subtle implication of minim via contrapuntal necessity.
Dumitrescu notes that Dygon had already “displayed a concern for this distinction” at the first treatise’s outset,
wherein three of the first four examples have “per minimam” or “per semibrevem” written into the music. See
Dygon, Proportiones practicabiles, 51.
130
Dygon, Proportiones practicabiles, 54
131
See Woodley, John Tucke.
132
Also found in London, Lambeth Palace (GB-Llp), 466

134
treatises “by no means conforms to orthodox notational practice,” as Dumitrescu notes.

Dumitrescu’s concluding assessment of the source’s import is worth quoting at length:

Such an employment of mensuration signs, in combinations of up to seven symbols,

divorced from any effect on relative rhythmic divisions, is unknown in continental theory.
It must be noted, furthermore, that the idea is not encountered elsewhere in the English
theoretical tradition, which in its treatment of proportions was generally in line with
international practice.133

While surviving musical sources offer nothing to suggest that Dygon’s system found widespread

practical use, a set of compositions in late-16th century composer John Baldwin’s Commonplace

Book134 display something approaching the combinatorial mensuration signs in GB-Ctc O.3.38

and are unquestionably works that fully explore the potential of the Tincto-Gaffurian and

Dygonian system of proportional music through their extreme proportional writing. However,

despite these late-16th century proportional works within Baldwin’s collection and the theories of

Dygon, the remarks of contemporary British music theorist Thomas Morley in his treatise Plaine

and Easie Introduction to Practicall Musicke published in 1597, clearly indicate a persistent

disdain among theorists for complex proportional music. Though, against the admonishments of

such prominent Continental theorists like Zarlino and Glaranus as well as influential Brittish

theorists like Morley, some composers retained an interest in the subject of rhythmic diversity up

to the beginning of the 17th century.

Thomas Morley’s A Plaine and Easie Introduction to Practicall Musicke became the

centerpiece of English music theory and education for nearly the next two centuries. Other than

its wide-reaching influence and its relatively thorough codification of standard contemporary

musical practice, Morley’s treatise is rather unremarkable in terms of exceptional practice.

133
Dygon. Proportiones practicabiles, 56
134
GB-Lbl R.M. 24.d.2, ff. 100v-115r.

135
Morley himself openly admits to the seeming uselessness of most, if not all, of the Tincto-

Gaffurian and Dygonian proportional practice. Morley critiques many of Dygon’s theories

(particularly those of “induction,” which we did not discuss above), and notes that the teaching

of extended proportions is pointless, beyond a basic knowledge of the common proportions and

their notation using numerals. The combinatorial mensural notation theory of Dygon and Tucke

goes unacknowledged by Morley.

Morley’s voice of the master in the Socratic treatise offers but a trifling introduction to

more advanced and uncommon proportions with this note to the treatise’s allegorical pupil:

Quintupla and Sextupla I have not seen used by any stranger135 in their songs (so far as I
remember), but here we use them, but not as they use their other proportions…. but that
is more by custom than reason.136

After this scant remark on quintuple, Morley disregards the remainder of Gaffurian proportional

studies by reassuring the pupil that

…these [few proportions] shall suffice at this time, for knowing these the rest are easily
learned; but if a man would engulf himself to learn to sing and set down all them which
Franchinus Gafurius hath set down his book De Proportionibus Musicis, he should find it
a matter not only hard but almost impossible.137

Despite his aversion to instruct in advanced proportional studies, he does offer the student a

parting table, from which one can translate between orthographic and numerical proportions (see

Figure 3-1 below)

But if you think you would be curious in Proportions and exercise yourself in them at
your leisure, here is a Table where you may learn them at full.138

135
By “stranger,” Morley means “foreigner.”
136
Morley, A Plaine and Easie Introduction, 56.
137
Ibid.
138
Ibid., 57.

136
 However, Morley cannot help but continuously disparage anyone who might choose to

write using such complex rhythmic ratios, using both the voice of the master and the pupil to

question the legitimacy of composition under such aesthetics. Once presented with this table, the

student remarks:

Here is a Table indeed, containing more than ever I mean to beat my brains about. As for
music, the principal thing we seek in it is to delight the ear, which cannot so perfectly be
done in these hard proportions as otherwise, therefore proceed to the rest of your music,
specially to the example of those proportions which you promised before.”139

Figure 3-1. Thomas Morley, Table of Proportions from his Plaine and easie Introduction to
Practicall Musicke (London, 1597).140

139
Ibid., 58.
140
Image in public domain. See Morley, A Plaine and Easie Introduction to Practicall Musicke. It is
interesting to note that in one of the Appendices on his book, Morley explains his insertion of this chart. His
presentation of it was not for the compositional exploration of advanced proportions, but for the reference of
performers who cannot read Hindu-Arabic numerals, a problem we must thusly assume was still present in England

137
Herein, Morley clearly means to suggest that music composed under complex proportions as

given by Gaffurius cannot “delight the ear” and are too “hard” to bother “beating one’s brains

about” attempting to execute, either compositionally or performatively.141

Morley continues discouraging students from exploring more exceptional uses of

proportional notation in music in his annotations to the first part of his treatise, annotations

which he adds to the first edition of his book because “when [he] had ended [his] book and

shown it to some of better skill in letter… [he] was by them requested to give some contentment

to the learned, both by setting down a reason why [he] had disagreed from the opinion of others,

as also to explain something which in the book itself might seem obscure.” 142

In regard to proportions, Morley reprimands himself for having gone during the first part

of this treatise into such great extent to discuss some of the simpler extended proportions that he

did bother to discuss. To this end, Morley writes, “as for Sesquitertia, Sesquiquarta, and such

like, it were folly to make many words of them, seeing they be altogether out of use, and it is a

matter almost impossible to make sweet music in that kind.”143 Further claiming that discussing

proportions further would be a ‘waste of time,’ Morley writes that if “[he] should go about

[saying] all that may be spoken of the Proportions [he] might be accounted one who knew not

at the end of the 16th century. To this end, Morley writes, “It resteth now to give a reason why I have placed that
Table of Proportions in my book, seeing it belongeth no more to music than any other part of arithmetic doth
(arithmetic you must not take here in that sense as it is commonly, for the art of calculation, but as it is taken by
Euclid, Nichomachus, Boethius, and others), but the reason why I set it there was to help the understanding of many
young practitioners who, though they see a song marked with numbers… yet do they not know what Proportion that
is, and therefore if they do but look upon the numbers and mark the concourse of the lines enclosing them, they shall
there plainly find set down what relation one of those numbers hath to another.” See Morley, A Plaine and Easie
Introduction, 137.
141
Here, we might wonder how English music might have progressed should Morley have been more
encouraging to this end.
142
Ibid., 100.
143
Ibid., 134-135.

138
how to employ [his] time.” In conclusion, Morley offers “one word” on proportions other than

those of simple multiplicity (e.g. 2:1, 3:1, 4:1, etc.), namely that such mixed proportions like

superparticulars and superpartients (e.g. 3:2, 4:3, etc., and 5:2, 7:4, etc., respectively) “carry

great difficulty, and have crept into music I know not how; but [perhaps it was] by means of the

Descanters who, striving to sing harder ways upon a plainsong than their fellows, brought in that

which neither could please the ears of other men, nor could by themselves be defended by

reason.”144

Despite all his cautions and discouragements against the use of atypical proportions,

Morley does offer a work which demonstrates the use of complex genres of proportions via the

use of simultaneous multiplex ratios (the simplest genre of proportion, which Morley does not

wholly discourage) in a three-voiced composition in the same manner as John Dygon suggested,

being a viable option for an easier production of more complex species. In this work Morley

presents in his treatise, he presents all multiplex values found in Muris, many of which are seen

in combination with each other. Simultaneous multiplex proportions found in this tricinium

include 9:2 (see Example 3-8, m. 88-93 and elsewhere), [Link] (see Example 3-8, m. 102-103),

7:2 (see Example 3-8, m. 108-11).

Ultimately, Morley represents the musical aesthetic of his age: an increasing disinterest in

rhythmically diverse music. However, despite Morley’s pedagogical acclaim, and the

increasingly conservative trends in music at the end of the 16th century, there was persistent,

albeit small, interest in Tincto-Gaffurian proportional aesthetics percolating throughout Britain.

144
Ibid., 136. By the ' Descanters’ Morley is probably referring to that class of singer who in the fourteenth
century provoked the famous decree of Pope John XXII, of which the following is an extract: ‘They chop up the
melodies by hochete, mollify them by discantus and triplum so that they [the melodies] rush around ceaselessly,
intoxicating the ear without quieting it, falsify the expression and disturb devotion instead of evoking it.’ See Lang,
Paul Henry. Music in Western Civilization. New York: W.W. Norton and Co, 1941. 139.

139
Like never yet seen in European notated music since the Middle Ages, there was a fashion

during the late 16th century among some court and chapel composers under Elizabeth I to write

rhythmically experimental music, well beyond the common typical, dyadic, triadic, and hexadic

hierarchies.145 The most exceptional and most extensive single collection of such works from

Britain at this time is John Baldwin’s Commonplace Book, a text which, taken in its whole

straddles the divide between the pragmatic and conservative bent of Morley and the speculative

and adventurous inclinations of Dygon and Tucke.

Copied by John Baldwin (d. 1615), a singer and clerk at St. George's Chapel in Windsor

Castle and gentleman of the Chapel Royal of Queen Elizabeth I starting in 1594, the Baldwin

Commonplace Book contains 174 pieces, the largest percentage of which are by William Byrd146

(23 total pieces), followed by Baldwin himself at 20 pieces. Other composers of fame with

multiple pieces in Baldwin’s book include Bull (3), Giles (7), Marenzio (13), Mundy (11),

Shepherd (7), Taverner (13), Tallis (3), and Tye (5). However, many works remain anonymous

(23) or by composers we know very little about (likely associates or friends of Baldwin).

145
Some examples of pentadically organized music include Nicholas Strogers’ in nomine, Osbert Parsley’s
Spes nostra, and Christohper Tye’s In nomine XIII ‘Trust,’ and Alwood’s Agnus Dei from his Missa Praise Him
Praiseworthy. Mallory’s Miserere is a unique example of hendectadic music. Along with these examples of music
organized around some particular level of non-dyadic or non-triadic subdivision are music wherein various rational
rhythmic strata are set against one another for extended periods in a poly-mensural conceit. One example of the
extensive use of this poly-mensural stratification is Picforth’s In nominee.
146
Baldwin makes a point in both his opening and closing poem of the book to praise Byrd as the greatest
among the composers listed in the poem and given in the manuscript. Baldwin writes, “[All the composers listed
here are] famous in their art, there is of that no doubt. / Their works no fess declare in every place about / Yet let not
strangers brag, nor they these so commend / For they may now give place, and set themselves behind / An English
man by name, William Byrd, for his skill, / Who I should have set first, for so it was my will, / Whose great skill and
knowledge doth excel all at this time, / And far to strange countries abroad his skill doth shine.” He goes on to wax
extensively on his admiration for Byrd, writing that despite the fame of composer in and outside of England, in all of
Europe there “is none like to our English man / Who doth so far exceed as truly I it scan. / As ye can not find out his
equal in all things / Throughout the world so wide, and so his fame now rings. / With fingers and with pen, he hath
now his peer / For in this world so wide, is none can him come near. / The rarest man he is, in music’s worthy art, /
That now on earth doth live – I speak it from my heart – / Or heretofore hath been or after him shall come / None
such I fear shall rise that may be called his son….” (fol. 189v-r, modernization by Jordan Alexander Key)

140
Particularly interesting to note is that the manuscript also contains one piece each by Dygon and

Morley, suggesting that Baldwin was at least familiar with these two composer’s musical works,

if not also their theoretical treatises. The genre contained therein also reflect a wide interest and

function, ranging from Mass Proper and Ordinary settings and sections along with other sacred

anthems, services, motets, and hymns to a variety of secular works, including Italian madrigals,

English consort music, and 44 textless instrumental pieces.

Baldwin’s Commonplace Book stands as a unique source for the music from some of the

above listed prominent 16th century British composers; however, we will focus our observations

herein on a lesser recognized sampling of pieces from the latter portion of the book. While these

works have been relegated to “pedagogical” works by nearly every scholar that has noted their

existence, it seems unlikely that they were intended only, or even at all, as teaching pieces, given

the book’s contents and the complexity of the works themselves. Baldwin provides a description

of the book’s contents in poems supplied at the beginning and ends of the manuscript. In the first

poem, the book itself addresses the reader, declaring that

I doe belonge of proper righte

and dutie owe him whom me boughte.
I bring no hurte to anie man
but doe him pleasure what I can
Such is the nature of sweete musicke
that she would all please none dislike… 147

In the concluding poem, Baldwin describes the book as

A store housse of treasure this booke maye be saide

of songes most excellente and the beste that is made
collected and chosen out of the best autours
both stranger and Englishe borne .... 148

147
Baldwin Commonplace Book, fol. 0v
148
Ibid., fol. 189v

141
 By Baldwin’s words, we can recognize that the works contained in the Commonplace

Book are meant to bring “pleasure” through “sweet musike,” not erudition through pedagogical

explanation. Rather than trifles for teaching, as those found in Gaffurius and Dygon, all the

works within this “storehouse” are “treasures” and “songs most excellent” and the “best that

[are] made.” In these remarks we should note that Baldwin must consequently stand counter to

Morley on proportional music. Since the contents of Baldwin’s book, which contain pieces of

proportional music, are designed for and capable of “pleasure” and are themselves “sweet,”

Baldwin cannot agree with Morley’s admonition that “it is a matter almost impossible to make

sweet music” from music with proportional complexity, which can “neither… please the ears…

nor could by themselves be defended by reason.”149 Thus, we should conclude, against past

appraisals, that Baldwin did not intend the complex music in his treatise as pedagogical material,

since there is no theoretical aspect to this text, nor are any of the other works in this source in

any way understood as teaching aids; furthermore, Baldwin clearly believes that the works in the

manuscript are all meritorious and things to be “treasured,” not overlooked or discounted.

Furthermore, these rhythmically complex pieces are quite numerous, totaling at least 20

of the 175 works in the manuscript. Unfortunately, it is not the purview of this document to make

of comprehensive study of these works, though a table below outlines all the proportionally

eccentric pieces within the book (see Table 3-1). Herein, we will only showcase one. Example 3-

9 presents one of the most complex and rhythmically diverse works, Nathaniel Giles 1594 duo

on Miserere, featuring, as the manuscript advertises, “38 proportions of sundry kinds,”

proportions which include all integers and prime subdivisions up to 9 along with various

multiples of these primes (see Figure 3-2). Furthermore, we see the systematic implementation of

149
Morley, A Plaine and Easie Introduction, 134-136.

142
the Tincto-Gaffurian system of numerical and orthographic proportional notation along with

some kind of pseudo-Dygonian sign notation.150

Ultimately, what are we to make of the variety of mensuration signs and proportions

available to the 15th and 16th century and the use of those signs? The many varieties of signs in

15th century music and the inconsistencies in their usage may give the impression that the

mensural system was no less unnecessarily complex and redundant than the multifarious system

of note shapes from the late-14th century. Indeed, to some extent, what we have surveyed in this

chapter is the relocation of agency in rhythmic structuring from the note itself to the signs

(orthographic, numerical, or semiotic) contextualizing those notes. While in the 14th century,

new rhythmic values were primarily demarcated through the shape, coloration, voiding, or

flagging of a note individually under a relatively small set of mensural possibilities, the 15th and

16th century saw a proliferation of mensural signs to diversify rhythm with the notes themselves

remaining rather fixed in their development and variations over this period.

150
What we consistently see in Baldwin’s manuscript within these rhythmically complex works, including
Giles, is a seemingly inconsistent usage of mensural signs in conjunction with numerical and orthographical
proportions with no relation to the system outlined by Dygon (see Figure 3-2); it is still unknown in what way
Baldwin intended these signs to be interpreted considering the given proportions. However, the proportions given
are clear enough and can be easily used to derive a work with correct counterpoint and voice leading.

143
Table 3-2. Rhythmically eccentric works within the Baldwin Commonplace Book.
Composer Voices Name Location

Baldwin, John 2 A Duo upon ut, re, mi, fa, 1595 ff. 100v – 101r
(c. 1560 – 1615) Spes mea ff. 108v – 109r
3 O lux ff. 122r – 123v
upon the plainsong ff. 125r – 126v
Sermone blando on Angelus ff. 112r – 113v
4 4 vocem,’proportions to the minum ff. 127r – 128v
Salve nos ff. 110r – 111v
Bedyngham, John (mid- 2 Manus Dei ff. 105v-106r
15th century or late 16th Vide Domine ff. 104v-105r
century)151
3 Salve Jesu ff. 106v – 107r
Giles, Nathaniel 2 A Duo, Miserere (1594) ff. 102v – 103r
(c. 1558 – 1634) In te domine speravi ff. 107v – 108r
3 Salvator mundi ff. 111r- 112v
Tye, Christopher 3 Sit fast ff. 113v-115r
(c. 1505 – c. 1573)
Woodson, Thomas (late 3 Upon ut, re, mi, fa ff. 101v-102r
16th century)
Anonymous 2 Four Agnus Dei ff. 96v-98r
(late 16th century) Exaudi me fol. 103v
Parce domine fol. 105r
4 Salve nos I fol. 110v
Salve nos II fol. 111r

151
There is some uncertainty whether this composer refers to the 15th century Johannes Bedyngham of
Westminster, London, reported between 1459 – 1460 or some other John Bedyngham contemporary to John
Baldwin at the end of the 16th century. If we suppose that this is the earlier Bedyngham, then we know that interest
in such rhythmically complex music in Britain extends well into the past, and contextualized Dygon and the Baldwin
Commonplace Book interestingly. However, Baldwin’s access to these pieces, should they be that old, and his
inclusion of them among composers generations after seems unlikely. Furthermore, should these pieces be from the
mid-15th century, then the rhythmic capabilities of this composer in England outpace the innovations of the Tincto-
Gaffurian system and are extremely forward thinking for their time. It seems more likely that these pieces by a
“John Bedyngham” are probably by a composer of the same name, but much later and contemporary to Baldwin.
Placing them in the context of the late-16th century makes more sense given the development and spread of
proportional notation systems across Europe and into the British Isles over the 15th and 16th century. Herein, we
assume the latter hypothesis.

144
Figure 3-2. Proportions given in Nathaniel Giles’ duo on Miserere.

145
 Conclusions to Chapter 3

These works presented in this chapter, as well as the works aesthetically orbiting within

the system of proportional music during the 15th and 16th century, embody the music Johannes

de Muris hypothesized in Notitia artis musicae as early 1321, when he advocated the possibility

of the proportional subdivisions of notes of the same shape from two to nine equal parts.

Significantly, not only did Muris imagine music that possess dramatic rhythmic diversity, but

also a notational system where the note shapes remained relatively fixed while the

contextualization of those shapes changed to accommodate differing interpretations. Even before

the outpouring of complex note shapes from the ars subtilior, Muris was able to see the utility of

few note shapes with many possible recontextualizations. It would take another 150 years before

a system would arise to make Muris’ music possible.

The Tincto-Gaffurian proportional notation system would ultimately point to the standard

model of time signature we enjoy today; however, the full implications of the system would go

nearly unrecognized until the 20th century, with only a handful of eccentric rhythmic pieces from

the 15th century. However, the speculative nature of Gaffurius’ book on proportions, printed at

the end of the 15th century, was compelling enough to find its way to the British Isles and into

the hands of a few theorists, who themselves found it worthy enough to copy into their own

theoretical tracts. Despite the curious few who found significance in the Tincto-Gaffurian

proportional possibilities, most prominent theorists of the age (Glarianus, Zarlino, and Morley to

name only a few), were strongly discouraging of such adventurous attitudes in music. Against

such damning judgments on the pedantry of Gaffurius’ catalogue of examples, a few theorists

and composers embraced, at least in part, the potential of diversity. John Baldwin, Nathaniel

Giles, and John Bedyngham, along with numerous other British composers dabbled in this subtle

146
art of rhythmically mathematical music at the close of the intellectual Dark Ages, just before a

different kind of “dark age” would descend on music itself: nearly three centuries of stagnation

in rhythmic development.

Again, what can we learn from this “collapse” or abandonment of a sophisticated notation

system? In attempting to reassess our own methods of rhythmic diversification in the early-21st

century, what could we learn from Tincto-Gaffurian and Dygonian systems that provided such

clear and rational paths to diversity yet could not impact enough aesthetic change to broadly

realize that full potential? Indeed, is there even a problem with such a system’s lack of

widespread utilization? Does proliferation signify goodness necessarily? Are such oddities

negligible as a consequence of their atypicality, or should they be more carefully considered as

radical experimentation and not merely discounted by some excuse of pedantry or pedagogy? If

our present system of time signature is derivative of this underrealized system of infinite variety,

what are we overlooking in our current practice that is already inherent within it that could allow

us access to such staggering musical possibilities?

Synthesis II

AXIOM S2.01: A system exists apart from time, in abstraction, though some of its
elements are pointers to concrete phenomena.

Definitions

DEFINITION S2.01A: the matric concatenation of a system’s elements is called

quantization.

DEFINITION S2.01B: The elements within such a quantization matrix are called the quanta.
There may be more quanta in a system’s quantization than elements within the system
itself, though all quanta must be system elements. A quantization may or may not be
systematic in its construction. The construction of the quantization, if systematic, cannot
be the same systemization as that which begat its unquantized self, since a quantization is
a system necessarily dependent on time, while the unquantized system is independent of
time (see Axiom 5.01).

147
 DEFINITION S2.01C: All quantizations may be partitioned; namely, any matrix
configuration of elements may be divided into sub-matrices. These sub-matrices are
called sub-quantizations.

DEFINITION S2.02: An operator is a particular kind of element within a system whose

function is to limit the interpretation and translation of other non-operator elements. They
are pointers to abstract logical modifiers, rather than concrete phenomena. A system may
have multiple operators that function to limit the interpretation of different parameters of
the system. In a quantized system, an operator may be assigned as an interpretative
limiter to individual quanta, sub-quantizations of many quanta, or the entire quantization
itself. Operators are classified into types and variations of those types.

DEFINITION S2.03A: An operator type assigns an operator’s general interpretative

function within the system. A non-operator element may be assigned multiple operator
types.

DEFINITION S2.03B: An operator type variation assigns an operator’s particular

interpretative function within the system, grouping that operator with others of a similar
general interpretative function within that same system. If a system contains variations of
operator types, then each variation is assigned to a subset of the system’s non-operator
elements, each subset interpreted according to its operator’s general type and particular
variation. Each non-operator element within a system may be assigned only one operator
type variation, though multiple operator types and variations may exist within a system.

Lemmas

LEMMA S2.01: An unquantized system may be interpreted and translated, but such a
translation will only return the general definitions of its elements. Consequently, the
interpretation and translation of an unquantized system will always return the same result.

LEMMA S2.02: Operators are not pointers to concrete phenomena but are interpretative
functions. Consequently, all operators, to operate, must be assigned to one or more non-
operative concrete-pointer elements; otherwise, they are inoperative within the system.

LEMMA S2.03: Assigning operators to non-operator elements is a concatenation of at least

two elements and is thus a quantization. Thus, an operating operator implies quantization.
Consequently, all operators are considered inoperative until assigned to elements.

LEMMA S2.04: A system is made a function of time through quantization since some of
the quanta of that quantization are pointers to concrete phenomena, which require time.

Observations

OBSERVATION S2.01: Note that a system may have as many operator types and type
variations as one wishes. Since operators are independent of the generative elements, one

148
 may arbitrarily apply any operator to a system without necessary making the system
unsystematic.

OBSERVATION S2.02: Note that while generally there need be no particular upper bound
on the number of operator types or type variations, it should be recognized that in
“practical circumstances,” (namely, systems applied to unaided human processing) a
relatively small number of operators should be employed for any system so that the
system remains reasonably coherent.

OBSERVATION S2.03: Some operator type variations will become trivial should the
process of variation be iterative or patterned in some way. Though we need not
enumerate such variations ad infinitium, and while nearly all such infinite variations will
be inapplicable to human-processed systems, we may still recognize the potentiality and
existence of such trivial elements, though most will be inoperative under some particular
quantization.

Conclusions

CONCLUSION S2.01: The potential of a system is only realized through the operators
applied to it. Since the application of operators to a system’s elements, in part or whole,
is quantization, a system is only meaningfully interpretable and translatable under
quantization.

CONCLUSION S2.02: If a system is systematic, then it contains one interpretation and

translation, even if the system itself contains an infinitude of trivial derivative elements
(see Lemma S1.01).

• COROLLARY: If an unquantized system has more than one possible interpretation

and translation, then it is unsystematic or else is not wholly unquantized.

• COROLLARY: To optimally interpret and translate a quantized system, one must

first fully unquantize it to determine the fundamental definitions of concrete and
logical pointers.

CONCLUSION S2.03: If a particular system and a set of operators is widely adopted for
interpretation and translation, it is advisable to derive any further operators from the
operators given, or else new operators need to be justifiably and wholly different from all
other forms, not generatable from some process and semiologically distinguishable, being
itself uniquely distinguished by its assigned logical modifier.

CONCLUSION S2.04: Recognizing the limited nature of human perception as a discrete

processor, as noted in Observation S1.01, the operators of the system must not only be
necessarily unique, and/or logically derivative from each other but must also have some
reasonable limit to variation or invention, disregarding trivial ad infinitium variation as in
Observation S1.02.

149
CONCLUSION S2.05: The translation of a quantized system will produce unique results
based on the particular quantization of that system. Such translations may or may not
reflect the general definitions of the individual elements in the quantization due to the
element’s logical modification by quantized operators. Consequentially, the interpretation
of elements may change under particular quantizations due to the operators applied to
them.

• COROLLARY: Thus, interpretation and translation of quantized systems will not

necessarily return the same results as the unquantized system’s translation.
Furthermore, different quantizations of the same system will not necessarily
return the same translation under interpretation, though they both be quantizations
of the same system.

• COROLLARY: Since a quantization, systematically or unsystematically created,

will not return the same interpretation and translation as the unquantized system,
and since there are arguably infinite operators and system quantizations under
such operators, any quantization has a high probably of a unique interpretation
and translation.

CONCLUSION S2.06: Operators allow for “pointer pivoting” within a system, as concluded
necessary in Conclusion S1.05 and appended corollaries.

150
 CHAPTER 4
THE DECLINE AND REBIRTH OF RHYTHMIC DIVERSITY: IVES, COWELL,
NANCARROW, CARTER AND RHYTHM IN THE EARLY 20th CENTURY

The Dark Age of Rhythm and The Genesis of Time Signature: 16th to 19th Century

By the start of the 17th century, interest in rhythmic experimentation was essentially dead.

The Baldwin Commonplace Book and a few scattered works from the Elizabethan avant-garde

stand as the last markers of proportionally diverse music before the next three centuries of

relative disinterest in rhythmic complexity. While Europe enters its philosophical and scientific

Enlightenment, and while this period is often seen as the “Classical” period of music –

embodying much, if not all, the musical values we hold dear today: tonality, tunefulness,

expressivity, individuality, balance, simplicity, and form – this period from the 17th century until

the early-20th century is the nadir of rhythmic development since Franco of Cologne. Before the

17th century, each successive century since the 13th had seen transformations, developments, and

innovations to the system of rhythmic notation and its ability to express different structures in

rhythm, whether utilized or not; however, beginning in the 17th century, we arguably see no

further expansion in the rhythmic capabilities of Western music notation for the next three

centuries. Furthermore, we see an abandonment of many of the developments from the 15th and

16th century, again witnessing a collapse to a system arguably similar to, if not simpler than, that

proposed by Franco of Cologne nearly half a millennium prior: a system intrinsically bound to

dyadic and triadic rhythmic hierarchies.

It is perhaps the fault of theorists at the end of the 16th century, who admonished or

belittled the exploration of exotic rhythmic proportions, that Europeans retreated deep into a

dyadic aesthetic. Like many of his contemporaries, Zarlino – a product of the reemergence of

Greek thinking, Humanism, and the end of Gothic aesthetics – ardently defined music’s beat

151
according to Aristotelian principles and discouraged the use of anything beyond the simplest and

clearest divisions of two and, occasionally, three. In the third book of Zarlino’s treatise on

counterpoint, Le intitutioni harmonische, in the chapter “Della Battuta,” the most preeminent

music theorist of the 16th century explains that that “beat” (Battuta), or “sounding time” (tempo

sonoro), namely that which Augustine (“the vey saintly doctor”) calls plauso, is a motion

consisting of two parts.1 These sentiments Zarlino expressed on the division of the beat are

consistent with most of his contemporaries; however, Zarlino’s evaluation is exaggerated with a

potpourri of ‘authoritative’ citations: Augustine, Galen, and Aristotle.2

Even before Zarlino, Martin Agricola’s Musica instrumentalis deudsch3 of 1528, only

discusses mensurations cut-C and C2, both of which stand for a dyadic rhythmic hierarchy. At

least as far back as 1511 (six years before the birth of Zarlino), German music theorist Sebastian

Virdung had done away with most non-dyadic mensurations in his Musica getutscht.4 Despite

1
Zarlino, Le istitutione harmoniche (Venice: Francesco de I Franceschi Senese, 1558), 207.
2
Such citations as given by Zarlino are part of a long and enduring tradition of music theorists relating the
study of music to natural philosophies, a cross-disciplinary relationship stretching back at least to the Ancient
Greeks. This relationship was apparent in music theory to greater and lesser extents depending on the aesthetics of
the age. In the Middle Ages, theorists often cited Aristotle’s Physics alongside discussions of beat. Marchetto of
Padua in his Pomerium of 1318-1324, explained beat as a motion wherein “every measure is [made] in a certain
quantity and time,” and “every time is the measure of change (according to the philosopher¸ Physics IV).” See
Marchetto of Padua, Pomerium, in Marcheti de Padua Pomerium, ed. Giuseppe Vecchi, Corpus Scriptorum de
Musica 6 (Rome: American Institute of Musicology, 1961), 75, as cited and translated in Blachly, “Mensura versus
Tactus,” 438-439. On the relationship between medieval Aristotelianism and theories of meter generally, see Dorit
Tanay, Noting Music, Marking Culture: The Intellectual Context of Rhythmic Notation, 1250-1400, Musicological
Studies and Documents 46 (American Institute of Musicology: Hanssler-Verlag, 1999). See also the earlier
discussion of this context in Max Haas’s “Studien zur mittelalterlichen Musiklehre: I, Eine Übersicht über die
Musiklehre im Kontext der Philosophie der 13. und frühen 14.. Jahrhunderts,“ in Aktuelle Fragen der
musikbezogenen Mittelalterforschung: Texte zu einem Basler Kolloquium des Jahres 1975, ed. Hans Oesch and
Wulf Arlt (Winterhur: Amadeus, 1982). 323-456; and Hans Eggebrecht, „Die Musiklehre im 13. Jahrhundert von
Johannes de Garlandia bis Franco,“ in Die mittelalterliche Lehre von der Mehrstimmigkeit, ed. Frieder ZamZaminer
(Darmstadt: Wissenschaftliche Buchgeselschaft, 1984), 89-159.
3
Martin Agricola, Musica instrumentalis deudsch (Wittemberg, 1528), fol. xxij r-v.
4
Sebastian Virdung, Musica getutscht (Basel 1511), fol. H ij v to H iij r; facsimile edition by Klaus
Wolfgang Niemöller (Kassel: Bärenreiter, 1983).

152
some radical theories surfacing perhaps in response to Gaffurius’ practica musicae - Sylvestro di

Ganassi’s la Fontegara5 of 1535 on the Continent and John Dygon’s contemporaneous

Proportiones practicabiles secundum Gaffurium on the British Isles – most 16th century theory

passively or actively discouraged the use of indivisible polyrhythms other than 3:2 or 2:3 and fell

into the ever-growing Aristotelian aesthetic of dyadic divisibility. Following in Zarlino’s

substantial theoretical footsteps, many other music theorists from this period cite similar

Aristotelian aesthetics of musical time.6 With the impetus of prominent Italian theorists like

Zarlino and Glareanus, German theorists like Agricola and Virdung, and English theorists like

Morley, the subjugation of the “imperfection of one single symbol up and downward to all other

symbols was an important step away from the complex mixture of perfect and imperfect values

towards the modern concept of a normally square organization.”7 Musicologist Roger Grant

shows that even as late as 1723, Nassarre’s Escuela musica segun la practica moderna still

included references to Aristotle in the context of rhythm and beat.8

5
Sylvestro di Ganassi, Opera intitulata Fontegara (Venice 1535), facs. Milan 1934, Ch. IX; See some
examples in Imogene Horsley, Improvised embellishment, in Journal of the American Musicological Society, Vol.
IV, 1951, 7. In general musical practice on the Continent, particularly in regard to notated music, most proportions
outside the multiplex and the sesquialtera were rare. However, one comprehensive example outside theoretical
spheres and discussing music as performed was Ganassi’s method for the flute, the Fontegara of 1535, which
teaches the art of improvised diminution and melodic embellishment by proportional rules, the first of which
subdivides the semibreve into 4 semiminims; the second, into 5; the third, into 6; the fourth, into 7 semiminims. The
4 and the 5 semiminims are in the ratio sesquiquarta; the 4 and the 7, supertripartiens; the 5 and the 6, sesquiquinta;
the 5 and the 7, superbipartiens; the 6 and the 7, sesquisexta.
6
See Agostino Pisa’s Breve dichiaratione della Battuta musicale, facsimile ed., ed. Piero Gargiulo, with
extracts from Battuta della musica (1611; Lucca: Libreria Musicale Italiana Editrice, 1996); and idem., Battuta della
musica (Rome: Bartolomeo Zanetti, 1611), facsimile ed. in Pisa, Breve dichiarazione della battue musicale [sic], ed.
Walther Dürr (Bologna: Forni, 1969).
7
Curt Sachs, Rhythm and Tempo, 226.
8
Roger Mathew Grant, Beating Time & Measuring Music in the Early Modern Era, 24. For more
information on Aristotle’s influence on Renaissance and early modern thought see Charles B. Schmitt, Aristotle and
the Renaissance (Cambridge, MA: Harvard University Press, 1983); and Paul F. Grendler, “Universities of the
Renaissance and Reformation,” Renaissance Quarterly 57, no. 1 (2004): 1-42. Grant poses a few interesting
questions that beg further investigation throughout Western aesthetic history, including music: “What are these
references to Aristotle doing in theories of the beat? More important than the lines of influence they may reflect is

153
 The declining popularity of triple subdivisions of the beat throughout the 15th and 16th

century can easily be demonstrated by sampling nearly any subset of music from this time.

Figure 4-1 below gives some impression of this mounting pull towards dyadic organization of

rhythm. While this small survey is somewhat misleading, since it fails to recognize binary

rhythmic divisions under which ternary subdivisions may be present, it accurately demonstrates a

retrospectively incontestable fact: the declining popularity of ternary beat divisions in favor of

binary rhythm.

By the time Petrucci printed his volume of Antione Brumel’s masses in 1503, aesthetic

tastes in music were bending in a new direction, the diverse rhythms, polyrhythms, structures,

and notations derivative from a more “Gothic” period were being forsaken. By the second half of

the 16th century, Zarlino sneers at “the speculations put forward by certain idle theorists of that

day,” who “treat of nothing but circles and semicircles, with and without points, whole or

divided not only once but two and three times, and in them one sees so many points, pauses,

colors, ciphers, signs, numbers against numbers, and other strange things that they sometimes

appear to be the books of a bewildered merchant.”9 The burgeoning new aesthetic trend in music,

bent towards poetry and Humanism, an aesthetic often called the “High Renaissance,” greatly

affected rhythm. Early-20th century musicologist, Curt Sachs, assessing this drastic shift in

musical tastes, poignantly noted that composers, both inside and outside of Italy, living through

the 16th century, such as Obrecht, Isaac, and Josquin des Pres, “chimed in with the visual arts and

their shared relationship to a general set of beliefs about time and motion, and the institutional control and
sanctioning of these beliefs.” See Grant, Beating Time, 24.
9
Oliver Strunk, Source Readings in Music History (New York: W.W. Norton, 1998), 456. Given our
discussion in the prior chapter on the influence of merchant arithmetic on the notation of proportional music via
Hindu-Arabic numerals, this remark by Zarlino is rather prescient.

154
led the way to the classicistic ideals of clarity and symmetry in one of the most impressive

apostasies in the history of music.”10

Figure 4-1. Sample survey of 15th-16th century repertoire in perfect time.11

The growing trend towards “square organization,” as many 16 th and 17th century theorists

would come to call this rhythmic hegemony, emphasizes that against centuries of development of

the complex French and Franco-Flemish proportional and mensural systems, the Italians, who, as

Curt Sachs brashly generalized as “always fond of simple, static, symmetrical forms,” were the

“winners in one battle in the eternal tug-of-war between the two elementary trends,” ultimately

imposing “clean-cut duple time upon the music of all countries.”12 The role of “perfection” and

10
Curt Sachs, Rhythm and Tempo, 239.
11
This small survey simply shows the number of works from each source which are in either major
(perfect) prolation or minor (imperfect) prolation without consideration of tempus, since prolation deals more with
“beat” and micro-rhythmic subdivision than tempus, which is more about macro-metrical delineation in most 15th
and 16th century music. This study, of course, is not comprehensive, but simply shows a fact that all musicologists
already agree to be true: major prolation declined in popularity throughout the 15th and 16th century. Thus, the chart
hopes to not prove something new, but merely visually demonstrate to the reader something already well known.
12
Ibid., 226.

155
“imperfection” would essentially swap, with dyadic divisions becoming the standard measure

while triadic divisions would become the exceptional warping of that regular dyadic

“perfection.” Curt Sachs’ remark regarding this change of attitude towards “perfection” and

“imperfection” in rhythmic stratification is worth giving in full:

… despite its disparaging name, imperfection was soon not only admitted to the rights of
perfection: [imperfection] grew stronger and stronger, gained victory eventually, and
tolerated short triple-time episodes only for the sake of contrast in the middle or at the
end of a piece. And far from being ‘perfect,’ these [short] episodes were, on the contrary,
incomplete diminutions of duple time.13

For nearly half a millennium, dyadic time was a subservient, if not warped, alternative to triadic

time, providing a novel, if not somewhat base, alternative to perfection. However, beginning in

the 16th century, triadic time became the “tolerated” diversion and was now considered the

fractured, diminished version of dyadic time.14

By the 17th century, most theorists that discuss mensuration and proportion, or rhythm

and tactus generally, only speak of C and cut-C and very rarely include any discussion of O or

any variety of it, let alone more obscure proportions. Furthermore, the concept of mensuration

and proportion migrate from note division to relative change in tempo, C and cut-C simply

signifying slower and faster tactus respectively. Over the 17th century, the system of mensuration

not only become drastically limited to a subset of the rather limited subset from the 15th and 16th

century but also migrates away from strict mathematical proportion to relative tempo.15 Sachs

13
Ibid., 227.
14
a change in fates still living with us today, evidenced by the greater frequency of triplet signs over duplet
signs in modern music notation.
15
For a detailed discussion of this history of mensuration and tempo during the 17th and 18th century see
George Houle, Meter in Music, 1600-1800: Performance, Perception and Notation (Bloomington: Indiana Univ.
Press, 2000), 1-34. One interesting counterexample to this trend is theorist Pier Francesco Valentini, who in 1643
devoted over 150 pages of his detailed manuscript, Trattato del tempo, e del modo, e della prolatione, to
proportions, devoting most of his discussion to 16th century music. Valentini gives examples of numerous
proportions, both duple and triple, and explores every possible proportion regardless of its practical use. Discussing

156
aptly summarized the sentiments at the outset of the 17 th century among the “monodists” of the

new emotionally oriented music of opera and madrigal.

The new emotionalism [of the Baroque], the urge to overwhelm the listener and make his
nerves vibrate— an urge so strong in the works of the early monodists around 1600 that
a weeping audience seemed to be the vital aim of composing —excluded rhythmic devices
born of sheer delight in purely rational structures without emotional values. The Gothic
proportions had to go.16

Starting in the 17th century, we begin to see the coalescence of our modern time

signatures and their meaning. First, we see a continued reliance on mensuration signs to

contextualize proportions, holding onto many of the chimeric forms given by Hothby nearly two

centuries prior. While there are examples of proportion signs used alone, the early-17th century

generally followed the semiotic trends of the 16th century, albeit with a more simplified

understanding of the chimeric signs. To such ends, in 1673 Giovanni Maria Bononcini wrote that

“it must be surely given that, in songs, the introduction of proportions without a sign of

mensuration is (as Valerio Bona says in his Rules of music) like sending soldiers to the

Giovanni Maria Bononcini:17

tripla maggiore: O3/1, C3/1, cut-O3/2, cut-C3/2,
tripla minore: C3/2, O3/1, C3/1, C3/2
C3/4, tripla di semiminime.
C3/8, tripla di crorne.
C6/4, sestupla di semiminime.
C6/8, sestuple di crorne.
Cl 2/8, dodecupla di crorne.
C12/16, dodecupla di semicrome.

Figure 4-2. Triple meter signatures as given by Giovanni Maria Bononcini in 1673.

more numerical signs than any other theorist of the time, Valentini included superparticular (such as 5/4, 7/6, and
10/9), multiplex (such as 5/1 and 7/1), and submultiplex proportions (such as 1/5 and 1/7). However, unlike the
examples from the prior chapter that have some practical counterparts, there seems to be no practical music
counterpart from the 17th century that reflects the proportional possibilities enumerated by Valentini. See Pier
Francesco Valentini, Trattato del tempo, e del modo, e della prolatione (Rome: Vatican Library Ms. Barb. lat. 4419,
1643), 300-459.
16
Curt Sachs, Rhythm and Tempo, 269.
17
Bononcini, Musico prattico, 20-23.

157
 Lorenzo Penna:18
3/1, tripola maggiore
3/2, tripola minore,
3/4, la tripola picciola,
3/8, la tripola crometta,
3/16, la semicrometta.
6/4, la sestupla maggiore.
6/8, la sestupla minore.
12/8, la dosdupla.

Figure 4-3. Triple meter signatures as given by Lorenzo Penna in 1684.

battlefield without a captain.”19 In the triple meter signatures of Bononcini himself, we begin to

see the impetus of our modern time signatures (see Figure 4-2).

However, just one decade later we see the divorcing of these “time signatures” from

their traditionally appended mensuration signs in the work of Lorenzo Penna (see Figure 4-

3), a trend that would continue into the 18th century that would ultimately and indefinitely

forsake all such signs, barring C and cut-C. Triple mensural proportions and signs, while still

used throughout the 17th century despite their relative decline in popularity through the 16th

century, were gradually transformed into the fractional numbers of modern time signatures

and made subservient to rationally dyadic rhythmic hierarchies.

18
Lorenzo Penna, Li primi albori musicali (4th ed., Bologna, 1684), 36-40. It is worth noting that Penna
calls them signs of “tripola,” rather than proportions, showing a further departure form time signatures understood as
proportions. It is also worth noting Penna’s theoretical parallel of Valentini from a generation prior. While the
number of signs given by Penna is few when compared to Valentini, Penna explains that he is only demonstrating
those most frequently used. Admitting the existence, however rare, of other rhythmic proportions, Penna somewhat
anachronistically includes a few additional proportions that were “formerly used”: hemiolia maggiore and minore as
well as the proportions 5/2 and 7/2, included as “tripola.” However, Penna’s explanation of these old and unused
proportions is brief, though he, in passing, admits that there “are [other proportions] in other forms.”
19
Giovanni Maria Bononcini, Musico prattico (Bologna: per Giacomo Monti, 1673), 14. “Per ultimo si
deue auuertire, che l’introdurre le proporzioni ne i canti, senza segno del Tempo e (come dice Valerio Bona nelle
sue Regole di musica) come mettere i soldati in Campo senza Capitano”

158
 By the end of the 17th century, Henry Purcell, perhaps echoing Morley’s sentiments a

century prior against rhythmically complex music, declared pronouncedly that the three

“signatures” (once mensurations) only indicated different tempi, nothing else. According to

Purcell, C “is a very slow movement,” cut-C “a little faster,” and cut-O a “brisk & airry time.”20

Herein, we see a new emancipation of “proportions” from mensuration signs; however, the 17 th

century emancipation guides proportionality in a wholly different direction than imagined by

Tinctoris and Gaffurius at the end of the 15th century when they attempted to give numbers a

musical power beyond traditional signed rhythmic hierarchies. By 1698, when Georg Muffat

gave a brief overview of the proportions, any theoretical discussion of “proportions” as such was

merely a “meaningless” deference to the past.21 While Tinctoris’ dream of proportions replacing

mensuration signs was finally coming into fruition, any proportion not equivalent with

mensurations signs in popular use (namely, primarily dyadic and secondarily hexadic

hierarchies) were passed over or forgotten.

By the outset of the 18th century, numbers were rapidly losing much of their proportional

potency, divorced from “speculative” musical possibilities, and mensuration signs held little to

no proportional significance, if used at all. In 1705, theorist Johann Peter Sperling used the word

Zeichung (“signatures”), rather than “proportions,” for meter signs. Houle remarks that “this

change in terminology recognized a new function of the fractional numbers… inherited from

mensural notation.” Sperling, by not joining numerical proportions with traditional mensuration

signs like “C,” identified such numerical ratios as independent signs in themselves rather than

proportions:22

20
Henry Purcell, A choice collection of lessons for the harpsichord or spinet (London, 1696).
21
Curt Sachs, Rhythm and Tempo, 270-271.

159
 When an unknown figure, one not explained in this third chapter, is encountered,
consider the two numbers written one above the other. First, the upper number shows
the quantity or how many notes are in the measure. Next, the bottom number shows
the quality or what kind of note makes up the number counted in the measure. If this
bottom number is 1, so many single-part notes or whole notes go to a tact. If the
number is 2, so many two-part notes or half notes go to a tact.23

Furthermore, Sperling designates new names for notes, names that correspond to the fraction

of the beat the note occupies. For example, “ein halbschlagig Note” (“a half-beat note”) is

half of a semibreve or “eintheilige Noten” (“whole note”), since one semibreve is equal to a

beat in his theory.

In the third chapter of his treatise, Sperling continues to explain his system of “time

signatures,” which included 59 in all, each systematically derived from all possible

combinations of 1, 2, 4, 8, 16 as numerators and denominators for “four-part measures” and

3, 6, 9, 12, 24 as numerators to 1, 2, 4, 8, 16 denominators for “three-part measures,” wholly

binding triadic subdivisions to dyadic hierarchies in a hexadic complex. Henceforth dyadic

rational hierarchies in Western Music would reign supreme.24

22
Houle, Meter in Music, 43.
23
Johann Peter Sperling, Principia musicae (Bautzen, 1705), 47. “Zu wissen ist nur/ class wann eine
unbekannte/ und in diesem dritten Capitel Night vorgestellte Zeichung des Tacts vorkommet/ man zu betrachten
habe solche aus zweyen über einander gesetzten numeris bestehende Zeichung; und zwar erstlich den obem
Numerum, welcher anzeiget die Quantität, wieviel nemlich noten aufm Tact gehen: Und hernacher den untern
Numerum, welcher anzeiget die Qualität/ was nemlich fur noten es seyn/ deren so viel auffm Tact gehen; Ist nun
solcher unterer Numerus/ 1. so gehen so viel eintheilige Noten/ das ist gantze/ auffm Tact: Ist dieser Numerus 2. so
gehen so viel zweytheilige Noten, das ist halbe/ auffm tact.“ Translation by Houle, Meter in Music, 43.
24
The classification of time signatures to distinguish between duple and triple was used by most early 18th
century German writers, including Martin Heinrich Fuhrmann, Johann Gottfried Walter, and Wolfgang Caspar
Printz. See Martin Heinrich Fuhrmann, Musikalischer Trichter (Frankfort an der Spree, 1706), 44; Johann Gottfried
Walther, Praecepta der Musikalischen Composition 1702 (Leipzig, 1955), 28-33; and Wolfgang Caspar Printz,
Compendium musicae signatoriae (Dresden, 1714), 16. In 1714, Wolfgang Caspar Printz echoed the contemporary
trend to abandon mensuration signs altogether for the sake of a purely numerical time signature when he wrote,
“Wenn der Gesang mit einer irrationalem Proportion anfangt/ lassen die meisten neuen Musici das Signum
quantitatis mensuralis weg/ und setzen unter die Zahlen/ so die Proportion andeuten/ allein: und zwar
Nicht ohne Ursache. Denn weil die untere Zahl der vorgeschriebenen Irrationalen Proportion schon die
Krafft hat die Lange des Tactes anzudeuten/ so ist das Signum quantitatis mensuralis uberfliissig/ unnothig/
und also/ vermoge . . . abzuschaffen.“

160
 Elsewhere on the Continent, we see the abandonment of note coloration in France, an

invention often credited to the French Ars Nova. Etienne Loulie’s Elements from 1696

criticized mensural signs and various outmoded notations like coloration and void notation;

in his treatise, Loulie remarks belittlingly that “there is more caprice than reason in most of

these signatures.”25 Loulie also gives the first system that defines meter signs by the number

and kind of beats contained in the measure and the classification system of meters, both

systems readily adopted by almost all 18th century theorists and still used today.26 Loulie’s

system classified meters into duple, triple, or quadruple and their compound versions. In

1709, Michel Pignolet de Monteclair reduced Loulie’s categories to two, recognizing the

redundancy of the quadruple meter with the duple and proposing only the duple and the triple

meters, both of which might be fast or slow and thus determine a different accenting.27

“If the song begins with an irrational proportion [3/1, 3/2, 3/4, 3/8], most of the new musicians omit the
mensural sign of quantity, and use only proportions as numbers alone. This is not without cause, because
the [denominator] of the prescribed irrational proportion already has the power to indicate the length of
the measure: consequently, the mensural sign is superfluous, unnecessary, and should be abolished.”
While in 1714, the time signature is still explained as a proportion, the omission of the mensural sign is justified in
response to the seeming meaninglessness of the proportional interpretation of the signature.
25
Etienne Loulie, Elements or Principles of Music, trans. Albert Cohen (Brooklyn,
1965), 26-33 and 59-62 (quotation on 61).
26
Houle, Meter in Music, 36. Loulie’s system of metrical classification defined six types of measures:
duple, triple, quadruple, compound duple, compound triple, and compound quadruple (in modern usage, the
quadruple classifications are considered redundant to duple and thus not typically considered).
• Duple meters were 2, cut-C, and 2/4.
• Triple meters were 3/1, 3/2, 3/4, 3/8, 3/16, and 3 (which Loulie equated to 3/4).
• Quadruple meters were C and 4/8, as well as cut-C again (the redundancy already apparent).
• Compound duple meters were 6/4, 6/8, and 6/16.
• Compound triple meters were 9/4, 9/8, and 9/16.
• Compound quadruple meters were 12/4, 12/8, and 12/16.
27
Michel Pignolet de Monteclair, Nouvelle methode pour apprendre la musique (Paris, 1709), 10. For
example, in a slow tempo the hand beats every note in the measure, whether duple or triple. However, when the
tempo is very fast, the hand compensates by omitting unaccented strokes (the second in the case of duple, the second
and third in the case of triple). When it is not possible to beat every note in a bar, measures are classified as
“compound” (“composée”). See Monteclair, Nouttelle methode, 15. Pierre Dupont and Francois David, and Jean-
Jacques Rousseau, as well as the Italian Vincenzo Manfredini all followed Loulie’s and Monteclair’s reduced
classification. See Pierre Dupont, Principes de musique (Paris, 1718), 16; Francois David, Methode nouvelle (Paris,
1732), 22-28; Jean-Jacques Rousseau, Dictionnaire de musique, 1768, s.v. “Mesure,” vol. 1, 417-23; and Vincenzo

161
 On the British Isles, the sentiments of Morley and Purcell resonate into the 18th

century. Alexander Malcolm in his A treatise of musick; speculative, practical and historical

from 1721 decries in a nearly two-century long castigation of proportional music that music

should “afford us pleasure” and must “not [be] difficultly perceived” due to “part of object[s]

[being] numerous” and “the relations perplexed.” In whole Malcolm states:

Things that are designed to affect our Senses must bear a due Proportion with them;
and so where the Parts of any Object are numerous, and their Relations perplext, and
not easily perceived, they can raise no agreeable Ideas; nor can we easily judge of
the Difference of Parts where it is great; therefore that the Proportion of the Time of
Notes may afford us Pleasure, they must be such as are not difficultly perceived: For
this Reason the only Ratios fit for Musick, besides those of Equality, are the double
and triple, or the Ratios of 2:1 and 3:1; of greater Differences we could not judge,
without a painful Attention.28

After limiting music to ratios of double or triple proportion, Malcolm goes further to add that

measures cannot be divided into anything but 2, 3, and multiples of them, excluding macro-

subdivisions of any greater primes like 5 or 7:

The Measures are only subdivided into 2 or 3 equal Parts; and if there are more, they
must be Multiples of these Numbers as 4 to 6 is composed of 2 and 3; again observe,
the measures of several Songs may agree in the total Quantity, yet differ in the
Subdivision and combination of the lesser Notes that fill up the measure. . . .29

The last remnants of pentadic or higher prime subdivisions appears as an aside in

James Grassineau A musical dictionary from London in 1740. Within a relatively standard

discussion of triple meter signs (approximately 20 in all), mostly in agreement with

contemporary Continental theories, Grassineau identifies some “obsolete” signs: the 5/2 and

7/2 meters discussed by Penna. In reference to these strange signs Grassineau writes that

Manfredini, Regole armoniche o sieno precetti ragionati (Venice, 1775), 2-3. See also Houle, Meter in Music, 36-
37.
28
Alexander Malcolm, A treatise of musick; speculative, practical and historical (Edinburgh, 1721), 390.
29
Ibid., 397-398.

162
“These raising some difficulty and confusion were rejected, and not admitted into the number

of mixed triples.”30 This would be the last reference to such “difficult” proportions until the

19th century, wherein composers would reinvigorate interest in non-dyadic/triadic macro-

metrical subdivisions.31

The 19th Century and New Horizons of Rhythmic Diversity

By the close of the 18th century, time signatures as we use today replaced almost all

mensural signs, albeit C still existing as a seemingly trivial stand-in for 4/4. In many ways,

Curt Sachs’ poetic summary of the next century, the 19th century, and the Age of

Romanticism, aptly characterizes that age’s primary concerns, aesthetics, and neglects:

Early Romanticism… strove for feeling and longing rather than action, for personal
experience rather than objective remoteness, for the mystic depths of night and death
rather than the open clarity of day and life. In the quest for emotional atmosphere, it
concentrated on the finer shades of harmony and orchestration, but neglected the
driving… force of rhythm.32

Rhythmic diversity was in such decline that, by the end of the fourth decade of the century,

Hector Berlioz regularly and pessimistically bemoaned the “vulgar” state of rhythm in music. In

30
James Grassineau, A musical dictionary (London, 1740), 302.
31
Interestingly, there is one extant theory from the 18th century that is bold enough to suggest the infinite
variability of proportional rhythmic diversity, perhaps a variability more radical and truly infinite than Tinctoris or
Gaffurius ever imagined. However, this theory does not come from the established musical institutions and figures
of the age. Rather, this theory comes from one of the most significant mathematicians of the 18th century and of all
time: Leonard Euler. Euler’s theory of harmony and rhythm marries harmonics with rhythm in the same manner as
done by Henry Cowell in the early-20th century. Consequently, Euler could recognize not only an infinite variety of
harmony but also rhythm. Furthermore, this theory could understand rhythm and the transformation of rhythm as a
process similar to harmonic progression and tuning. However, since Euler’s theories were unknown to music
theorists, composers, and performers of his day, we will not consider them as a musical “development” of his day,
since it existed nearly entirely outside the purview of music until its rediscovery in the late 20th century with no
evidence of any, even the most modest, applications. For more information on Euler’s theory of harmony and
rhythm see both Roger Mathew Grant, "Leonhard Euler's Unfinished Theory of Rhythm" in Journal of Music
Theory. 57, no. 2 (2012), 245-286 as well as my own article on the topic: Jordan Alexander Key, "Euler, Cowell,
Polyhedra and the Music Genome" (The Neglected Composer, 2019).
32
Curt Sachs, Rhythm and Tempo, 331.

163
his À travers chants, a collection of some of the composer’s writing compiled in 1862, Berlioz

proclaims the ability to perceive subtle rhythmic forms “one of the rarest” faculties, but it is a

faculty which is unfortunately untaught, leading to rhythm’s atrophy in his generation:

[We] don't teach the musician to find beautiful rhythmic forms; the particular faculty
which enables him to discover [such rhythmic forms] is one of the rarest. Rhythm, of all
parts of music, seems to us today to be the least advanced.33

Berlioz’s complaints over the languishing state of rhythm were not merely limited to abstract

grandiloquence; he even targeted such idols of the day as Rossini in his review of Guillaume Tell

in 1834, wherein he wrote with polite antipathy that “[Rossini] would have done well to abandon

the square-cut rhythms.” While he offers some positive remarks to the overture, he qualifies such

praise by writing that the overture

[lacks] originality in theme and rhythm, and despite its somewhat vulgar use of the bass
drum, [is] most disagreeable at certain moments, constantly pounding away on the strong
beats as in a pas redouble [a quick march] or the music of a country dance.34

Berlioz goes on at a further point in his review of the work to imagine a conversation between

himself and Rossini, wherein he chastises Rossini for unapologetically and lazily foisting a

hackneyed rhythmic construction upon the audience:

…the piece as a whole would be captivating were it not for the torture inflicted upon the
listener who is at all sensitive by the innumerable strokes of the bass drum on the strong
beats, whose effect is the more unfortunate since they again call attention to rhythmic
constructions that are completely lacking in originality. To all this I am sure that Rossini
will reply: ‘Those constructions which seem to you so contemptible are precisely the ones
that the public understands the most readily.’

‘Granted,’ I should answer; ‘but if you profess so great a respect for the propensities of
the vulgar, you ought also to limit yourself to the most commonplace things in melody,
harmony, and instrumentation. This is just what you have taken care not to do. Why, then,

33
Hector Berlioz, À travers chants: études musicales, adorations, boutades et critiques (Paris, Michel Lévy
frères, 1862), 11. “Division symétrique du temps par les sons. On n’apprend pas au musician à trouver de belles
forms rhythmiques; la faculté particulière qui les lui fait découvrir est l’unedes plus rares. Le rhythme, de toutes les
parties de la musique, nous paraît être aujourd’hui la moins avancée.”
34
Strunk, Oliver. Source Readings in Music History (New York: W.W. Norton, 1998), 1128

164
 do you condemn rhythm alone to vulgarity?... No, frankly; excuses of this kind are
unacceptable. You have written a commonplace rhythm, not because the public would
have rejected another, but because it was easier and above all quicker to repeat what had
already been used over and over again than to search for more novel and more
distinguished combinations.’35

Despite Berlioz’s disparaging remarks on the state of rhythm among his contemporaries

and during the 19th century generally, we should take note of some interesting exceptions. While

no composer entirely broke free of the dyadic hegemony during this century, many composers

began to point to a needed diversification in the rhythmic dimension of music, many seeking

such diversity in the expansion of partitionings within dyadic hierarchies rather than employ

non-dyadic (or non-hexadic) hierarchies. While such partitions are unremarkable by standards of

rhythmic exceptionalism in the 20th century, such prime partitions of the whole-note (or FTU)

into five, seven, eleven, thirteen, or even higher prime factors within a dyadic rhythm tree are

exceptional in the 19th century and give evidence to a mounting reemerging interests in rhythmic

diversification.

Berlioz himself was one early exception to the rhythmic “dearth” of the 19 th century. Not

only does Berlioz begin to employ asymmetric rhythmic groupings in his music, suggestive of

larger partitionings of the FTU, but he also directly prescribes such large non-dyadic/triadic

partitions in some of his works. As early as 1838, it can be argued that Berlioz uses asymmetrical

beat patterns in the overture to his opera Benvenuto Cellini, which begins with a rhythm

seemingly contrary to the alla breve prescribed by the composer; an examination of the work

shows the pattern 3+2+3+2+2+2 in the brass choir. By 1854, Berlioz is unequivocally

prescribing non-dyadic/triadic partitions of the FTU in his L’Enfance du Christ, wherein he calls

for a 7/4 time signature.

35
Ibid., 1133.

165
 The richness and eccentricity of Berlioz’s rhythmic conceits did not escape his

contemporaries. Robert Schumann commented on the rhythmic complexity of Berlioz’s

Symphonie Fantastique in 1835 in the his review of this work in Neue Zeitschrift für Musik,

wherein he wrote that “[t]he modern period has perhaps produced no other work in which equal

and unequal mensural and rhythmic relationships have been combined and employed so freely as

in this one.”36 While Symphonie Fantastique presents us with no rhythmic complexity

commensurate to The Baldwin Commonplace Book or to rhythmically experimental composers

of the 20th century, Berlioz’s rhythmic aesthetic, both in words and deeds, called his

contemporaries to imagine more interesting possibilities for the rhythmic dimension of music,

possibilities which all composers of the past two to three centuries were likely unaware of.

The 1830s were witnessing a swing in rhythm’s aesthetic pengulus. Alongside Berlioz,

Spohr was also producing poly-metered works, albeit still under the guise of dyadic hierarchies.

In 1832, Spohr’s symphonic Tongemalde, Op. 86, Die Weihe der Tone, included a section

wherein the orchestra was divided into two rhythmic strata: every three 3/8 measures of one

stratum coincided with two 9/16 measures in the other. Concurrently, Schuman and Chopin

explored polyrhythm; Sachs notes that in the hands of these two composers “the conflicting

coincidence of different rhythms reached a new heyday…. [in] their pianistic polyrhythm,

[wherein] we can easily distinguish between two varieties: coincident beats [with] conflicting

accents; and… conflicting beats [with] coincident accents.”37

36
Schumann, Robert. Neue Zeitschrift für Musik (Leipzig: J.A. Barth, 1835). See also John Schuster-Craig,
"Schumann Encounters Berlioz: The Symphonie fantastique and Schumann's String Quartets, Op. 41" (Faculty
Scholarly Dissemination Grants, 2010), 105.
37
Curt Sachs, Rhythm and Tempo, 335.

166
 In 1837, Schumann’s piano work Des Abends, No. 1 in the Phantasiestiicke, Op. 12

employs two time signatures - 3/8 over 6/16 – which allow for coincident individual beats but

present an opposing three melodic accents to only two accompanimental (namely, demonstrating

Sachs’ coincident beats with conflicting accents). Inversely, Chopin’s F minor etudes from his

Drei Neue Etuden presents quarter triplets over eighths (3:4) between the two hands, creating

two coincident accented beats per measure with the unaccented beats indivisibly non-coincident

(namely, demonstrating Sachs’ conflicting beats with coincident accents). This same

phenomenon is also present in another F minor Etude (Op. 25, No. 2) with 12/8 over 6/4.38 Of all

composers of the 19th century, Chopin is perhaps the most substantial counterexample to the

dyadic hegemony of the age since he was the composer to perhaps most frequently employ

extensive uses of indivisible rhythmic subdivisions; however, Chopin’s purview hardly reached

beyond the juxtaposition of dyadic and triadic hierarchies, since his use of polyrhythms generally

set some power of 2 against some power of 3.

While the impetus to partition the whole-note into values greater than 4 was given by the

French as early as the 1830s,39 quintuple time became an eccentric fashion among elite

38
In the subsequent generation, Brahms’ would follow Chopin’s example. In 1866, Brahms’ makes both
beat and accent disagree in his Paganini Variations, Op. 35, wherein the time signatures 2/4 and 3/8 work against
one another.
39
Of course, there are examples of pentadic, heptadic, and even hendectadic partitioning of the FTU from
the Renaissance. Even after the decline of rhythmic complexity, rare examples of pentadic partitioning of the FTU
appear in post-Renaissance music (albeit still under dyadic subdivision hierarchy). For example, Handel used
quintuple time as early as 1732 in his opera Orlando (Act III, scene 11), Furthermore, Benedetto Marcello (1686-
1739) prescribed quintuple time in his cantata for soprano and harpsichord, wherein he not only commented on the
piece’s difficulty when he wrote “Senza gran pena non si giunge al fine” (“Without hard labor one cannot get to the
end”) but also noted the novelty of its design when, in the cantata’s final aria which is in 5/4 time, he wrote, “di
novita stravagante” (see Benedetto Marcello, La stravaganza, cantata a una voce sola, aria No. 2). At least two
examples of quintuple time can also be found from the 17th century: Heinrich Albert’s German arias (1638-1650),
wherein five, six, and four beats alternate. See Arnold Schering, Geschichte der Musik in Beispielen (Leipzig, 1931),
233 f; and Christoph Demantius’ Conviviorum deliciae (1608), wherein one of the intradas is in quintuple time (see
again Schering, Geschichte der Musik, 158 f).

167
Table 4-1. Examples of quintuple, septuple, and hendectuple time by Romantic composers.
Metric Time Country Date Composer Piece Notes
Quintuple Russia 1836 Glinka Life for the Czar, woman’s
Time (5x/2n) chorus
1873/1895 Rimsky- Pskovityanka, III.1
Korsakov
1874 Mussorgsky Boris Godunov, Act II alternation
between 3/4 and
5/4 (3+3+2)
1893 Tchaikovsky Pathetique Symphony,
“Allegro”
France 1825 Boieldieu La Dame blanche, II.2,
Cavatina
1892 Saint-Saens 2nd piano trio in E minor
1897 d’lndy 2nd string quartet, Op. 45
unknown Pierne Sonata for Violin and Piano, (6/8 in violin
Op. 36, mvt. 1 “allegretto” against 10/16 in
piano)
1893 Debussy first string quartet, mvt. 2 Simultaneous use
of 15/8 and 9 + 6
1912 Debussy Rondes de Printemps also with 15/8
1864 Gounod Mireille, II “chanson de Simultaneous use
Magali” of 15/8 and 9 + 6
Germany unknown Brahms Agnes (lied), Op. 59
unknown Wagner Tristan und Isolde, III.2 alternates 3/4 and
2/4
Poland unknown Chopin C min. Sonata, Op. 4, mvt. 3
unknown Paderevsky Chants des Voyageurs, No.4,
“Andantino mistico”
Finnish 1899/1900 Sibelius Finlandia 4/4 time w/
accents every 5
beats

Septuple Austrian 1904 Mahler Sixth Symphony, “Scherzo” alternation

Time (7x/2n) between 3/8 and
4/8
Hungarian 1854-1861 Liszt Faust Symphony alternation
between 3/4 and
4/4 (= 7/4)
Swiss c. 1859 Raff 2nd Grosse Sonate für periodic
Pianoforte und Violine, alternation of 3/4,
“Sehr rasch” 5/4, 4/4, and 2/4 (=
14/4)
English 1898 Elgar Caractacus, “Lament” Alternates 3/4 and
4/4
Hendectuple Russian 1874 Mussorgsky Boris Godunov, first chorus alternates 2x 3/4's
Time (“Promenade recur”) and 5/4 (= 11/4)
(11x/2n) 1882 Rimsky- Snow Maiden, final chorus prescribed 11/4
Korsakov

168
composers across Europe over the ensuing century. Examples of quintuple and other hyper-

tertian-prime times from this period include those in Table 4-1 above.

The above examples are not comprehensive, but they do demonstrate a growing interest

in expanding rhythm’s capabilities towards the beginning of the 20th century. By the early-20th

century, the use of asymmetrical times was relatively common among prominent classical

composers, so much so that its use becomes rather unexceptional in light of other rhythmic

innovations of that century. Consequently, we will not linger long on the use of asymmetrical

combinations of meters and the use of meters with prime partitions greater than the FTU. It is

worth noting some examples to demonstrate the continued interest in this arena opened by prime

partitions greater than 3, but a detailed discussion of them is not our prerogative, since such

metrics, despite their prime partitions greater than 3, are still operating mostly in either dyadic or

triadic rhythmic hierarchies, albeit the one (usually first) subdivision in which the prime

partitioning takes place.40

40
It might be worth noting a prominent time signature reform during the early 20 th century, a reform which
continues to be used, albeit not as much as the common practice. Music educator Émile Jaques-Dalcroze (1865 -
1950) attempted to improve time signatures by replacing the often-arbitrary time unit by a simple beat value, a
change more significant to the interpretation of compound time signatures. His reform structures time signatures like
fractions as before; however, numerators indicated the number of beats and denominators the semiotic note-value of
the beats. For example, when a 6/8 is a compound duple meter and has two beats per measures (rather than 6 as the
traditional time signature would suggest), the signature should be written as 2/”dotted-quarter note,” or similarly
12/16 would be 4/”dotted-eighth note.” However, this innovation, however much it still persists today, is not
relevant to our discussion herein. Curt Sachs, Rhythm and Tempo, 361. For early 20th century reforms on the
notation of rhythm and time signatures see also Jaques-Dalcroze, Eurhythmies, New York, 1930; Rhythm, music,
and education, New York, 1921. Also J. Pennington, The importance of being rhythmic, New York and London,
1925. Another suggestion to improve time signature, less popular but perhaps more interesting and powerful, come
from Carlos Vega in Argentina. In his reform, the time signature’s denominator indicates the note-unit (halves,
quarters, eighths, etc.) and the numerator indicates the number of those units in a measure. In addition to this basic
premise, the values in the time signature are braced on the right by a smaller figure that indicates the subdivisions of
those beats. For example, standard compound-duple 6/8 gets a figure “3” to its right because the two macro-beats
are ternary; similarly, 3/4 could equally appear as a 6/8 with a figure “2,” because of 3/4’s binary macro-beats. It
becomes apparent then that Vega could use this system to indicate any value for the subdivision of the macro-beats
of his measures; however, he never does this and only ever uses numbers that make sense given the number of
eighth-notes that should be in the measure. For example, Vega could have a 3/4 measure with a “5,” which would
indicate that each quarter note should be divided into a 1/20th-note. See also Carlos Vega, Fraseologia, in La Musica
popular argentina (Buenos Aires, 1941, Vol I), 48-49.

169
 Charles Ives: The Rebirth of Rhythmic Diversity in the 20th Century

The beginning of the 20th century witnessed the rapid expansion and hyperbole of mixed

meter and asymmetric meter. Curt Sachs’ poetic assessment of these few pivotal decades

between 1900 and 1930 and the publication of Henry Cowell’s New Musical Resources

colorfully points us to the inevitable change of fads that ultimately begat the more exceptional

breaks from rhythmic traditionalism we will examine in the mid-20th century:

When the 20th century buried Romanticism and…, the tradition of five hundred years of
musical evolution, clean-cut divisive rhythm with its neat 4/4, 3/4, 6/8 [lost] its hold.
Following the law of the pendulum, other forms of rhythm [came] to the fore again.41

These “other forms” during these three formative decades were capricious use of syncopation,

the consistent use of additive quintuple or greater prime-numbered times, and the shifting from

one irregular group to another in successive measures. Naturally, nearly all of this was in the

guise of dyadic-rational time hierarchies. For example, Bartok in his Mikrokosmos piece No. 140

used 35 time signatures that continually changes between 3, 4, 5, 6, 7, 8, 9 eighth-notes, and his

pieces from the same collection labeled “in Bulgar rhythm” along with his Outdoors, No. 4

(1926) continually change between duple, triple, quintuple, and septuple time. Stravinsky’s Fire

Bird and Rite of Spring ballets (1910-1913) also have numerous examples of rapid alternation

between duple, triple, quintuple, and septuple time (metric and rhythmic).

From Russia, with Dimitri Shostakovich’s regulated alteration of binary and ternary

measures in his 6th and 7th symphonies in 1939 and 1941, to the United States of America, with

Roger Sessions’ unpredictable shifts to 4/8, 5/8, 6/8, 7/8, 8/8, and 7/16 in his Piano Sonata in

1931, what had been “experiments” and exoticisms in 19th century music had become the

expectation of 20th century composers. Such techniques had become so typical by the middle of

41
Curt Sachs, Rhythm and Tempo, 370.

170
the century that composer Daniel Jones (1912-1993) noted in his 1950 commentary on

contemporary rhythmic experiments that “indeed, the occurrence of unusual looking time-

signatures, especially in ambiguous form, is one of the commonplaces of modem music.”42 For

such reasoning, mixed meters will not be addressed in detail. While it formed much of the

foundation upon which the new Renaissance of Rhythm in Western Classical Music43 would be

built, mixed meters ultimately became trivial in the light of the experiments to come after.

Such later experiments of the early-20th century, particularly those in the United States,

were already foreshadowed by Charles Ives (1874 – 1954) by the late-19th century. As early as

1900, Charles Ives displayed interest in extreme rhythmic schemes in From the Steeples and the

Mountains for brass and bells.44 Over the first fifteen years of the century, Ives presaged nearly

all rhythmic experiments of the 20th century.45 Over these formative years, Ives explored layered

time signatures, multifarious use and layering of complex tuplets, embedded tuplets, and the use

of tempo to change the perception of rhythm (e.g. metric modulation and layered tempos).

Charles Ives comments on his interest in multifarious rhythms in his observations on his

1910 work Over the Pavement in a memo. Ives remarks on the natural rhythms of the city:

42
Daniel Jones, Some metrical experiments, in The Score, Nr. 3, June, 1950, 32-48. In this article, Jones
himself prints a sonata for three kettledrums, Jones uses his own scheme of unpatterned asymmetrical metrical
patterns, using his own proposes system of “complex meters,” which was fully developed in this "Sonata for Three
Non-Chromatic Kettle-Drums" (1947). In his 1950 article, he described this system: “The unifying element of fixed
pattern is present, but the pattern itself is asymmetrical, therefore with a powerful means of satisfying structural
requirements there would seem to be possible both a greater variety and a greater subtlety in the rhythm-metre
relationship.” It is also interesting to note that in this same year – 1950 – Elliott Carter wrote his famous Suite for
Tympani, which would itself lay the groundwork for a different system of rhythmic diversity we will examine later.
43
Herein, we call the 20th century the “Renaissance” of rhythm, since the Renaissance of Western music
was not a rebirth in rhythmic notation but rather a continued development of it since the 14th century, a development
which had nothing to do with notational practice from Ancient Greece. Since we witnessed a decline in the
sophisticated notation of complex rhythms during the 17th and 18th centuries with a rekindled interested in rhythmic
diversity in the 19th century and a subsequent return to rhythmic complexity in the 20th century, we arguably see the
“rebirth” of sophisticated rhythmic notation in the 20th century rather than the 14th through 16th century.
44
See measures 38-41.
45
The one major exception being developments in time signature, which will be discussed later.

171
 In the early morning, the sounds of people going to and fro, all different steps, and
sometimes all the same – the horses, fast trop, canter, sometimes slowing up into a
walk… an occasional trolley throwing all rhythm out (footsteps, house, and man) – then
back again. I was struck with how many different … beats, time, rhythms, etc. went on
together – but quite naturally, or at least not unnaturally when you got used to it….46

We can see the consequence of these observations in bars 59-61 of Over the Pavement, wherein

there are not only proportions like 3:5, 9:5, 9:10, 2:3, 4:3, 12:9, and 6:9, unique relative to

common practice, but also those same tuplets within a pentadic partitioned time signature with a

metric modulation47 based on the 9:10 8th-note proportion (see Example 4-1).

Even earlier than Over the Pavement, Ives was whimsically employing advanced tuplets

in complex stratification in his Scherzo: Holding Your Own from 1903 to generate relatively

unique rhythmic proportions for the early 20th century (see Example 4-2). In this brief excerpt we

can see 3:2 (three levels), 3:4 (two levels), 9:2, 9:12, 5:1, and 20:3. By the second decade of the

century, Ives’ use of layered tuplets reached levels of complexity unseen since the 16th century

with works like Tone Roads No. 3, wherein one can see [Link] all contextualized with complex

dyadic syncopations, and “The Housatonic at Stockbridge” from Three Places in New England,

wherein one can see [Link] and [Link] along with tuplets crossing bar lines and filling non-

dyadic partitions of the macro-beat (see Examples 4.3 and 4.4 respectively)

By 1908 in The Unanswered Question, Ives is even using nested tuplets to non-dyadically

redefine the tuplet’s subdivision beyond its initial subdivision. In Figure 4-5 we see that Ives is

showing that the second subdivision level within the tuplet is, in fact, not the dyadic subdivision

of the “third-note” (“half-note triplet”), which would be a sixth-note, but rather the triadic

subdivision of the third-note, which is a ninth-note.

46
Charles Ives and John Kirkpatrick, Memos (New York: W.W. Norton, 1972), 62-63.
47
We will discuss metric modulation later in this chapter in our discussion of Elliott Carter.

172
Example 4-1. Charles Ives Over the Pavement (1910); bars 59-61.48

48
Charles Ives, Scherzo: Over the pavement (New York: Peer International, 1954)..

173
Example 4-2. Charles Ives, Scherzo: Holding Your Own (1903); bars 22-28.49

49
Charles Ives, Scherzo, for 2 violins, viola, and cello (New York: Peer International Corp, 1958).

174
Example 4-3. Charles Ives, Tone Roads No. 3 (c. 1915); bars 18-21.50

50
Charles Ives, Tone roads no. 3 (New York: Peer international, 1952).

175
Example 4-4. Charles Ives, “The Housatonic at Stockbridge” from Three Places in New England
(c. 1911); bars 1-4.51

51
Charles Ives, Three places in New England, III. The Housatonic at Stockbridge (New York: Mercury
Music Corporation, 1935).

176
Example 4-4 (continued)

177
Example 4-5. Charles Ives, The Unanswered Question (1908, rev. 1935); bars 47-53.52

Concurrent with his adventurous use of tuplets, Ives also explored proportional,

asymmetric, and syncopated polymeter. In 1909, his Largo Risoluto No. 1 uses relatively wild

asymmetric and syncopated polymeters throughout. While most of the music maintains dyadic

hierarchies (though he uses tuplets sporadically), the meters in each instrument are constantly

changing, often asynchronously with the other instruments (see Example 4-6 and 4-7). In his

slightly later String Quartet No. 2, the second movement displays proportional polymeters,

giving such metrical macro-proportions such as 3:8, albeit all couched within a dyadic rhythmic

hierarchy (see Figure 4-8).

Again, in The Unanswered Question, Ives displays an intriguing manner of polymeter

through the use of disproportionate tempos. Examining the score in Example 4-9, one can

observe that there is no seeming coincidence between the meter of the winds and the

trumpet/orchestra. If there was a more quantitative tempo given, we might be able to determine

the precise ratio at work here, but given the measuring by Ives, the rates of the two ensembles

seem rather incommensurate.53 Herein, Ives is using stratified tempos as a method to layer

relatively simple dyadic hierarchies in a more complex polyrhythmic relationship.

52
Charles Ives, The Unanswered Question (New York: Southern Music Pub. Co, 1953).

178
 By 1916, Ives was employing all of these techniques freely in his compositions. His work

from that year, In Re Con Moto et al, demonstrates all of these techniques, barring the

simultaneous use of differing tempos; it also demonstrates a shared interest in larger, non-dyadic

measure partitionings, as examined earlier in this chapter. In Example 4-10 from bars 7-12, we

can see such use of large and non-dyadically partitioned measures (15/8, 21/8, and 33/8) along

with creative use of additive and subtractive rhythms. Example 4-11 from bars 24-25

demonstrates a syncopated (by one quarter note), asymmetric polymeter – 7/4+5/4 against

(3x)4/4 – along with the free use of tuplets.

Example 4-6. Charles Ives, Largo Risoluto No. 1 (1909); bars 10-22.54

53
though it seems there is either 1) 3 measures of flute choir or every 2 of orchestra with one beat offset,
providing a polymetric syncopation, or 2) the proportional tempo between these two choirs is highly indivisible.
54
Charles Ives, Largo Risoluto, Piano Quintet No. 2 (New York: Peer International, 1961).

179
Example 4-7. Charles Ives, Largo Risoluto No. 1; bars 10-22.55

Example 4-8. Charles Ives, mvt. II “Argument” from String Quartet No 2 (1913); bars 17-30.56
55
Ibid.

180
Example 4-9. Charles Ives, The Unanswered Question; bars 14-19.57

56
Charles Ives, String quartet no. 2 (New York: Peer International Corp, 1980).
57
Ives, The Unanswered Question.

181
Example 4-10. Charles Ives, In Re Con Moto et al (1916); bars 7-12.58

Example 4-11. Charles Ives, In Re Con Moto et al; bars 24-25.59

58
Charles Ives, In re con moto et al; for piano quintet (New York: Peer International Corp, 1968).
59
Ibid.

182
Example 4-12. Charles Ives, In Re Con Moto et al; bar 42-4560

60
Ibid.

183
 By measure 43 in Example 4-12, Ives employs nested tuplets, establishing a two-level

triadic hierarchy of third- and ninth-notes in the strings followed by a pentadectadic (15)

hierarchy in measure 44, wherein each subdivision order of the hierarchy tree can branch to

either pentadically or triadically. In this example, it is also worth noting Ives’ explicit (via slurs)

palindromic prime groupings of rhythmic units in the upper voices (sequentially 2, 3, 5, 7, 11, 7,

5, 3, 2 16th-notes).

The climax comes to fruition in measure 59-61 (see Example 4-13), with the wildest array of

rhythmic stratification yet. The tuplets used herein are so multifarious, it is difficult to define the

precise proportional scheme at work, since the common multiple of these various stratified prime

tuplets, along with the various prime-valued nested tuplets, amounts to an inhumanly

performable fundamental minimum subunit. In Re Con Moto et al, we see the below listed

rhythmic proportions (See Figure 4-4)

• Non-Dyadic Tuplets:
o 10:8
o 7:4
o 3:4
o 10:6
o 11:6

• Non-Dyadic Nested Tuplets:

o (11:8):(3:4) = 33:32
o (10:8):(3:4) = 30:32 = 15:16
o (14:8):(3:4) = 42:32 = 16:21
o (3:2):(3:4) = 9:8
o (5:2):(3:4) = 15:8
o (5:2):(11:6) = 55:12

Figure 4-4. Rhythmic proportions found in Charles Ives’ In Re Con Moto et al, bars 59-61.

184
Example 4-13: Charles Ives, In Re Con Moto et al; bars 59-61.61

61
Ibid.

185
Furthermore, there are also the implied proportions with the various combinations of the

proportion in Figure 4-4 are played concurrently (i.e. 10:7, 7:3, 10:3, 11:10, etc.). The

combinations, while finite, are staggering in comparison to the limited set of possibilities under a

completely dyadic hierarchy; the aural effect is extremely rich and complex.

By 1916, Ives was extending the limits of rhythmic notation in the common system

available to him. Some of his metrically off-set non-dyadic tuplets within non-dyadic partitions

of the whole-note reached the limits of not only accurate performability but also clear and

accurate notation. Ives successfully explored all aspects of prescriptive rhythm available to him;

however, unavailable to him was the free use of non-dyadic rhythmic notes, not bounded by the

limits of the tuplet bracket, able to freely move within a measure and intermingle with dyadic

rhythmic notes. Furthermore, Ives, while having used the most exotic time signatures available to

him, had not explored the possibility of new time signatures that could not only more concisely

express some of his rhythmic conceits but also allow for unexplored rhythmic constructs.

Ives’ music, however obscure during its time of composition, prepared the path for

further experiments in the 20th century, ultimately inspiring many of the subsequent composers

who would follow Ives’ new rhythmic paths to their full extent, often necessitating new

notational devices to express these exceptional rhythmic proportions, structures, and hierarchies.

Beswick, Cowell, and Bartolozzi: The Early-20th Century and the Re-emancipation of
Rhythm

While composers across the world were beginning to discover the limitations of the

classical system of music notation handed to them from the 17 th and 18th centuries and were

attempting to uncover new rhythmic structures within that limited system, some were attempting

to emancipate themselves from these limitations not by re-contextualizing them under new

186
macro-groupings of dyadic hierarchies but by embedding non-dyadic hierarchies into the

prevailing dyadic system. While many, such as Charles Ives in the United States and Egon

Wellesz62 in Austria, were using myriad tuplets in multifarious combinations to create rich, often

indivisible, poly-rhythmic strata, some were beginning to imagine a better system altogether, a

system that would emancipate non-dyadic rhythms (often referred to as “irrational” rhythms by

contemporaries)63 from subservience to dyadicism and the prison of the tuplet bracket, wherein

any such non-dyadic rhythms have been kept in bondage (particularly those greater than triple

subdivisions) arguably since their notational inception under proportional mensuration in the 15 th

century.

The first such pioneer to imagine a notational system less subservient to dyadicism was

Laura Beswick, a relatively unknown musical presence in America during the early 20 th century

whose “invention” in 1900 points to a glaring flaw in our still used notation system and attempts

to offer a solution to that flaw. In 1903 the US Patent Office trademarked a little-known addition

to music notation by Laura Beswick,64 which gave a method to notate the tripartite division of

the whole-note into third, sixth, twelfth, twenty-fourth, and forty-eighth notes.65 Furthermore, her

proposed addition to the standard system, by being an “addition,” has the advantage of melding

nonstandard note values with conventional durations and freeing a composer from the obligation

62
Here, we particularly think of Egon Wellesz’s interesting a creative use of superimpositions of half-note
triplets in 5/4 time in the second act of his opera Die Bakchantinnen in 1930.
63
An interesting appellation that points to the long denigration of such rhythms, stretching as far back as
the 16 century with Morley’s, Glarean’s, and Zarlino’s denunciation of such “surly” rhythms that defy any
th

“rationality.” Why such rhythms should be “irrational,” since they are intrinsically mathematically rational (as
Tinctoris and Gaffurius would have recognized) is clearly a result of “rational” thought, but limited aesthetic
inclinations. Consequently, we will not refer to such rhythmic structures as “irrational” herein, since this branding
lacks rational consideration in itself.
64
See Laura H. Beswick, Musical Notation (U.S. Patent 733,351, U.S. Patent and Trademark Office, 1903).
65
Significantly antedating Henry Cowell’s similar unorthodox note values by at least 14 years.

187
to group tripartite divisions into complete triplet groupings, possibly spreading tripartite

divisions (triadic hierarchies) freely among bipartite subdivisions (dyadic hierarchies). Of her

own system, Beswick writes into the patent that

Prominent objects of the invention are to make possible the indication and use of certain
tone lengths and pauses not at present employed in the rendering and noting of music; to
facilitate the designation of other tone lengths at present so employed; to provide simple,
effective, and practical means for designating such tone lengths and pauses; to secure
certain novel, desirable, and striking effects in both vocal and instrumental music not
possible to attain under the system of notation at present in use, and to arrange for the
better, in fact, for the substantially perfect adaptation of music to words in vocal music.66

The addition Beswick adds to standard music notation is given if Figure 4-5. Herein,

Beswick notes that her notation employs a “style similar in general arrangement to the notes of

the musical system now in common use[:] … notes having a head and a vertically arranged stem

extending upwardly or downwardly from the head,” and remarks that this is a “preferred

arrangement” since the “employment of notes similar to [standard notes] has the advantage of

being simple and practical, while at the same time it gives a composition in which the two

systems of notation are employed a uniform appearance.”67 Simply, Beswick places dots on the

typical note heads+stems, rather than flags, each dot indicating the progressive halving of the

third note, which appears like a half note surmounted by a dot (e.g. one dot void note = 3rd note;

one dot black note = 6th note; two dots = 12th note; three dots = 24th note; etc.).

While many of the practical musical examples Beswick gives can be simply notated

using traditional triplet figures, she does provide one example in her last figure that cannot easily

be notated using triplets, since the tripartite subdivisions are dispersed among the dyadic rhythms

(see Figure 4-6). The top stave in Figure 4-6 is Beswick’s example; the bottom stave is the

66
Ibid. 1.
67
Ibid.

188
traditional equivalent, which obscures the quarter note in the final two measures by the

requirement to incorporate it into the two necessitated triplet figures.

Figure 4-5. Laura H. Beswick’s free tripartite subdivision of the whole note (notes and rests).68

Figure 4-6. Example given by Beswick of intermingling of dyadic and triadic subdivisions
within a measure using her new system of tripartite subdivision notation (see final two
measures).69

She addresses this unique potential numerous times throughout her patent. She begins

with a cursory explanation of the potential to emancipate oneself from the necessity of the triplet

when she writes:

68
Ibid.. Patent expired, image in public domain.
69
Ibid.. Patent expired, image in public domain.

189
 My improved system of musical notation…. [uses] notes and rests [that] designate their
respective tones and silences individually – that is to say, each note or rest indicates the
length or duration of the tone or silence by itself without any further associated symbol
or symbols in the music-score.70

She later directly notes the problematic limitations of the triplet and its grouping requirements

and reinforces the freedom her system allows when she writes:

In my notation it will be observed that each of the notes and rests designates its tone or
pause length… by itself alone – without refence to any associated symbol or character in
the score. In this the notation differs from the triplet, for in the triplet the time length of
the notes … composing the triplet is indicated by the entire structure – that is to say, by
the three notes…, the slur joining the three together, and the numerals “3” arranged
above or below the slur. Among the results secured by this difference are … convenience
in notation… [and capability of] separating the various notes from one another and thus
permitting their combination in any desired way with other [dyadic or triadic] notes….71

Beswick mentions numerous times throughout the rather extensive patent that this system gives

the composer the capability to reorganize triplet values freely throughout a musical structure,

even choosing to not include all, previously necessary, tripartite subunits. This final observation,

the fact that one could use not only a separated triplet but an incomplete one as well (also

possibly dispersed) under this new system is quite revolutionary. To this end, Beswick writes:

Regarding the second feature [of my system – namely, the ability to separate a triple
grouping], it will be observed that in the triplet all three notes remain together. By the
nature of the combination-symbol [triplet] the [bracket] ties the three notes together, and
the numeral “3” indicates that the three are to be played together. It, therefore, is
impossible [under the traditional notation system] to separate one [tripartite unit] from
the other two or to separate all three from one another. Consequently, if it is desired to
arrange one of the triplet notes on one side of a note of different character and the other
two one the other side of that note this is impossible by the triplet. In my system, however,
this can be readily done….72

70
Ibid.. Here, Beswick refers to the additional symbol necessary to indicate tuplets (a horizontal bracket
above appropriately grouped notes with the non-dyadic subdivision listed on the bracket).
71
Ibid., 3.
72
Ibid..

190
She goes on to point out the capability of choosing any number of triadic subdivisions (3 rd-, 6th,

12th, 24th-notes).

Regarding the third [feature of my notation system] it will be seen that only three or six
or nine or like number of triplet notes can be employed. This must follow from the fact
that the triplet provides for only three notes or multiples of three. In my system, however,
any number of notes can be employed, it being necessary only to arrange for such notes
as are desired and then employ them, be the number two, four five, seven, or any other
number.73

Beswick’s notation allows notes of a tripartite subdivision to be used “in combination with notes

of the present musical system in any desired way,” so that “notes and rests indicating thirds,

sixths, twelfths, and like divisional parts of a whole tone can be associated in any way desired

with notes and rests indicating halves, quarters, eighths, and like divisional parts of a whole

tone.”74 Beswick recognizes the broader implications of her system’s emancipation of tripartite

subdivisions when she proudly proclaims:

[T]he combination of the notes of my system with those of the present system, produces
novel and striking results whose number and variety are unlimited and whose utility is
unquestioned.75

While Laura Beswick’s system was utterly unused and unknown as far as any surviving music

shows,76 it points to a stark lack in music notation at the start of a renewed interest in rhythmic

diversity. The concern over this lack would only continue to grow over the first decades of the

73
Ibid..
74
Ibid., 1.
75
Ibid..
76
While this might be because of Beswick’s sex, which would historically limit her ability to compose
music or have much, if any, composed music taken seriously by the largely male music community, the failure of
this system might also simply be because Beswick entertainingly stipulated in her patent that any piece of music
written using this new system would be rightfully hers. In her patent she writes, “What I claim as my invention is…
[a list of eight things pertaining to her notation and]… 9) As an article, a music-score containing characters adapted
to designate tones or rests, or both, having lengths equal to an even divisional part of a third of a whole tone, in
combination with characters adapted to designate tones or rests or both, having lengths equal to one-half or an even
divisional part thereof of such whole tone.” See Beswick, Music Notation, 4.

191
century, provoking a handful of composers to seek new alternatives to notate rhythmic structures

not easily notated or utterly unable to be notated. The first comprehensive attempt to give a

solution to this mounting problem was Henry Cowell, who created his own system of non-dyadic

rhythmic notation as part of a new piece in c. 1917 – Fabric for solo piano – which would lay the

groundwork for his more comprehensive pan-rational system of rhythmic notation found in his

seminal New Musical Resources, published in 1930.

For many composers of the 20th century, especially those in the United States and those

retrospectively assessing the musical culture of the early-20th century, Charles Ives was the

harbinger of musical diversification and experimentation whereas Henry Cowell was the

cumulative archetype of that diversity. Cowell, in fact, took much of his inspiration from Charles

Ives in his own experimentation, having written the first book on the music of Charles Ives in

1955. Seeing the many aesthetic directions in which Ives’ music pointed, Cowell, as a personal

project assigned by his teacher Charles Seeger during his studies at the University of California,

Berkeley, beginning in the Fall of 1914, decided to outline the avenues down which he felt music

could expand in his book New Musical Resources.77 This book began in its first form in the

second decade of the century, around the time numerous experimental pieces were penned by

Cowell. One such experimental work was Fabric for solo piano, wherein Cowell outlined a

77
Cowell recounts this formative time in his life and discusses the impetus for his experiments in
polyrhythm in the score notes for his Quartet Romantic and Quartet Euphometric. Therein, Cowell writes that “a
graduate student in the physics department at Stanford, with whom [Cowell] discussed the matter [of the similarity
between the overtone series and counter-rhythms], mentioned that he [the physics student] had access to a pair of
sirens. [Their] experiments with two simultaneous sirens showed that if they are tuned in the relationship 3:2, [the
sirens] will sound the interval of a perfect fifth; [however,] if [the sirens] are both slowed down, keeping the same
3:2 relationship, they arrive at a rhythm of 3 against 2, heard as gentle bumps but also visible in tiny puffs of air
through the holes in the sirens, and so easily confirmed.” They discovered, not to their surprise, that if they tuned the
sirens to any other harmonic ratio the sirens could be modulated to demonstrate the proportional rhythmic ratio.
Cowell rightly claimed that this experiment proved that “these ratios express a single physical relationship which is
heard as rhythm when slow and as pitch when fast.” See Henry Cowell, Quartet Romantic and Quartet Euphometric
(New York: Peters, 1974), 1.

192
system of rhythmic notation that would allow for a more “natural” notation of non-dyadic

rhythmic values. While similar in aim to Beswick’s reforms, Cowell’s was more comprehensive,

proposing shaped notes for divisions of the whole note between 2 and 15 parts, all of which were

dyadically sub-divisible and combinable.

Differentiating the methodology of Cowell’s system from Beswick’s was Cowell’s

modification of note heads, rather than the stems/flags, to represent these non-dyadic rhythmic

values. For example, in Cowell’s system, the series of dyadic subdivisions of the “third-note” is

based on a triangular note head. The 3rd-note is a void triangle-note with a stem. This note can be

dyadically subdivided or combined. If combined into a 2/3rds-note, the stem is removed to make

a triangle with no stem; Cowell did not provide any method to combine the 2/3rds-note into a

4/3rds note. If dyadically subdivided, the void notehead becomes black with a stem, which

represents a 6th-note. By adding a flag to the 6th-note, one creates the 12th-note. As is traditional,

with each additional flag, the note is subdivided into two parts. The complete chart provided by

Cowell in Fabric is given below in Figure 4-7.

Cowell recognizes the same deficiency in modern music notation: the inability to

recognize the individual existence of such rhythmic values, even as simple as the 3rd-note. In the

introduction to this short work, Cowell axiomatically states that the whole note is the “unit by

which all shorter time values are measured.” From this axiom, he quickly deduces the limitations

of the dyadic system in which he composes wherein tuplets expressing non-dyadic subdivisions

stand not as integral part of the system, but errant, perhaps “irrational” outliers. To this end,

Cowell writes:

The only regular system of subdividing a whole note is by twos into halves, quarters,
eighths, etc.. If notes of other time values, for instance notes occupying one twelfth of a

193
 whole note are desired, they are called ‘eighth notes triplets’ and written as eighth notes,
with a figure 3 over them…. Why not call them twelfth notes, as would seem natural?78

From this rhetorical question, Cowell goes on to propose that these “irregular time values” be

identified by their “correct names,” rather than the apologetic “tuplet” appellation:

It is here proposed that all these irregular time values be called by their correct names,
according to the part of a whole note they occupy…. Although heretofore not suggested
in notation, it will be seen that third, sixth, twelfth and twenty-fourth notes form a related
series; fifth, tenth and twentieth notes another, and in fact, that a new series can be
formed on each odd number and its divisions by two.79

While rhythmically revolutionary, harkening back both to the proliferation of note-

shaped during the late 14th and early 15th century, as well as the limitless proportions of the

Tincto-Gaffurian system of rhythmic proportions, the lack of vision in Fabric’s rhythmic

notation is glaring and twofold. First, Cowell still binds his system to dyadic subdivisions, no

matter the fundamental subdivision of the whole note itself.80 Furthermore, Cowell still uses

these new note values in complete groupings rather than free arrangements, as suggested and

demonstrated in Beswick’s system. Throughout Fabric, no note value appears without the

completion of its rhythmic “whole” (namely, the whole note summation of all equal subunits).

Thus, the only effect Cowell’s system allows in this piece is a presentation without tuplet

brackets. However, since the note heads are all different, the score is arguably just as

complicated looking, if not more so.

78
Henry Cowell, Fabric: for piano (New York: Breitkopf, 1922), 1.
79
Ibid..
80
Consequently, Cowell’s system outlines mixed mensurations according to my theory of time signature.
For example, the series of 5th, 10th, 20th, 40th notes is really the combination of pentadic and dyadic hierarchies (p=2,
q=5) with a time signature of X/10. Thus, while Cowell sees this as a series of “five” it fundamentally is a series
based on the prime factors of both 2 and 5, and, more specifically, just one branch of this dectadic hierarchy: that
which is always divisible by 2, never by 5. Thus, Cowell’s system is limited significantly in this way. Contrastingly,
my system would allow for either dyadic or pentadic division on any level of a dectadic hierarchy. For example, a
5th-note could subdivide into either a 10th-note or a 25th-note. Significantly, one should note that Cowell’s system
has no allowance for a 25th-note.

194
Figure 4-7. Table of note-shapes and rhythmic values in Henry Cowell’s Fabric.81

The next step in Cowell’s exploration in the new rhythmic possibility afforded by his

shaped-note system was in Quartet Romantic (1915-1917), written concurrently with Fabric, and

Quartet Euphometric (1916-1919). In both of these quartets, Cowell used his system in tandem

with tuplets, often requiring fractional portions of the tuplet to complete the measure of music

given (see Example 4-14 below); however, most of these fractional tuplets merely represent

multi-measure macro-tuplets, which must be tied over from one measure to the other due to the

81
Ibid..

195
strictures of the bar line. For example, the 3 and ¾ half-notes tuplet in the third voice of the first

system could also be written as 15/8 8th-notes for one measure or 15/8 quarter-notes over two

measures. Similarly, the bottom voice of the second system is written as 2 and 2/3 half-notes

tuplet per measure (note the use of the 2/3rds-note and 1/3rds-note) could be written as 8/6 half-

notes over 3 measures. In this work, Cowell rarely intermingles different incompletely

partitioned non-dyadic hierarchical groups.82 Interestingly, in his post-1930 work Rhythmicana,

Cowell would not hesitate to use multi-measure tuplets and avoid the unusual and perhaps

problematic fractional tuplet values (Example 4-15 below).

In 6 Ings from 1922 (after the above-mentioned quartets), Cowell opts for stratified time

signatures to loosely represent indivisible polyrhythms in the first movement and forgoes tuplets

altogether for implied non-dyadic subdivisions via beaming in the final movement, surprisingly

abandoning his thoroughly articulated system from the decade prior. In the first movement,

Cowell juxtaposes 3/4 and 4/4 to create a 6:4 polyrhythm, occasionally resorting to tuplets to

alter the subdivision of the whole-note to 7:4 or 7:8 (see Example 4-16 below). In the sixth

movement, Cowell occasionally provides tuplets when rhythmic subdivisions change, but his

application of tuplets is inconsistent and freely used with implied non-dyadic subdivisions via

the grouping of similar note values via beaming (see Example 4-17 below).

Ultimately, Cowell never realized the complete potential of his system as codified in New

Musical Resources, though he came close in Quartet Romantic. However, his outlining of the

system in his book from 1930 and the further suggestions for rhythmic exploration he provided

therein would spur later developments in 20th century rhythmic diversity.

82
namely, he does not mix various non-dyadic rhythms outside of providing a complete non-dyadic
aggregate

196
Example 4-14. Henry Cowell, Quartet Romantic, bars 10-15.83

Example 4-15. Henry Cowell, Rhythmicana, bars 10-12.84

83
Henry Cowell, Quartet Romantic and Quartet Euphometric (New York: Peters, 1974).
84
Henry Cowell, Rhythmicana for piano (New York: Associated Music Publishers, Inc, 1975).

197
Example 4-16. Henry Cowell, 6 Ings, mvt. 1 “Floating,” bars 1-11.85

Example 4-17. Henry Cowell, 6 Ings, mvt. 6 “Seething,” 5-9.86

85
Henry Cowell, Six ings: for piano (New York: Associated Music Publishers, 1950).
86
Ibid.

198
 Like Beswick, Cowell begins his chapter on rhythm by noting the limitations of our

current system of rhythmic notation, recognizing that it is “incapable of representing any except

the most primary divisions of the whole note.”87 Furthermore, Cowell states his belief that if we

are to have “rhythmical progress” or “cope with some rhythms already in use,” then “new ways

of writing must be devised to indicate instantly the actual time-value of each note.”88 To this end

he suggests that composers begin to think of meter not in terms of quantities of quarter-notes, but

rather how the whole note is divided into equal parts.89 Cowell logically demonstrates how this is

reasonable, given the typical nature of the whole-note’s subdivision and its ability to be

subdivided into variously many parts through the applications of tuplets; he proposes that tuplets

be mostly done away with, given their limitations and clumsiness, and replaced with a modified

understanding of the whole-note as possibly divisible into any number of subunits:

However great a variety in time effect is made possible by [the] existing system, certain
limitations at once suggest themselves. We are always at liberty to divide a whole note
into two halves…. But if we wish to introduce into composition a whole measure of
normal length divided into three notes of equal length, there is no way of doing so except
by the clumsy expedient of writing the figure 3 over three successive half-notes filling a
measure. In other words, the notes as written down have a certain time-value impossible
under the circumstances, and the discrepancy is reconciled by explaining that in reality
notes of a different time-value are intended. Were the use of such notes of rare
occurrence, this method might be justifiable; since, however, these notes and others
having a similar discrepancy in time are very often used, should not an independent
method of notation be found for them?90

Cowell goes on to propose nomenclature for “these notes” of which he speaks:

At present a note occupying one-fourth the time of a whole note is called a quarter-note.
Consequently, instead of calling a note occupying a third the time of a whole note a
“half-note triplet,” why not refer to it as a third-note? Hereafter in this work a note will

87
Cowell, New Musical Resources, 56.
88
Ibid.
89
Ibid.. Of course, this is in fact how the system outlined in chapter 1 proposes we begin conceptualizing
time signature.
90
Ibid., 49-50.

199
 be designated by a fractional number indicating what portion of a whole note it
occupies.91

For the most part, Cowell’s system as found in New Musical Resources is the same as that given

in the score notes to Quartet Romantic and Quarter Euphometric. Cowell provides not only for

dyadic subdivisions of non-dyadic notes,92 but also allows, as before, for the notes to be

dyadically combined:

As is at present the practice in the series now in use, that of whole notes, the divisions to
obtain fractional notes in the proposed new series will be by two; and since the bases of
these new series are themselves fractional notes…[notes] obtained by multiplying any
basic note by two are included in the series to which the basic note belongs. Thus, a
third-note multiplied by two gives a two-thirds note, which is not only twice the length of
a third-note in value, but will form three equal notes in the time of two whole notes—six
third-notes.93

Furthermore, Cowell also recognizes, as Beswick before him, the potential to order triadically-

rational notes in any order imaginable among traditionally dyadic-rational notes when he writes

that in the “old notation three triplet notes or their equivalent must always be used together; in

the new notation perhaps only one triplet note will be used between quarter-notes,”94 though he

also broadens this concept beyond Beswick’s to include any note from either his new system or

the traditional one:

Still another possibility opened up by the new notation is that of separating notes of
triplet or other time-values by placing between them notes of other systems.95

and recognizes the ability to employ these non-dyadic notes in any combination within a

measure:

91
Ibid., 53-54.
92
Again, perhaps a flawed limitation of his system.
93
Ibid., 57.
94
Ibid., 59.
95
Ibid..

200
 In reality… there is nothing to prevent more than one time-scheme from being included
within a single measure; such a scheme does not have to fall within the limits of a whole
note; all that is necessary is to see that the fractional divisions of each time-scheme fill
the time allotted to that time-scheme.96

Importantly, Cowell does fix one problematic aspect of his earlier system from the rhythmic

quartets. Previously, Cowell prescribed note shapes for all odd rhythmic subdivisions up to 15.

Probably recognizing this arbitrary limiter and foreseeing the need to develop more note-shapes

from an already abundant set, Cowell provided a new modular system of rhythmic designation

still using the same note shapes from before (up to 9) and then providing a method of modifying

the basic note-shapes to generate subdivisions larger than 10 through the application of a slash

through the notehead. Though, it is significant to note that Cowell never developed a system of

notating non-dyadic rests, as he admits in his description of his note-shaped system below:

In our present notation the shape of notes is the same; their time-value, whether whole
notes, half-notes, quarter-notes, etc., is designated by printing the note as open or solid
and by adding stems and hooks. All that need be done, then, is to provide new shapes for
notes of a different time-value—triangular, diamond-shaped, etc. The use of open and
solid notes, of stems, and of hooks is equally applicable to these notes of varied shape. A
few adjustments in regard to designating rests would make the system complete. The
shapes of the new notes are necessarily arbitrary, except in the use of triangular notes for
the third-note series, since a shape suggestive of the number of a series would be too
complicated for practical purposes. Notes of new shape are provided for the new series
up to the ninth-note series. For the basic note of all new series above a tenth-note in
value, a stroke drawn diagonally downwards from left to right through the basic note of
an earlier series indicates that the number ten has been added to its original value. This
obviates the necessity of inventing new shapes for notes of higher series. 97

However, there is still some problematic inconsistency here, with multiple possible notations for

rhythmic values as a consequence of his modular system. For example, a 12th-note is technically

the second dyadic-subdivision of the 3rd-note (void-triangle) and is thus a triangular note-shape

with one flag. However, the 12th-note is the modular form of the half-note and should then

96
Ibid..
97
Ibid., 56-57.

201
alternatively be notated as a void ovaloid note with a slash. This is also a problem with the 18 th

note, which is either a modular 8th-note or the dyadic subdivision of the 9th-note. Many such

multiple methods of notating a value begin to arise, largely due to the system’s reliance on

dyadic subdivision of all values.

Regardless of Cowell’s overlooked inconsistencies and recognized incompleteness of his

system, his proposal is nonetheless seminal. For the first time since Johannes Tinctoris, a

musician has proposed a method to completely, or at least nearly completely, emancipate

rhythm. While Beswick provided a system to freely notate dyadic and triadic subdivisions of

beat, it was Cowell who proposed the complete free notation of any rhythmic subdivision and

combination, albeit forgetting many rhythmic possibilities himself. Countering centuries of

derision towards rhythmic diversity, stretching as far back as Morley, Zarlino, and Glarean,

Cowell proclaims the aesthetic legitimacy of complex polyrhythm, encouraging composers to

explore more fruitfully this nearly unexplored dimension of musical discourse when he writes:

Nor can it be assumed that a complicated relationship is unrhythmical; if a group of nine

notes is played against a group of eleven, the result may sound chaotic to a listener
unfamiliar with such a procedure; but to musicians accustomed to cross-rhythms, or to
certain peoples who use more complex rhythms, the combination is not only intelligible,
but possibly moving.98

While Cowell abandoned nearly all of his rhythmic experimentation following the 1930s

after his Rhythmicana for Rhythmicon & Orchestra (1931) and Rhythmicana for solo piano

(1938),99 New Musical Resources and the ideas contained therein continued to influence musical

98
Ibid., 46.
99
Some speculate that Cowell’s abandonment of much of his experimental aesthetic after the 1930s is tied
to his changed attitude after his arrest and multi-year incarceration. Despite his early parole in 1940 and pardon in
1942, his arrest for “moral crimes” of homosexuality and consequent incarceration and negative notoriety had a
devastating effect on Cowell. On first meeting Cowell in 1947, Conlon Nancarrow notes that Cowell "was a terrified
person, with a feeling that 'they're going to get him.’” According to musicologist, theorist, and composer Kyle Gann,
this experience had a profound and lasting effect on Cowell’s music. Soon after his release from prison, Cowell's
output turned relatively conservative, simpler rhythmically and harmonically, with his signature provocativeness

202
experimentation and diversification, overtly or covertly, throughout the remainder of the century.

For the remainder of this chapter, we will examine those strands of rhythmic experimentation

that spring from many of the ideas outlined by Cowell in 1930.

Though Cowell followed seriously few of the rhythmic ideas he proposed, many

composer’s forged entire creative careers from the germs of Cowell’s suggestions, some of

which are seemingly ancillary remarks to his core ideas of rhythmic progressions following

polyrhythms according to the harmonic series. The most obvious example of this phenomenon is

the work of Conlon Nancarrow, which is directly attributable to Cowell’s remark that while

“some of the rhythms developed through the present [rhythmo-acoustical] investigation could

not be played by any living performer, [they] could easily be cut on a player-piano roll.”100

Nancarrow’s use of tempo proportions to create combinatorial structures (sometimes truly

irrational) polyrhythms (discussed later in this chapter) was also inspired by New Music

Resources.101 To this end Cowell noted this further application of his system:

We have but to use different tempi simultaneously in different parts, choosing and
relating them properly according to the tempo scale…. It is obvious that when more than
one tempo is used simultaneously, the number of units of the quicker tempo will be
increased in proportion so that musical sections will end simultaneously.102

Furthermore, Cowell’s thoughts on tempo, as outlined above and in the excerpt below,

anticipated Elliott Carter’s metric modulations (discussed later in this chapter) and John Cage’s

temporal complexities in Music of Changes from 1951, as well as many of their European

largely abandoned. As Nancarrow observed, Cowell's imprisonment had the consequence of compelling Cowell to
"[keep] his mouth completely shut,” forsaking anything radical – politically, morally, or artistically. See Kyle
Gann, The Music of Conlon Nancarrow (Cambridge, New York, and Melbourne: Cambridge University Press,
1995), 44. See also Michael Hicks, Henry Cowell, Bohemian (Urbana: University of Illinois Press, 2002).
100
Cowell, New Musical Resources, 64-65.
101
Of course, this was itself inspired by Charles Ives use of differing simultaneous tempi.
102
Ibid., 92-93.

203
counterparts such as Brian Ferneyhough and Michael Finnissy, who would explore temporal

relationships through complex hierarchical tuplets (see Chapter 5). Cowell recognizes that

proportional use of tempi is not a novel concept but suggests that the “mathematical ratio

between tempi” is something not well systematized.103

Cowell even suggests new notions of time signature that would be explored modestly in

the mid-20th century by Carlos Chavez, Karen Khachaturian and more thoroughly by Thomas

Adès in the late-20th century (see Chapter 5). Extending his ideas about different fractional

subdivisions of the whole note and recognizing that his ideas of non-dyadic fractional

subdivision need not only apply to note values specifically, Cowell suggests that “new meters

could be made by using the new kinds of notes suggested in the time-scheme…. such new meters

[would be] invaluable in combining meter and time-ratios.”104

Cowell’s new note designs to indicate non-dyadic rhythmic values would also inspire

later attempts to rationally quantify rhythmic quanta. Italian composer Bruno Bartolozzi

proposed a system in 1961 that attempted to systematically delineate non-dyadic rhythmic

values, though rather than accomplish this through the modification of noteheads like Cowell or

flags like Beswick, Bartolozzi used combinations of stems and augmentation dots. In his 1961

article in the Journal of Music Theory, Bartolozzi stated the premise of his system:

To obtain those values not attainable with the present system of notation, one augments
the value necessary for completion of the note by adding, above or below the note, the
tail [flag] or dot corresponding to the deficient value.105

103
Ibid., 92.
104
Ibid., 66.
105
Bruno Bartolozzi, "Proposals for Changes in Musical Notation” (Journal of Music Theory. 5, no. 2: 297-
301, 1961), 300.

204
In essence, Bartolozzi’s system adds extra stems to traditional noteheads, with or without flags,

adding the value represented by that stem to the appended notehead; augmentation dots are

included when necessary to derive any intermediate value between the notes derivable by the

composite symbol of stems and notehead. Bartolozzi concisely explains his system below:

[T]o augment the value necessary for completion of a note, the five following symbols are
adopted [along with the augmentation] dot. The augmentation is applied above or below
the note: 106

Here, one can see that a half-note with an additional 8th-note stem takes the four 8th-notes of the

half-note and adds them to the singular 8th-note of the added stem, making a combined symbol of

five 8th-notes; this is also done with 8+1 16th-notes, 4+1 32nd-notes. For the whole-note,

Bartolozzi has given a “double-whole-note” (by the bars on either side) with an augmentation dot

(indicating an additional half of the double-whole-note), and a stem, which we must assume

represent a quarter value given the sum needing to total 13 (8+4+1 quarter notes).

The possible advantage over the later system when compared to Cowell’s is that it can

express non-dyadic rhythmic partitions with only the symbols present in common practice

(typical noteheads, stems, flags, and augmentation dots); however, there is obviously some visual

complication by the application of so many appendages to the notehead. For example, the

quantity of data given on each notehead could make it difficult to differentiate an 11/8th-note

from a 14/8th-note or a 5/4th-note from a 10/8th-note. The table of values given by Bartolozzi is

given below in Figure 4.6.

106
Ibid..

205
Figure 4-6. Bruno Bartolozzi’s complete table of rhythmic notation.107

Furthermore, what is perhaps obvious from this table above is that Bartolozzi’s system is

entirely dyadically oriented. While one can create many non-dyadic partitions previously

unavailable (i.e. 5, 9, 10, 11, 13, and 15 32nd-, 16th--, 8th-, and quarter-notes – 7 and 14 being

available via double-dotting in the traditional system), these are still all merely partitions of note

subdivisions that are intrinsically part of a completely dyadic rhythmic hierarchy. In this system

one still needs tuplets and is bound by the completion of those tuplets within a measure. Thus,

this system, even though proceeding Cowell’s by 30 years, is arguably less visionary than

Cowell’s, since it only provides easier ways to notate partitions, not subdivisions.

It is not necessarily surprising to realize that the rhythmic innovations of Beswick,

Cowell, and Bartolozzi represent nearly all systematic efforts to diversify prescriptive rhythmic

practice beyond the traditional dyadic system throughout the entire 20 th century. Musicologist

Gardner Read notes theorists and composers proclivity towards pitch over rhythmic innovation

when he writes that “Most inventors of unorthodox notational systems [throughout the 17th, 18th,

19th, and 20th century] retained common-practice methods in depicting rhythmic and durational

values, concentrating their reform efforts on pitch clarification. Only when the procedure that

individualized pitch made time values unclear did the reformers search for new approaches to

107
Ibid..

206
indication of duration.”108 Of the over 100 different rhythmic notation reform proposals outlined

in Gardner Read’s comprehensive survey of musical notation reforms in the 17th through 20th

centuries, only the above three offer a method of notating rhythms outside the typical dyadic-

rational system of subdivisions and partitions. Like Beswick, Cowell, and Bartolozzi, Read

addresses the glaring inadequacy of our present notational system’s ability to express diverse and

nuanced rhythmic values; he goes on to even question why nearly no composer or music theorist,

in almost half a millennium, seems at all concerned about this deficiency while many have been

historically concerned about the precise and diverse notation of pitch quality since the inception

of music notation in the West. It is worth quoting Read at length to these ends:

It is curious indeed that the purpose of most notehead reforms has been to distinguish
pitch levels rather than to define rhythmic values. Yet our present mensural notation is
notably, if not notoriously, inadequate in expressing complex and irregular rhythmic
schemes. Why were theorists and composers not motivated to eradicate this notational
ineptitude, long before they addressed the question of inherited pitch symbology? Given
the intense experimentation with new staves, clef signs, and pitch factors, it is hard to
believe that from 1657 – when Joannes Van der Elst suggested radical changes in
rhythmic note forms109 – until Henry Cowell’s 1917 exploration of new rhythmic
notation, few significant systems that dealt with this vital aspect of notation were
proposed…. The one duration innovation now embraced by musicians the world over is
non-mensural, which does not depend on note-head color, size, or form to make its
point.110
108
Gardner Read, Source Book of Proposed Music Notation Reforms (New York: Greenwood Press, 1987),
135. For a complete discussion of attempted modern note-shape designs see pages 135-281. Nearly every proposed
system either applies note-shape design only to pitch, many such systems ignoring the durational dimension entirely
or else applying only a system that encapsulated rhythm as typical.
109
Johannes Van der Elst, Notae Augustinianae, sive musices figurae seu notae novae concinendis modulis
faciliores tabulatoris organicis exhibendis aptiores (Augustinian note-symbols, whether old or new notations, made
easier for group singing and better adapted to clear instrumental tablatures) (Gandavi: Maximiliani Graet, 1657).
This “radical” change was not as radical as Read suggests, since it was totally predicated on dyadic hierarchies and
was far less radical than something like Tinctoris’, Gaffurius’, or Dygon’s system of proportional notation, which
allowed for note values unheard of in notated Western music at that time. The “innovations” of Van der Elst, in light
of what came a century prior to him, are trivial and not worth mentioning at length herein. In the context of Read’s
book, which does not discuss the Middle Ages or Renaissance, Van der Elst’s system from the 17th century is
interesting as one of the few surviving examples of rhythmic notation reform, but when contextualized by the
centuries prior to Van der Elst, they are insignificant.
110
Here, Read is referencing non-prescriptive rhythmic notations (freely interpretative rhythms based on
approximate spacing of musical events within a score). Since this dissertation is only concerned with prescriptive
rhythms, such innovations will not be discussed. However, it is interesting to note that the only rhythmic innovation

207
 Ultimately, Cowell’s system is that which is the most rhythmically diversifying and

influential from the 20th century. However, while Cowell’s notational system was far more

comprehensive than Beswick’s or Bartolozzi’s, it was still fundamentally limited and perhaps

problematically proliferate in its shapes. The proliferation of symbols half a millennium before,

while successfully diversifying rhythm, were overwhelmingly complex and often showed

contradictions and redundancies. Cowell’s system not only has the clear problem in its reliance

of dyadicism and its consequently missing note values, but also begets many redundancies with

many symbolic pointers pointing to the same definition, not to mention its complete lack of rest

notation. Furthermore, there is the question of playability, especially once many diverse shapes

begin to intermingle and not be bound to their complete rhythmic aggregate; how easy will it be

to read perhaps up to eight differing note heads along with the accompanying subdivisions and

combinations in one piece or even one measure, let alone perform such a complex potpourri of

micro-temporally flexible values? Perhaps these questions, limitations, inconsistencies, and

incompletenesses of Cowell’s system are some of the reasons of its relegation to the

metaphorical museum of musical oddities. Even Cowell himself never extensively employed his

rhythmic notation and structural theories. However, they do offer us some lessons on how to

approach (and perhaps not approach) the fashioning of such a system. Though we leave Cowell’s

notation in the 1930s, as we continue to press forward through the 20th century, the rhythmic

ideas proposed by Cowell would beget new ideas in rhythmic diversification that would point to

new concepts of metrical and temporal stratification as well as a renewed interest in non-dyadic

time signatures.

that has been widely adopted is one that frees the composer from prescribing rhythm precisely (of course such “free”
rhythm is not new to Western music, originating in chant notation from the Middle Ages, and existing in various
forms to the present, including, for example, the free-rhythmical preludes from Baroque French harpsichord suites.
See Read, Source Book, 272.

208
 Conlon Nancarrow: The Emancipation of Meter in the Mid-20th Century

Conlon Nancarrow (1912 - 1997) was born in Arkansas. Starting at the age of 21,

Nancarrow received his first formal composition training with Nicolas Slonimsky, Walter Piston,

and Roger Sessions in Boston between the years of 1933 and 1936. During this time, Nancarrow

began to develop his own unique desire for musical experimentation in the domain of time and

rhythm.111 The extreme difficulty posed by his music in the early 1930s resulted in only a few

satisfactory performances prior to leaving the United States to fight in the Spanish Civil War.

Neither performers’ frustration with his music nor Nancarrow’s interest in complex rhythms

waned by his return from the war in 1939. In that same year, with a need of adequate

performances of his music, Nancarrow discovered Henry Cowell’s treatise New Musical

Resources, which not only contained many rhythmic ideas reinforcing his own but also ideas he

had yet to explore.

Nancarrow realized the implications for Cowell’s new paradigm of rhythm were perhaps

limitless, but the same problem still confronted him: if his music were already too difficult to

perform, implementing such rhythmic structures would only make it more so. Conveniently,

Cowell recognized this difficulty and offered this offhand suggestion, which would form the

primary concern of Nancarrow’s life work:

Some of the rhythms … could not be played by any living performer; but these highly
engrossing rhythmical complexities could easily be cut on a player-piano roll. This would
give a real reason for writing music specially for player-piano, such as music written for
it at present does not seem to have, because almost any of it could be played instead by
two good pianists at the keyboard.112

111
Nancarrow, "Tempus perfectum," 266.
112
Cowell, New Musical Resources, 108.

209
 When in 1940 the United States government denied renewal of Nancarrow’s passport due

to his prior involvement in the Communist Party, he immigrated to Mexico and, finding the

musical climate there equally unfavorable towards his musical experimentations with rhythm,

undertook creating music for the player piano, superimposing tempi in wholly polyrhythmic

pieces. The piano rolls and the clockwork mechanism of the instrument ultimately gave

Nancarrow more temporal control over music than had ever previously been possible. Composed

primarily from the 1950s through 1980s, Nancarrow’s innovative 49 numbered Studies for

Player Piano are works of incredible rhythmic complexity whose hallmark is the temporal

conflict among simultaneous layers of music, which he called “Temporal Dissonance.” Within

these pieces, Nancarrow employed rhythmic and “rhythmo-tempic”113 proportions in various

ways, largely influenced by the ideas of Henry Cowell as set forth in New Musical Resources.

Perhaps most significant for Nancarrow among Cowell’s rhythmic suggestions (other

than using a player piano) was Cowell’s suggestion to use a system in which correspondences

are established between polyrhythm and proportional tempi, very similar to a prolation canon

from the Renaissance or Middle Ages.114 This experimental use of different simultaneous tempo

values, particularly those not related by powers of 2, was central to Nancarrow’s work.

Generally, there are three broad categories of Nancarrow’s use of ratios in his music: tempo

relationship, melodic/harmonic material, and structural design. In structure, Nancarrow used

ratios to determine temporal positionings and rates of imitations between voices as well as other

structural concerns on micro and macro levels. Melodic and harmonic materials are sometimes

derived from ratios employed elsewhere in the music; for example, sometimes central pitch sets

113
By this, we mean the explicit use of stratified proportional tempos to create a polyrhythmic complex,
rather than the use of tempo to determine the pace of music generally.
114
Recall from Ciconia’s le ray au soleyl with the three voices unfolding in a tempo proportion of [Link].

210
are reflective of the tempo proportions of the canonic voices in a work. Tempo is often also

determined by ratios, sometimes successively and sometimes simultaneously. The category of

tempo in simultaneous proportions is what concerns us here most, since it is the device used to

create highly complex and unique rhythmic structures.

In his very first player piano study, “Rhythm Study No. 1,” Nancarrow began exploring

relationships between different simultaneous tempos, particularly 120 and 210 which reduce to a

4:7 ratio (see Example 4-18). In this work, we can see a complex array of simultaneous, stratified

polyrhythmic proportions: 3:2 (1st and 2nd staves), itself subdivided into proportion 9:14; 5:4 (1st

and 5th staves), itself subdivided into proportion 15:8; 4:3 (2nd and 4th staves); 5:3 (2nd and 5th

staves); and 5:8 (4th and 5th staves). The combined continual effect in this work is utterly novel in

Western music since the far less complex stratified continuous polyrhythms found in

mensuration canons of the 14th, 15th, and early 16th centuries.

Nearly all of Nancarrow’s tempo canons have subtitles that list the ratios at work in the

canon’s tempo markings for individual voices; usually, the number of elements in the ratio

indicate the number of canonic voices115 in the study. Table 4-2 below describes most of

Nancarrow’s tempo canons and the tempo markings used in the scores. Nancarrow wrote

numerous “tempo canons,” borrowing directly from Cowell’s ideas on proportional tempo

relationships (what Cowell described as “a harmony of several different rhythms played

together”) to create complex polyrhythms. Fascinatingly and significantly, Nancarrow also

explored simultaneous tempo proportions involving irrational numbers, to create wholly

incommensurate polyrhythms.

115
It should be clarified that canonic “voice” in Nancarrow’s work is not necessarily a monophonic musical
line but is often a dense polyphonic complex.

211
Example 4-18. Conlon Nancarrow, excerpt from Study for Player Piano No. 1.

With his first study for player piano punched between 1949 and 1950, over the next

fifteen years between 1951 and 1965 Nancarrow wrote the remainder of his first thirty studies for

player piano, all of which explored ever complexifying rational relationships between musical

strata. A pivotal turn in his explorations, particularly in achieving irrational approximations

occurred after his 20th study, when Nancarrow made some custom improvements on his player

piano roll hole-punching mechanism, modifying it from a ratcheted device, only allowing fixed

divisions of metric units, to one with continuous movement, allowing for fluid and even more

precise metric transformations.116 The first piece composed under this new freedom was

Nancarrow’s “Canon X,” or Study No. 21, which explored continuous and gradual accelerandos

and decelerandos. By Study No. 31 Nancarrow had increased his ratio complexity to 21/24/25.

However, it would not be until his Study No. 33, composed sometime in his late 50s in the 1960s,

that Nancarrow would realize the full temporal potential of his new mechanized marvel.

116
Drott, "Conlon Nancarrow and the technological sublime," 542.

212
 It is easy to see in Table 4-2 the progressive rhythm-tempic complexity Nancarrow

reached in some of his later canons. Nancarrow’s stated purpose in writing these canons was

[to explore] my interest in temporally dissonant relationships. Temporal dissonance is as

hard to define as tonal dissonance. I certainly would not define a temporal relation of 1
to 2 as dissonant, but I would call a 2 to 3 relation mildly dissonant, and more and more
so up to the extreme of the irrational ones.117

The first and foremost implication of Nancarrow’s use of irrational ratios is that there will never

be a metric convergence in the music if precisely realized. Since, as proven in Chapter 1, there is

no rational number such that √2 can be expressed as some ratio of integers, if we have two

metric systems progressing under an irrational relationship, they will continuously move out of

phase.118 Such a metric complex maximizes what Nancarrow calls “temporal dissonance.”

Harkening back to his studies of Henry Cowell’s New Musical Resources in his youth,

Nancarrow uses this term to describe, in terms of harmony, the relationship between independent

contrapuntal lines progressing in different tempi.119 The more extreme the difference, the more

the relationship between the lines is dissonant. Nancarrow considered most of his early works

before Study No. 33 to be relatively consonant. While perhaps perceptually unusual for people

conditioned to a musical system under the hegemony of the dyadic rational, two voices at a ratio

of 4:5 are not dissonant, only representing a harmonic major third. However, voices in an

irrational relationship are perhaps as dissonant as they can be,120 since they will never have a

common denominator.121

Reynolds, Roger. “Conlon Nancarrow: Interviews in Mexico City and San Francisco,” American Music
117

2/2 (Summer 1984): 1-24.

118
This is assuming that they start at some offset not the inverse of their given tempo relationship and that
there will never be any alteration in their temporal relationship.
119
Nancarrow, "Tempus perfectum,” 272.
120
Ibid..

213
Table 4-2. Relationship of Canon Ratios and Tempo Markings in Nancarrow Studies
Class Study Tempo Markings (q = quarter, h = half, w =
whole)
No. 14, “Canon 4:5” q = 88, q = 110
No. 15, “Canon 3:4” h = 165, h = 220
No. 17, “Canon [Link]” h = 138, h = 172.5, h = 230
No. 18, “Canon 3:4” h = 168, h = 224
No. 19, “Canon [Link]” q = 144, q = 180, q = 240
No. 24, “Canon [Link]” q. = 149.33, q. = 160, q. = 170.66,
q. = 224, q. = 240, q. = 256
No. 31, “Canon [Link]” h = 105, h = 120, h = 125
No. 32, “Canon [Link]” h. = 85, h. = 102, h. = 119, h. = 136
Rational

9 10 11 h = 72, h = 80, h = 88; the application of [Link]

No. 34, “Canon : :
[Link] [Link] [Link]
ratio results in progressively faster tempos in each
voice, peaking at h = 264 in the fastest.
No. 36, “Canon [Link]” w = 85, w = 90, w = 95, w = 100
No. 37, “Canon 150 : 1605⁄7 : 1683⁄4: 180 : q = 150, q = 160.7, q = 168.75, q = 180, q = 187.5,
q = 200, q = 210, q = 225, q = 240, q = 250, q =
1871⁄2: 200 : 210 : 225 : 240 : 250 : 2621⁄2 : 262.5, q = 281.25

2811⁄4” 122

No. 43, “Canon 24:25” w = 120, w = 125

No. 48, “Canon 60:61” q = 120, q = 122
No. 33, “Canon √2: 2" q = 140√2, q = 280

No. 40, “Canon 𝑒⁄𝜋 Tempo marking not used – Nancarrow notates a
simple diagram which gives timings that establish
a relationship of roughly this irrational ratio.
1 No tempo markings; in C canons play
No. 41A, “Canon ⁄ √𝜋 simultaneously. Nancarrow established these
2
relationships using total playing times:
Irrational

√3
1
1 = 5;00”
3
𝜋
No. 41B, “Canon √𝜋
⁄3 13 2
√16 √3 = 7’40”
1
1 1 3 = 4’35”
3 √𝜋
No. 41C (A:B), “Canon√𝜋 ⁄ 2 :
√𝜋
⁄3 13 " 3 13
√3 √16 √16 = 6’10”

121
Correlating to harmonic principal of the “fundamental tone.”

It is worth noting that Nancarrow directly borrowed this series of proportions from Cowell’s New
122

Musical Resources, wherein these tempo proportions are given as representatives of the 12 pitches of the justly-
tuned chromatic scale

214
 Temporal common denominators form a particularly significant structural function in

much of Nancarrow’s oeuvre. Kyle Gann calls the resultant metric coincidences created by the

common denominators “convergence points.” 123 Under the paradigm of consonance and

“dissonance,” Nancarrow’s continually transforming metric relationships in his tempo and

prolation canons create progressive levels of dissonance with movement towards and arrival on

“convergence points” representing tonal return and cadence. As Margaret Thomas eloquently

explains, "the momentum of many of the studies may be best understood as a progression toward

the resolution of their temporal dissonance via a process of convergence to a simultaneity."124

Consequently, the closer the canonic voices are to their convergence point, the more

clearly one can hear their imitative relationship. Equally, as the voices approach their maximal

distance from the convergence point, their imitative relationship moves into maximal obscurity.

Thomas described this as the “progression of the dissonance”125 Such structural demarcations

become particularly significant under the irrationality of Nancarrow’s later studies, wherein there

might be only one such convergence point, if there is any at all. For Nancarrow to fabricate such

arrival points, he must rely on arch forms in many of his irrational canons, wherein he swaps the

temporal relationship at some point in the piece, so that, by a retrograde process, they return to

their outset tempo and temporally converge as they began.126

Thus, in his Study No. 33 Nancarrow does not simply initiate the two canonic tempi and

let them continue their de-phase ad infinitum. Rather, Nancarrow creates a series of five

123
Drott, "Conlon Nancarrow and the technological sublime," 540.
124
Thomas, Conlon Nancarrow's, 4.
125
Ibid., 137.
126
This should not be confused with melodic retrograde. While Nancarrow might swap the temporal
relationship between the voices so that they follow their phasing process backwards to their original metric
coincidence, the other musical parameters need not, and usually do not, change.

215
concatenated arch forms, wherein the voices are either initiated at some time interval such that

the canon ends at the point of convergence or directed to invert their temporal ratio at the

canon’s midpoint. Consequently, by not minimizing the number of possible convergence points

to one (or none) and constructing points of convergence through various transformational

methods, Nancarrow presents a greater level of comprehensibility than would otherwise be

possible. Interestingly however, Nancarrow evades every point of convergence in his Study No.

33, arriving at the point without any sounding convergence; instead of a convergence, he begins

a new canonic section. Thus, what was the point of convergence is now a new point of

transformation; as much as it might be an arrival, it is also a departure.

As both Gann and Thomas have demonstrated, due to the progressive transformation of

imitative structure between voices due to their differing rates of unfolding, the degree to which

we perceive this imitation changes throughout the piece. Take for example the third canon in

Study No. 33, a notated approximation of which appears below (see Example 4-19).

Example 4-19. Approximated rendition of third canon from Nancarrow’s Study No. 33 for Player
Piano. Transcription by Paul Usher, who approximates 2:√2 as 7:5.127

Callender, Clifton. "Performing the Irrational: Paul Usher’s arrangement of Nancarrow’s Study No. 33,
127

Canon 2:√2" from Conlon Nancarrow, Life and Music: Online Symposium September 27 - October 27, 2012.

216
 Here, the point of imitation begins almost simultaneously, but it expands as the piece

progresses. At the midpoint of the section (not shown above), due to Nancarrow’s inversion of

the tempo ratio, the point of imitation begins to contract, ultimately returning to the convergence

point at the end of the section. Here, Thomas’s “progression of the dissonance” is manifested as

the point of imitation distally migrates, increasing the difficultly in perceiving canonic

relationships between the voices. As we progress towards the maximal “temporal dissonance”

we move from relative canonic simplicity to complexity. However, there is no clear point of

division signaling the shift from this textural simplicity to complexity, but such a shift does take

place. This shift is significant because it compels a transformation in the listener's orientation

toward the work, metamorphosing from a perception of canon to free polyphony and ultimately

back again.

While many of Nancarrow’s pieces are pre-designed such that the various canonic voices

will at some point converge, all his “irrational” canons use ratios that, by their nature, can never

converge, only doing so under arbitrary transformations.128 Again, take for example the

beginning of the third canon from Nancarrow’s Study No. 33. The beginning of the section in

Nancarrow’s original notation is given below (Example 4-20). Here, we see that Nancarrow

does, in fact, not begin the voices precisely together. Since he does not begin the voice

coincidentally, there is no point, even under his arch forms, at which the voices will truly

coincide. The voices can either expand apart infinitely or contract to an infinitesimal imitative

distance. Since his player piano is naturally limited in the micro precisions that it can make, there

is a limit to which Nancarrow can approach such an infinitesimal difference. Consequently,

when we approach the convergence point at the end of this section, we can hear the collapsing

128
Such as the arch-form process previously outlined.

217
imitative interval, but in abstraction, if this process were never halted, we would never reach an

actual convergence point. Once the voices get “close enough,” or as close as they can get,

Nancarrow stops the process and begins a new section (see Example 4-21).

As Nancarrow reached his late 50s, he became more and more committed to his

conviction that “time is the last frontier of music.”129 By using his mechanical musician,

Nancarrow emancipated himself from the concerns of performability, allowing himself to exploit

the possibilities of musical tempo within polyphonic forms, creating a flexible, fluctuating

musical space and time.130 His compositions lacked most of the qualities one would associate

with “traditional” music: melodiousness, harmoniousness, regularity, and emotion.

Example 4-20. The opening of the third canon Example 4-21. The closing of the third
in Nancarrow’s Study No. 33.131 canon in Nancarrow’s Study No. 33.132

129
Nancarrow, quoted in Garland, Americas, 185. See also Thomas, "Nancarrow's ‘Temporal Dissonance’:
Issues of Tempo Proportions,” 137.
130
It is worth noting that due to his music’s asynchronous rhythm, his composition’s formulaic method, his
medium’s mechanized character, and his instrument’s immobility and scarcity, his music was inaccessible to most.
Nancarrow, for much of his career, could only play his music in his studio unless he laboriously moved his player
piano to a concert venue; he was far too unrenowned for most of his career for any venue to provide him a player
piano. Nancarrow was also reticent to play his music for anyone but himself, only demonstrating his experiments
when he felt one was truly interested. See Koonce, "Interview."
131
Conlon Nancarrow, Studies no. 2, 6, 7, 14, 20, 21, 24, 26 and 33 for player piano (Mainz: Schott, 1988).
132
Ibid.

218
Nancarrow was too interested in the sonic implications of his mathematical experiments to be

concerned with emotion. Kyle Gann, perhaps the greatest Nancarrow scholar and promoter of his

works, recalls Nancarrow’s opinion on music’s ability to express emotion, Gann writes:

… Nancarrow was one of a trio of composers, along with John Cage and Milton Babbitt,
who didn’t believe in music’s ability to express emotion. I remember Conlon saying that
to him music was just an interesting pattern of sounds with no emotive connotations.133

Alistair Riddell was correct when he observed that “there’s a feeling that time is being

manipulated in an entirely different manner” in Nancarrow’s music.134 Composer Roger

Reynolds once remarked that "it doesn't seem possible that art like [Conlon Nancarrow’s] could

exist."135 Under past paradigms, Reynolds was correct; Nancarrow’s music, in many ways, was

impossible before the 20th century, though mechanized technology did exist such that it could

have physically been done.136 What was not present, however, was the musical climate enabling

such questions to be asked and their answers sought. Composers at the beginning of the 20 th

century, like Nancarrow, could avail themselves of such a climate due to the proliferation of

Ives’ and Cowell’s experimentations by the middle of the 20th century and a consequential

renewed interest in rhythmic diversification.

Conlon Nancarrow’s obsession to realize Cowell’s rhythmic ideas from New Musical

Resources ultimately led him well beyond Cowell’s fundamental suggestions. Nancarrow

extended Cowell’s ideas well beyond simple just-intoned pitch proportions to rational ratios as

close as 60:61 and surreally complex incommensurate ratios of irrational numbers. Barring

133
Gann, “Outside the Feedback Loop,” 3.
134
Nancarrow, "Tempus perfectum,”269.
135
Greeson, Conlon Nancarrow: Virtuoso of the Player Piano (Univ of Arkansas Press, 2012).
136
The player piano was an outdated form of technology during Nancarrow’s life. In fact, it had been
patented in 1863 by Fourneaux, though technologies like the player piano predate this machine for centuries, such as
music boxes and automated carillons and pipe organs.

219
Nancarrow’s player piano studies, the use of rhythm-tempic ratios or large-scale polyrhythmic

schemes in 20th-century music was largely confined to simultaneous expressions of different

tempos of relatively simple ratios. For instance, Elliott Carter, another composer heavily

influenced by Cowell’s New Musical Resources, used techniques similar to Nancarrow’s in

establishing tempo relationships, but did so with human performance in mind; consequently, his

polyrhythms and rhythm-tempi concepts never become as indivisible or incommensurate as

Nancarrow’s, but they do present a slightly different path from Cowell’s suggestion towards

proportional rhythms and tempos. Concurrently with Nancarrow, Carter would develop a system

of metric modulation through the use of proportionally arranged tuplets.

Elliott Carter: Metric Modulation and the Extension of the Tuplet

When only 17 years old, Elliott Carter met Charles Ives, with whom Carter shared his

adolescent songs; on this encounter, Ives encouraged Carter to become a composer. During these

formative years, Carter, with a letter of recommendation from Ives, studied at Harvard with

Walter Piston. After graduation, Carter, on the council of Piston, went to France and studied with

Nadia Boulanger. The following years of compositions output would be relatively conservative

within Carter’s oeuvre.137 However, by 1945 Carter’s compositional motivations took a different

137
Many Carter scholars have noted that Carter’s initial works were strongly influenced by Boulanger as a
consequence of studying with her. When compared to his later works, these early pieces are very traditional,
consonant, and conservative, doing little experimentation with form, harmony, or rhythm, which would typify his
later oeuvre. David Schiff writes that "Boulanger's impact on Carter's composing [was] evident, perhaps negatively,
in the works he wrote in the decade after he left Paris, which show no traces of the ultra-modernist music he admired
in the 1920s;” see David Schiff, "Carter, Elliott," (Grove Music Online, Oxford Music Online, Oxford University
Press). In 1936 with his ballet Pocahontas, Carter appeared to attempt his first break away from these conservative
styles. During this time, Carter became influenced by Aaron Copland’s simplicity and ‘American quality,’ as
demonstrated in Carter’s works Holiday Overture (1944) and First Symphony (1942). Carter's Piano Sonata (1945 -
1946) was perhaps the pivotal work demonstrating Carter’s compositional transformation during the mid-20th
century; in this piece, Carter first employed the dramatic scale and invention that characterizes many of his later
works.

220
route, one which eventually led to his fascination with musical time, its ability to flex, partition,

and subdivide; his development would ultimately become known as ‘metric modulation’. Carter

discusses this period during his late 30s, during which he spent a year living and composing in

southern Arizona:

One of the aspects of this period, in my life, was the notion of the experience of time. Of
course, it's a very old idea, people have been working on time since Adam and Eve, I
expect. But the idea of treating time as a very vivid and strong thing is certainly very
important in Marcel Proust and James Joyce. Those were writers I read when I was too
young to read such nasty pieces, but in any case this has been something that is very
important to me. I have tried in my pieces to give the concept of the passage of time as a
dramatic idea, so that the pieces change as they go along in one way or another;
different kinds of rhythm conflict with each other and so on. This was a sense that I
wanted to give because after all, as we live our own lives, we are constantly involved in
all sorts of different aspects of time.138

Following this compositional shift, Carter’s works began to operate through tension

between independent and highly contrasting elements, either in concatenation or stratification,

and he began regularly experimenting with dynamically changing and opposing musical time

systems and senses of musical time. He became progressively dissatisfied with prior influences,

particularly the Neo-Classical styles of Piston and Boulanger, as well as Charles Ives, with

whom Carter’s relationship had strained since the mid 1930’s. Musicologist and scholar of

Carter’s music Jonathan Bernard notes that “Ives's innovations... for all the initial excitement

they provoked, eventually appeared to [Carter] deeply problematic.”139 Carter thought Ives’

138
Elliott Carter and Sue Knussen, "Elliott Carter in Interview" (Tempo, no. 197, 1996), 5
139
Jonathan W. Bernard, "The Evolution of Elliott Carter's Rhythmic Practice" (Perspectives of New Music
26, no. 2, 1988), 165.

221
music too confused and not optimally effective in its rhythmic manipulations.140 Central to this

assessment was Carter’s perspective given by his new influences during this time: Henry Cowell

and Conlon Nancarrow.

Carter met and befriended Nancarrow when Nancarrow had just returned to New York

from the Spanish Civil War before his life-long exile in Mexico. Sharing a unique interest in

complex stratified polyrhythms, the two continued to correspond and share ideas after

Nancarrow left the United States; Carter even published some of Nancarrow’s ideas about

rhythmic experimentation and a few of Nancarrow’s early works for player piano in 1951.

Carter’s new philosophy on music and time was affirmed by these new influences, because

interestingly, they were each intriguing but ultimately unsatisfactory to Carter, who was

searching for something rhythmically emancipating but inclined to human performance. In

Carter’s own opinion, “all three [composers’] approaches to rhythmic emancipation proved, in

one way or another, unsatisfactory.”141

In a number of articles published during this period, Carter discussed the development he

underwent. He described his preoccupation during the 1950s as being "a desire to find a new

flow of music thought and expression - a tendency to which the previous efforts [of Cowell and

Nancarrow] seemed be leading."142 Disregarding many of the post-World War II European

trends, considering idioms like serialism a "return to old-fashioned avantgardism" and not "really

140
Carter never interested himself in ‘whimsical’ or uncalculated polyrhythmic relationships between
musical strata like Charles Ives; indeed, Carter specifically rebutted relating musical strata "in any one of the
possible uncoordinated ways that have been used either by Ives or by others in recent years,” feeling that such
procedures produced "a form of entropy, a degrading of the possibilities of communication." Consequently, Carter
always inclined himself toward notating precisely as possible the temporal relationships between polyphonic and
polyrhythmic layers. See Carter, The Writings of Elliott Carter, 353 and 355.
141
Ibid., 165.
142
Elliott Carter, Else Stone, and Kurt Stone, The Writings of Elliott Carter: An American Composer Looks
at Modern Music (Bloomington: Indiana Univ. Press, 1977), 351

222
experimental or advanced" regarding musical time, he instead sought “a more significant

temporal thought."143

Carter’s interest in musical time manifested itself in two differing yet complementary

manners: first, "the speeding up and slowing down of themes"144 by means of tempo

manipulation to create an aggregate musical structure;145 second, "a special dimension of time,

that of 'multiple perspective,' in which various contrasting characters are presented

simultaneously"146 by means of stratified polyrhythmic layers. In Carter’s opinion, such "double

and sometimes manifold character simultaneities" led to richer, more expressive music than had

theretofore been achieved.147

Carter’s development into his mature aesthetic is well demonstrated in two works: Cello

Sonata (1948) and String Quartet (1950 -1951), the latter composed during his year in Arizona.

Conductor and music critic of the 20th century, Richard Franko Goldman, notes that:

[The cello] sonata may well be one of the influential works of the century, for in it Carter
used for the first time a principle that he had been developing for some time. This
principle has been described as "metrical modulation.”148

Goldman also comments on Carter’s first string quartet, remarking on its seminal place in mid-

20th century American music:

[String Quartet] is almost without doubt the most important and imposing
accomplishment of American music in the last decade. It involves a texture in which non-

143
Carter, The Writings of Elliott Carter, 351.
144
Ibid., 246.
145
such procedures are demonstrated in the 1950s by Carter’s String Quartet No. 1 (1951) and Variations
for Orchestra (1954).
146
Ibid., 355.
147
Ibid., 356.
148
Richard Franko Goldman, "The Music of Elliott Carter" (Musical Quarterly. 43: 151-170, 1957), 162.

223
 simultaneous changes of speed [namely, "metrical modulations"] in the four instruments
become the essence of the contrapuntal texture.149

Example 4-22 shows an example of Carter’s early use of metric modulation in String Quartet.

Slightly later in his Sonata for Flute, Oboe, Cello and Harpsichord in 1952, Carter is

avidly employing metric and tempo modulations against complex and alternating time-strata.

The third movement of this 1952 sonata presents a warped baroque-style, couched in the style of

Nancarrow, testing the limits of “baroqueness,” with the harpsichord superimposing a faster

tempo in cross-rhythm against the ensemble’s gentle forlana rhythm.150 Traversing the

movement, we find numerous and rapidly emerging metric modulations. David Schiff,

Example 4-22. Elliott Carter, String Quartet151 (1950 - 1951), mvt. 1, bars 36 - 43.

149
Ibid., 162
150
See Elliott Carter, Sonata for flute, oboe, cello, and harpsichord (New York: Associated Music
Publishers, 1997).

224
Carter scholar, describes the work as “a study in motion and color,” with the continuity of the

third movement “crosscut like a movie - at times it superimposes one dance on another.”152

In these works, and others from the 1950s, Carter explores the possibility of temporal

manipulation via metric modulation, develop the technique he himself called ‘temporal

modulation.’ Concerning his fascination with this technique, Carter writes:

The effect that I am interested in producing is... one of perceived large-scale rhythmic
tension, sometimes involving the anticipation of an impending final coincidence of all the
disparate rhythmic layers at some key moment... [O]ne of the things that I like about this
kind of effect [temporal modulation] at slow speed is that at first these points of rhythm
don’t seem to have any graspable relation to each other and appear perplexing or
perhaps chaotic... Then, as these beats begin to converge toward a unison, you begin to
become aware of their pattern and to grasp the emerging rhythmic convergences.
Conversely, the rhythm may at first appear clearly directional and structured and then
seem to disintegrate into a floating, apparent incoherence. This sense of progression into
extreme irregularity and back to a perceptible order appears in many of my works.153

Carter's ideas regarding musical time gradually developed throughout the 1950s,

begetting a significant oeuvre, for which he is now famed. His fascination with "multiple

perspectives" and virtuosic, idiomatic instrumental writing led to a style wherein each instrument

played differing music with several musical streams occurring simultaneously. This stratified

music with parts relating intrinsically yet loosely to one another is his music’s hallmark since the

1950s. Throughout the 1950s, Carter continued polishing his new compositional and rhythmic

151
Elliott Carter, String quartet no. 1 (New York, NY: Associated Music Publishers, 1994).
152
David Schiff, The Music of Elliott Carter (London: Eulenburg Books, 1983), 116-117.
153
Elliott Carter, "Music and the Time Screen" (Current Thoughts in Musicology, n.d.), 1.

225
aesthetics and techniques, ultimately producing his seminal works, Second String Quartet (1959)

and Double Concerto for Harpsichord and Piano (1961).154

In this Double Concerto, Carter employs two methods for notating tempo fluctuation:

metric modulation and very precisely notated accelerandos/ritardandos.155 Richard Goldman

defines metric/temporal156 modulation as “a means of going smoothly, but with complete

accuracy, from one absolute metronomic speed to another, by lengthening or shortening the basic

note unit.”157 Lorraine Vaillancourt, conductor, pianist, and interpreter of Carter’s music

describes metric modulations as:

...a change (modulation) from one time signature/tempo (meter) to another, wherein a
note value from the first is made equivalent to a note value in the second, like a pivot. The
term was invented to describe the practice of Elliott Carter, who prefers to call it
temporal modulation. For metric modulation to exist 3 things have to occur:

1. There has to be an exact relationship between two different tempi

2. A common pulse must exist between these two tempi
3. The name and function of the pulse changes.158

Such a technique was not new, by Carter’s own admission, recognizing its place in music of

the 14th, 15th, and 16th century as well as in some of Carter’s immediate predecessors. 159

However, this technique as employed to an extreme by Carter, was unprecedented, the scale

154
Worth noting is that it was also during this period at the outset of the 1960s that Carter composed Eight
Pieces for Four Timpani (1950 -1966), in which these techniques are present, and which has become to idiomatic
tool for teaching this technique in curricula on 20th century music.
155
the former method found throughout the work (solo cadenzas as well as Allegro scherzando and Presto
sections) while the latter is found chiefly in the central Adagio
156
The first of these terms is the one in popular use; however, the second is what Carter himself preferred
to call it.
157
Goldman, "The Music of Elliott Carter,” 161.
158
Elliott Carter and Lorraine Vaillancourt, Elliott Carter (Outremont: Disques Atma, 2001), 15.
159
For example, Igor Stravinsky used three tempos (72, 108 and 144 bpm) in proportion [Link] in
Symphonies d’instruments a vent from 1920. Charles Ives also used forms of metric modulation on numerous
occasions.

226
of tempos far broader, the ratios more complex, and the changes happening within, rather

than between, musical phrases. In an interview with Cole Gagne, Carter speaks at length on

the technique of metric/tempo modulation. It is worth excerpting this explanation and

historical narrative from Carter at length. Carter says:

I can't explain how I came up with it. It appeared in my mind at the time that I was
writing my Cello Sonata, which was in 1948. But it's obvious that this is not a new
thing with me [being present in many forms of earlier music]…. but it didn't occur to
me to write a piece that employed [this technique] thoroughly until my Cello Sonata.
In that work, I made an analogue to the key changes, a kind of modulation from one
speed to another, step by step. The whole Cello Sonata starts and ends with the same
speed, and it goes through a great variety of speeds that are all strung together one
right after the other.… There are superimpositions of one speed over another, and
one becomes more dominant than the previous one, and that leads into the next
section…. When I finally began to think more about [this technique], it seemed to me
to be a summation of all the different efforts that modern music had been making in
the matter of rhythm--the various changings, going from slow to fast or from fast to
slow, having irregular rhythms like Stravinsky, accents on sevens and fives and
threes, and alternations. One could make a complete connection between all the
different possibilities modern music had been developing. It seemed to me a
conclusion that everybody was trying to find, and I felt that I had found it.160

A clear example of metric modulation from Double Concerto can be seen in Example 4-

23, taken from mm. 109-114 of the harpsichord cadenza. In the space of just a few bars, the

tempo fluctuates thrice – q = 98 to q = 56 to q = 70 – in the ratio [Link]. The first tempo is

established by continual 8th-notes in the left hand, which is contrasted by the right hand, which in

measure 110 introduces a new septpartite-partitioned pulse expressed by "beats" equal to seven

16th-notes, which become the new quarter-note beat at the resultant new tempo, q = 58 in

measure 111. Measure 112 introduces quintuplet 16th-notes, which are then reinterpreted in

measure 113 as dyadic 16th-notes, consequently establishing the third tempo at q = 70.

160
Gagne, Soundpieces, 89.

227
Example 4-23. Elliott Carter, Double Concerto, bars 1099-114.161

In all the harpsichord and piano cadenzas of Double Concerto, Carter transforms his

music with numerous such metric modulations. In the Allegro scherzando of the concerto, there

are altogether six rhythmo-tempic fluctuations executed by this technique, while in the Presto

there are three before the first piano cadenza and two following the second. The second method

of time manipulation in Double Concerto is the precise notation of accelerations and

decelerations, found previously and slightly more abstractly in the punched player piano rolls of

Conlon Nancarrow. In the concerto’s Adagio, there are five such notated continuous

transformations.

Many of Carter’s later works, occupying the majority of his remarkably long life, use

the techniques of metric modulation and continuous tempo change within highly stratified,

complex atonal and rhythmic textures. It is worth noting a few of these later works, though

the complete list is expansive. One such example from the 1970s is his song cycle on poems

161
Elliott Carter, Double Concerto for Harpsichord and Piano with Two Chamber Orchestras (New York:
Associated Music Publishers, 1994).

228
by Elizabeth Bishop, A Mirror on which to Dwell from 1975.162 In Anaphora, Carter follows

Cowell’s suggestion of rhythmic designs suggestive of harmonic progression and sets the

rhythm of the song as a structurally conceived ‘harmony.’163 David Schiff notes that ‘a

rhythmic backbone is formed by a cross-pulse between piano and vibraphone (sometimes

passing to the other instruments) in a pulse ratio of 65:69, with the two instruments

coinciding “only once, at bar 23, the poem’s center.”164 The unified ensemble from Anaphora

splits into two ensembles for Argument, superimposing two arguments: the voice, supported

by alto flute, bass clarinet, cello, and double bass against a counter dialogue from the piano

and bongos. The arguments set against each other in stratification each have their own

polyrhythmic and metrical design, often undulating in precisely notated rubato. Schiff

demonstrates how the instrumental design of O Breath is based on a “three-part polyrhythm”

of every 43/6, 37/5, and 65/9 of a beat. According to Schiff, this pattern is not meant to be

recognized, “the complex cross-play of very slow pulses (MM 8.372, 8.108 and 8.308)

[giving] the impression of a regular pulse surrounded by slightly irregular swelling and

fading.” By contrast, the soprano’s rhythm is utterly erratic in this movement.165

From the early 1980s, Triple Duo (1982)166 is written for three demarcated duos,

differentiated in timbre, style, harmony, and rhythm.167 Throughout the work persists the

162
See Elliott Carter, A mirror on which to dwell: (six poems of Elizabeth Bishop): for soprano and
chamber orchestra (New York: Associated Music Publishers, 2001).
163
Schiff, The Music of Elliott Carter, 172.
164
Ibid..
165
Ibid., 178.
166
See Elliott Carter, Triple duo (New York: Hendon Music, 1990).
167
See Schiff, The Music of Elliott Carter, 120-121. Schiff colorfully describes this ensemble, writing that
“the woodwinds gurgle, shriek and coo like a pair of amorous birds, the strings scrape and pluck comically, and the
percussion and piano evoke the more angular variety of free jazz.”

229
 Basic tempo q = 100 bpm
Related tempi:
• 100 x 2/3 = 67 bpm
• 100 x 3/2 = 150 bpm
• 100 x 5/4 = 125 bpm
• 100 x 3/4 = 75 bpm
• 100 x 5/6 = 84 bpm

Figure 4-8. Elliott Carter, Triple Duo: tempo relationships.

three duos’ indivisible polyrhythmic stratification in proportion [Link] (winds, strings, piano-

percussion respectively). Schiff notes that though there are eight tempi in the piece, “they all

are closely related to the basic tempo of MM 100 to the quarter” (see Figure 4-8 above for

tempi).168 From the early 1990s, Carter’s work Quintet for Piano and Winds (1991)169 also

demonstrates a complex tripartite rhythmic framework, the ensemble again divided into three

polyrhythmic sub-groups: woodwinds, horn, and piano. The winds occupy a dyadic

hierarchy, playing rhythms largely based on 16th-notes; less-dyadically oriented, the horn

plays quintuplets while the piano sextuplets. Schiff notes that Carter devised a “five-way

polyrhythm” based on the metronomic speeds of MM-7 for the oboe, 6.75 for the clarinet,

7.56 for bassoon, 8.859 for the piano, and 7.5 for the horn,170 or essentially

[Link] continuous macro-scale polyrhythm.

While Carter’s full realization of his rhythmo-tempic techniques is highly elaborate, his

notation of them meticulous and exacting, it is the overwhelming intricacy of his notation that

paradoxically operates as an implosive force, the details of the writing encumbering its

168
Ibid., 122.
169
See Elliott Carter, Quintet for piano and winds (Boosey & Hawkes, 1991).
170
Schiff, The Music of Elliott Carter, 125.

230
interpretation and straining the performer’s ability to coordinate accurately with the ensemble.

Consequently, while Carter’s oeuvre, including Double Concerto, illuminate new routes for

rhythmic exploration, such complexity simultaneously begs serious questions and concerns over

the notation, realization, and comprehension of such highly complex musical structures. Robert

Strizich notes that, “the requirements of [Carter’s] notation are so explicit as to seem overly

fussy and even unreasonable;” for example, in the Introduction (whose tempo is q = 105), Carter

“several times requires the players to subdivide a quarter note into exactly ten parts, rather than

prescribing an unmeasured roll or tremolo - which would, of course, be the aural effect of his

overly-explicit notation anyway.”171 Gardner Read, in his 1978 book on modern music notation,

presages Strizich’s remarks in his somewhat cautionary critique of Carter’s music:

The Double Concerto, the Piano Concerto, the Concerto for Orchestra, and all three
string quartets bring to mind Schoenberg's remark - applied to his own music - that such
music must be realized with "inexorable severity." The players of Carter's works cannot
"interpret" the music; they can only follow as best they can his minutely conceived and
rapidly fluctuating indicia. The scores, as they stand, vividly illustrate the pitfall into
which the over-zealous notator can tumble: their hyper-exactitude becomes more
inhibiting than the most calculated obscurity.172

Carter himself must have noted many of the notation’s pitfalls in his Double Concerto, since he

edited it various times after hearing it played on numerous occasions. Yet, even with his attempt

to make this score foolproof, errors still abound in its pages, enough so that Robert Strizich could

write a whole dissertation on the topic.173 One such error discussed by Strizich is Carter’s

notation of “non-rational”174 subdivisions of the beat, particularly Carter’s notation of quintuplet

and septuplet subdivisions.

171
Strizich, "Notation," 62.
172
Gardner Read, Modern Rhythmic Notation (1978), 48-49.
173
See Strizich, "Notation.”

231
 In his Double Concerto, Carter has subdivided quarter-notes in simple meters into five

16th-notes when writing quintuplets, which is correct by standard practices; however, when

subdividing a half-note into a septuplet, he has also chosen to use 16th-notes, which is counter to

standard practice of using 8th-notes. Furthermore, Carter subdivides a dotted quarter-note in

compound meters into five quintuplet dotted 16th-notes, counter to the typical practice of five

undotted 8th-notes. Carter explains that this unorthodox system of tuplet notation is based on

Renee Longy-Miquelle’s text from 1925: The Principles of Musical Theory.175 In a letter to the

editor of the Journal of Music Theory, Carter writes:

The question of the notation of artificial divisions of dotted units has even less uniform
agreement. It seems to me that Mme. Miquelle's method, cumbersome as it looks, has the
advantage (in music that is in alternations of 6/8 and 3/4) of making the different groups
of artificial divisions look different at once.176

In essence, Carter defends his perhaps unorthodox dotting of non-dyadic subdivisions in

compound meters by noting that when such notes are dotted, they are more distinguishable from

non-dyadic subdivisions in simple meters. Gardner Read again takes a critical stance against

Carter’s rhythmic complexity and the notation thereof in his article “Some Problems of

Rhythmic Notation,” in the Journal of Music Theory. Read writes concerning Carter’s notation

that:

it is not possible, however, to go along with Carter in advocating the method of notation
for artificial divisions expounded in an outmoded text by Renee Longy-Miquelle. This
system uses identical note-values when writing quintuplets and sextuplets against duplets
or quadruplets. But the next smaller note-value is employed for septuplets. The author's
reasoning - to me, completely fallacious- is that seven is closer to eight than to four (the
first division of a quadruplet); therefore, one should use the note-value that would be
employed for eighth-notes as the unit. Far more accurate in the mathematical sense, and
174
Again, a poorly chosen appellation since all of these rhythms are purely rational. We will only use
Strizich’s terminology as such this one time.
175
See Renée Longy-Miquelle, Principles of Musical Theory (Boston: E.C. Schirmer, 1925 [reprint,
1949]).
176
Elliott Carter, "Letters to the Editor,” in Journal of Music Theory, Vol. 9, No. 1 (1965), 272.

232
 more generally applied by composers today, is the principle advocated by Paul
Hindemith. He would retain the same note-value for both the regular and the
extraordinary groups until the notes in the irregular group exceed twice the value of the
normal group. This is only another way of saying that the note-value in the extraordinary
group does not change until the number of notes contained therein enters the second
division of the basic unit.177

Thus, Read would advocate the use of 8th-notes (rather than 16th-notes) for writing a septuple

against a half-note, since seven does not yet exceed twice the normal number of eighths for the

unit.

Of course, Read is not against the use of tuplets generally, but rather how Carter has

notated the values within the tuplets; he questions Carter's practice of dotting values within a

tuplet when those values are not actually following the principle of the augmentation dot, but

rather stand to show that the values themselves are a subdivision of a dotted note (though not

actually themselves augmented using a dot). Read, understandably, thinks this practice is utterly

counterintuitive, writing that:

The patent illogicality of dotting each note (or rest symbol) within a quintuplet or
septuplet, in either simple or compound time, is not a matter of personal taste and
discretion but of simple mathematics. Whereas duplets and quadruplets can, and even
should, consist of dotted notes in compound time, all other irrational figures cannot. If
the augmentation dot still means what it always has since its invention in the seventeenth
century - the increase by one-half of the note value it follows - it cannot be applied to
rhythmic figures whose total value would then exceed, or fall short of, the normal number
of units within the prevailing pulse or measure.178

Of course, what Carter is perhaps failing to recognize is that one can simply place above the

tuplet bracket the value from which the tuplet is subdivided, and thus not need to dot the tuplet

notes to indicate that the tuplet is derivative of a dotted note. Considering purely visual

simplicity and communicative efficacy, a comparison between Carter's notation of non-dyadic

177
Gardner Read, “Some Problems of Rhythmic Notation,” in Journal of Music Theory, Vol. 9, No. 1
(1965), 259.
178
Gardner Read, Modern Rhythmic Notation, 46-47.

233
subdivisions, Read's suggestion via Hindemith, and the further addition of note ratios above the

tuplet demonstrates how muddled Carter’s notations is (see Table 4-3 below). As noted by

Strizich in his assessment of notation in Carter’s Double Concerto, in Carter’s “zeal” for precise

Table 4-3. Comparison of Longy-Miquelle (Carter) and Hindemith (Read) tuplet notations.

Notation School
/ Meter-Beat
(duple) (compound)

Longy-Miquelle
(Elliott Carter)

Paul Hindemith
(Gardner Read)

Modern with
Ratio

234
notation “the composer has perhaps become overly complicated, rather than clear and

unequivocal.179

Carter wrestled with the issues of non-dyadic notations throughout his career, never

clearly deciding on a specific practice of tuplet notation in compound meters. However, even

long after the mild controversy of notation in the Double Concerto, Carter writes that he did not

think that such non-dyadic subdivisions could have been notated more clearly; to such ends he

wrote:

In [the Double Concerto] there are many problems of notation - the awkward dotted
notations, used for quintuplets and septuplets in compound time, that have bothered
performers in Warsaw and elsewhere. In defense of them, I do not think many passages
which use dotted and undotted notes simultaneously could have been notated as
clearly.180

Such issues regarding the proper and exact notation of tuplets will become of central interest

over the remainder of the 20th century for composers exploring greater levels of rhythmic

diversification. For Carter’s work, the issue concerns the problem of notating non-dyadic

subdivisions in either simple or compound meters; however, this particular issue would

compound its effects in the hyper-rhythmic music of some composers in the second half of the

20th century.

Conclusions to Chapter 4

Through the impetus of Ives and Cowell, by the middle of the 20 th century, the return of

rhythmic diversity and an interest in investigating the limits of non-dyadic rhythmic complexity

had returned to Western musical discourse. While “an argument against the development of more

diversified rhythms might be their difficulty of performance,” writes Cowell in his New Musical

179
Strizich, "Notation," 70.
180
Carter, The Writings of Elliott Carter, 296.

235
Resources, composers nevertheless continued to push the limits of the possible under acoustic

performance, perhaps recognizing as Cowell did that even if “a few minutes a day are seriously

devoted to mastering [polyrhythms], surprising results [can be] obtained,” believing that such

complex, indivisible rhythmic structures are “ as well worth learning as the scales, which

students sometimes practice hours a day for years.”181 What Ives, Cowell, Nancarrow, and Carter

had begun, along with various other early 20th century composers using extreme non-dyadic

rhythmic structures, is the longest uninterrupted proliferation in rhythmic diversity in Western

music history, lasting, with continuous and evolving interest, from the outset of the 20th century

to the present day.

The late-20th century and the exceptional rhythmic practices therein witness the

development of some ideas outlined above while some, like Cowell’s rhythmic notation, died

nearly as soon as they were conceived. Ultimately, the 20th century would progress with the

traditional rhythmic notation system intact, almost unchanged from its 18th and 19th century

form; however, the changes we will see continue are results of Ives’, Nancarrow’s, and Carter’s

work in the extreme use of tuplets and complete partitions of non-dyadic subdivisions. Rather

than attempt to fashion new forms of musical communication, composers like Milton Babbitt (b.

1916), Brian Ferneyhough (b. 1943), and Michael Finnissy (b. 1946) would seek the limits of the

notation system they inherited, employing tuplets to extents unrivaled even by Ives or Carter.

While Nancarrow’s later works pointed to music unperformable by humans due to highly

indivisible and even incommensurate polyrhythms, composers of subsequent generations would

follow this aesthetic with moderation, writing music for human performance employing

continuous polyrhythmic stratification.

181
Cowell, New Musical Resources, 64-65.

236
 By the end of the century, such continued developments in rhythmic diversification

would ultimately turn composers of the 20th century on the same course as composers of the 15th

century: to reassess the meaning and potential of time signature, relocating rhythmic demarcation

from the note-shape to the “mensuration” in which the note operates. Though Cowell’s suggested

innovations had failed, like the note-shape reforms of the ars subtilior of the 14th century, later

composers would see the efficacy of diversifying rhythm not through a confusing proliferation of

semiotics, but rather through the fluid reassigning of the semiotics given under global operators.

Meaningful questions to ask are why notational modifications offered by Cowell, or even

the simpler and more widely applicable ones of Beswick and Bartolozzi, did not find wider

appeal or persist. Why is it that note-shape reforms tend to not endure, though reforms in global

rhythmic operators seem to carry forward, even if simplified from more complex predecessors

(i.e. mensuration signs to proportions to time signatures)? What is the optimal medial point that

compromises between the need for freely moveable non-dyadic note values and performability or

readability? Is the adoption of new note-shapes even reasonable to attempt – such adoptions have

been successful in the past (e.g. the Franconian revolution of the late 13th and early 14th century)?

What made the adoption of Franco’s note-shapes persistent compared to the note-shapes of the

ars subtilior or Henry Cowell?

The above are questions perhaps answerable with educated hypotheses based on our

survey thus far. Franco’s system offered a simplification of the notation system prior to it, rather

than a complexification of it, while the note-shape systems of the ars subtilior and New Musical

Resources complicated their contemporary systems. Though such additional complexification

might be advantageous to the expression of rhythms previously unable to be notated, they

237
perhaps demand too much from the processor (here, human) of the notation, let alone any issues

of incompleteness or inconsistency in those new layers of complexity.

Of course, the systems of Beswick and Bartolozzi have their own issues that perhaps

suggest why they were not adopted. Beswick’s was, first, a system made by a woman, which, in

its time, was unfortunately not a strength in garnering popularity, regardless of its quality.

Furthermore, Beswick herself might have been myopic in patenting her system for exclusive

ownership of not only the system but seemingly anything written using her system; either she

wished no one to use her system, or she did not understand the consequences of protecting her

ownership of the idea to its possible flourishing. Bartolozzi’s problem perhaps lies not in its

complexification of a system, for his rules are simple and employ all present semiotics in a

logical and systematic fashion; however, all he provides is a method of notating partitions and

combinations of dyadic notes, a task already accomplishable through the use of dots and ties.

While his system provides a cleaner method of notating such partitions, it provides nothing

intrinsically new to the discourse of rhythm and is thus perhaps unnecessary. Thus, Bartolozzi’s

system only adds suspect clarity to the older system (i.e. it is not necessarily true that forgoing

ties will necessarily make music notation clearer) and definitively adds nothing new that could

not be notated.

Thus, it seems necessary for a new notational system, or an addition to a living system, to

be either 1) an unquestionable improvement over the communicative clarity of the older or

unmodified system (such as Franco’s system’s clarity over modal rhythmic notation) or 2) add

something intrinsically new to the system (such as Franco’s system’s ability to prescribe various

levels of hierarchical rhythmic divisions). The first of these is more of a necessity than the latter;

it seems that the addition of something new does not likely persist if there is substantial

238
complexity or confusion added. However, some additional complexity does seem acceptable,

since various levels of additional notational complexity have been successful throughout history:

Franco’s addition of new notes shapes and hierarchical divisions of basic rhythms, Tinctoris’

addition of what would essentially evolve into the tuplet, and the ability to notate non-dyadic

partitions of the whole note using dyadic time signatures (e.g. 7/4, 5/8, 11/4, etc.).

Ultimately, there seems to be some acceptable limit to modification in music notation,

one that relies on clarity, continuity, and useful innovation. Such is perhaps the reason why

Cowell’s useful innovations did not persist; they lacked clarity though they had continuity with

the past system and possessed genuinely new tools. However, this also shows why Ives’,

Nancarrow’s, and Carter’s ideas persisted, though they were not modifications to contemporary

notational practices; they were reasonably clear, since they entirely relied on the notational

system as given (consequently continuous from it), and they provided useful new avenues for

exploration within that system: the extreme use of tuplets, the extreme and continuous

stratification of indivisible rhythmic proportions, and a renewed perspective on time signature

and tempo and their relation to rhythmic design (namely, a mensural approach to time signature

and a poly-metrical approach to tempo). Even though these “innovations” offered nothing not

already inherent in the system, they pointed to unexplored extremes possible in the system. Thus,

though they were complex, they were able to be notated clearly, the notation of them was

perfectly derivative from the past system, and they provided novel musical structures and

sounds.

However, in terms of human performance, the music of Ives and Carter was constrained

by all the limitations of our present system of notation as outlined above. While Nancarrow

could essentially write whatever he wanted on his player piano rolls, not confined by any

239
limitations of classical music notation in the punched holes of his rolls, Nancarrow produced

nothing that, within one musical voice, intermingled incomplete partitions of different non-

dyadic subdivisions of the whole-note. Yes, Nancarrow provides intermingling of differing non-

dyadic hierarchies through the juxtaposition of two or more voices in differing hierarchies;

however, he failed to musically realize the ability to discretely move, within a single musical

stratum, between differing hierarchies emancipated from the tuplet’s necessity for completion.

Strangely, even Henry Cowell failed to accomplish this emancipation from the tuplet’s

aggregation, though his system perfectly allowed for it.

The late 20th century would continue to confront these same limitations, even in the

music of hyper-rhythmicity. Not until the end of the 20th century would we see a route to

emancipate music from dyadic hegemony and free composers from the tuplet’s aggregate in non-

dyadic time signatures. However, even these new musical operators would possess their own

problems and incompletenesses.

Synthesis III

From the historical observations made in Chapter 4, we can distill some principals upon

which we can expand our theory of notation systems. Before adding to our set of definitions, we

will add an additional axiom, which seems pertinent after the observations of Chapter 4.

Following this axiom, we will supply further features of notational systems along with the

observations and conclusions coming from those definitions.

AXIOM S3.01: If a system is systematic, then:

• There is no purely efficacious or inefficacious system.
• There is no purely optimized or specialized quantization.
• There is no purely coherent or incoherent quantization.

240
Figure 4-10: Diagram of System Quantization Classification

241
Definitions

DEFINITION S3.01A+B: A system is considered efficacious the more unique quantizations

a system can provide. A system is considered inefficacious the fewer quantizations it
allows.

DEFINITION S3.02A+B: Any quantization (or sub-quantization) is optimized if it contains

the fewest possible elements and operators to accomplish the intended interpretation and
translation of that quantization. A quantization that clearly lacks optimization is
considered specialized. There can be degrees of specialization.

DEFINITION S3.03A+B: A quantization is considered coherent if it allows few to one

unique interpretation. A quantization is considered incoherent the more unique
interpretations it allows.

The logical outlay of the above definitions is given in Figure 4-10 above; note that both

systematically efficacious and systematically inefficacious branches are neither better nor more

desirable than the other, but are rather two alternative approaches, each of which might be better

in particular circumstances.

Observations

OBSERVATION S3.01: Efficacious and Inefficacious systems allow for both optimized and
specialized quantizations.

OBSERVATION S3.02: There might be no perfectly optimized form of any particular

quantization, only more or less optimized quantizations.

OBSERVATION S3.03: An incoherent quantization is not necessarily a result of an

unsystematic system, since the “interpretation” of some element, quanta, or quantization
is not the same as the definitions of elements or operators and relies on the processor
rather than the system itself.

OBSERVATION S3.04: In regard to incoherence of a quantization, we should keep in mind

that interpretation is the process of synthesizing the definitions of elements, effected by
operators, in quantization. The interpretation is based on definitions of elements and
operators, however is not the definitions of those elements and operators necessarily but
rather the combined effect of each on the other. The more elements and operators
simultaneously under interpretation in any one moment of quantization, the more variable
the interpretations might become (incoherent).

242
 OBSERVATION S3.05A: The more optimized a quantization is, the easier it is to interpret
and translate, though it is not necessarily coherent.

OBSERVATION S3.05B: The more specialized a quantization is, the more difficult it is to
interpret and translate, though it is not necessarily incoherent.

OBSERVATION S3.05C: Difficulty in interpretation does not mean that there are numerous
interpretations. A specialized system is not necessarily incoherent.

OBSERVATION S3.06: Depending on the complexity and necessary precision of the

interpretation and translation of a quantization, a system might benefit from being more
or less efficacious, accounting for the point furthest from incoherence in a particular
quantization.

Conclusions

CONCLUSION S3.01: An optimized quantization is easy to interpret and translate. A

specialized quantization is difficult to interpret and translate.

CONCLUSION S3.02: Depending on the complexity and consequent nuanced prescription

of the desired interpretation and translation of a system’s quantization, over-optimization
of the quantization might be undesirable, leading to incoherence. Depending on the
simplicity and consequent basic prescription of the desired interpretation and translation
of a system’s quantization, over-specialization might be undesirable, leading to
incoherence.

CONCLUSION S3.03A: There is likely a necessary balance between system efficacy and
inefficacy for a quantization to be coherent.

CONCLUSION S3.03B: There is likely a necessary balance between optimization and

specialization for a quantization to be coherent.

The definitions, observations, and conclusions of Synthesis I, II, and III will all be

applied in the meta-synthesis at the conclusion of Chapter 5. Therein, we will examine some

theories of what makes a good system of musical notation and compare our abstract conclusions

to the practical assumptions of notational theorists from the 20th century. In so doing, we will

demonstrate the utility of our abstract theory and show its practical consequences. Through this

meta-synthesis, we will be better equipped to methodologically approach the refashioning of our

present system of musical notation to the ends of pan-rational rhythmic prescription.

243
 CHAPTER 5
NEW HORIZONS FOR RHYTHMIC DIVERSITY:
RHYTHMIC EXCEPTIONALISM AND ITS NOTATION IN THE LATE-20th CENTURY

The Limits of Tuplet Notation and Hyper-rhythmics1 in the Late-20th Century

Princeton, New Jersey, between 1962 and 1965, witnessed an evolution in the concept of

time in music, the coalescence of a new discourse in time’s composition, and the emergence of a

controversy that would slowly work to dismantle the preeminence of serialism in American

music for the next three decades. In 1962, composer Milton Babbitt outlined his theory of “time-

point” composition and “durational sets,” first used in his piece Composition for 12 Instruments

from 1954, in an article published in the soon-to-be American academic epicenter of printed

serial music discourse, Perspectives in New Music. In creating a system by which to serialize

musical events, or “time-points,” Babbitt was refining Olivier Messiaen’s older concept of a

“chromatic scale of time values,” as demonstrated in his work, Mode de valeurs et d'intensités,

from 1949.2 While Messiaen’s original idea created a separate serialization system for time

1
Herein, we are avoiding the use of concepts like “New Complexity,” due to the controversial baggage
such classifications carry. New Complexity has been defined as a “group,” “movement,” “branch of new music,”
“journalistic slogan,” “supermarket labeling,” “school of thought,” “radical aesthetic,” “broad aesthetic” and a
“resistance to musical post-modernism,” but primarily the term has been used to refer to “the density of black notes
per page;” see Stuart Paul Duncan, The Concept of New Complexity: Notation, Interpretation and Analysis (D.M.A.
Thesis, Cornell University, May 2010), 4. New Complexity can be a useful generalizing term, but has met with
some scrutiny not only from musicologists and theorists, but also the composers lumped into the aesthetic. Thus,
since New Complexity as a term is challenged and encompasses “complexities” both within and outside of the realm
of rhythm, we will not use “New Complexity” to identify any composer, aesthetic, or school. Rather, we will be
discussing composers and aesthetics using extreme rhythmic complexity, which I will term “hyper-rhythmicity.”
Composers practicing hyper-rhythmicity include composers typically identified with New Complexity and not. Even
retrospectively, composers of the ars subtilior, Renaissance, and early 20th century (e.g. Ives, Cowell, Nancarrow,
and Cater later) can also be considered “hyper-rhythmicists” relative to the general practice of their time.
2
We have not discussed this rhythmic innovation since it is intrinsically tied to dyadic rational rhythmic
hierarchies and thus is not “exceptional” rhythmic practice by the mid-20th century; however, one might consider it
“exceptional” rhythmic organization, a development of such experiments in rhythmic organization perhaps
derivative of mixed and asymmetrical meters of the late 19th and early 20th century.

244
values, Babbitt required mapping durational values with the chromatic scale of pitches, not

holding them as independent musical structures.

Thus, the rhythmic aesthetic in which Babbitt operates fluidly moves between one

wherein musical events are organized, relatively traditionally, along a dyadically-demarcated

timeline, often employing dyadically-hierarchical rhythms, and one wherein events themselves,

once their outset is determined, often evolve complexly down non-dyadic rhythmic hierarchies.

Thus, herein we are not so much concerned with Babbitt’s use of serialized time-points along a

dyadically demarcated unidimensional event-matrix; rather, we are concerned with what happens

within those events themselves, particularly when Babbitt employs his most intricate use of

tuplets. Arguably, Milton Babbitt, particularly in his rhythmically rich works, stands as one

composer influenced by Ives and Carter and presages the late 20th century aesthetic of hyper-

rhythmicity, an aesthetic intrinsically oriented towards dense rhythmic interplay, development,

and diversity primarily through the use of tuplets and embedded tuplets. To the end of

understanding the troublesome place of tuplets during the latter half of the 20th century,

following the impetus of Ives, Cowell, and Carter before Babbitt, we will first examine two

works by Babbitt from the 1960s.

In his seminal article, Babbitt clearly begins to define time similarly to the traditional 12-

tone equal temperament understanding of pitch:

Since duration is a measure of distance between pitch points, we begin by interpreting

interval as duration. Then, pitch number is interpretable as the point of initiation of
temporal event, that is, as a time-point number.3

Babbitt’s ultimate proposal for his new system of pitch-time serial synchronicity can be

concisely encapsulated by mapping the twelve equal-tempered pitches of the octave onto twelve

3
Babbitt, “Twelve-Tone Rhythmic Structure,” 63.

245
equally spaced time-points of a measure of music; naturally, the conventional musical measure

best suited to such a system is either 3/4 or its compounded equivalent, 6/8, both of which

contain twelve 16th-notes. Each of these 16th-notes in a 3/4 measure could then correspond to the

same mod-12 system used for decades to serialize pitches, allowing a composer to express of

row of twelve distinct integers (usually 0 to 11 or 1 to 12) both by a sequence of pitches and a

conceptually related series of time-point events within a 3/4 measure.4

Since Babbitt’s music regularly incorporates complex rhythmic events like non-dyadic

and non-triadic rational tuplets and embedded tuplets, both frequently carried over these bar lines

and beginning on complicated subdivisions of a seemingly arbitrary “beat,” adding the additional

layer of purely theoretical barring for the performer(s)5 to navigate might be non-optimal. The

4
As one might already foresee, this new method of time organization met with skeptical scrutinizers, even
within the serialist fold of Princeton itself. Just as one might question the arbitrariness of the supremacy of a mod-12
system of octave-based, equal-tempered pitches given the theoretically infinite space of frequency modulation, one
might equally, if not more stringently, question the arbitrariness of Babbitt’s system, especially since this arguably
arbitrary system is itself based on an arbitrary system. Musicologist and music theorist, Richard Taruskin, succinctly
addresses the concern of some of Babbitt’s peers, and Taruskin’s remarks are worth excerpting at length: “…
analysts were quick to note… that an irreducible arbitrary (or ‘notational’) element remained at the heart of
Babbitt’s procedure, namely the deceptively simple notated meter. Conceived solely as a container for the time
points, and never articulated in terms of recurrent rhythmic or accentual patterns, Babbitt’s 3/4 or 6/8 measure was
not in fact “a measure in the usual musical metrical sense,” but (like the bars in a transcription of a “medieval” or
“Renaissance” motet) just a notational convenience. Rebar the music in 4/4 or 7/8, or just shift the bars an eighth
note to the right or left, and everything will change for the analyst, although nothing has changed for the listener.
What is analyzed, then, are the relationships that are demonstrable in the music as seen on the page, not the music as
heard” (see Taruskin, "Chapter 4: The Third Revolution,” 168). In his article, Babbitt himself acknowledged, “[a]
pitch representative of a pitch-class system is identifiable in isolation; a time-point representative cannot
conceivably be, by its purely dispositional character” (see Babbitt, “Twelve-Tone Rhythmic Structure,” 63). This
seemingly deterministic measure of 3/4 or 6/8 allows for a conversation with the analyst and composer - revealing a
path for reverse engineering the compositional process - but theoretically creates no audible consequence intrinsic to
its nature.
5
While some of Babbitt’s oeuvre involves no human performance (such as Composition for Synthesizer,
1961), most of his work is acoustic, having abandoned electronic music in 1975. See Gross, “Milton Babbitt,” non-
paginated. Here, Babbitt in an interview with Gross says, “I never went to computers. I could have started with
computers with Bell Labs with Max Matthews in 1957. But you couldn't imagine what it was like at that time. With
the turnaround time, you might as well have gone out and hired an orchestra. You had the punch cards and the
mainframe computer in which you had to pour your work into. I knew enough about it to know that this was not for
me.… I just decided that I couldn't do anything like [computerized music]. Life is too short. I haven't done any
electronic music since. In 1975, I turned back to instrumental music, not that I wasn't always writing that.”

246
problem seen by some of Babbitt’s colleagues, such as Peter Westergard in his aporetic article6

published in 1965 in response to Babbitt’s 1962 publication, is that Babbitt’s bar lines are only

artifacts of an arbitrary choice in how to serialize rhythm.

Following his time-point idea outlined in Composition for 12 Instruments, Babbitt was

afforded the possibility of realizing with utmost accuracy his most extreme hyper-rhythmic

conceits under the power of new computerized, programmable synthesizers. While this period

begat some purely electronic pieces, Babbitt also composed a few electro-acoustic pieces for

both synthesized sound and human performer. Two examples canonized today are the sister

works, Vision and Prayer (1961) and Philomel (1964), both for soprano singer and synthesized

sounds. Since these works, requiring virtuosically precise execution, are dependent in-part on

human interpretation, the manner and efficacy of their notation becomes paramount.

This history of the notation of these two pieces, particularly Babbitt’s experience in

notating the computer-programed sound structures composed for them, is important to consider,

for while they are highly similar in their basic materials, forces, and compositional methods, and

while their notation appears aesthetically similar on the surface, these pieces represent Babbitt’s

navigation of a practical issue in exceptional rhythmic notation: mathematical accuracy.

In Eric Chasalow’s interview with Milton Babbitt in 1997, Babbitt discusses the notation

of the tape part from Vision and Prayer. Babbitt remarks that he decided to “fake a little bit and

6
Westergard, “Some Problems,” 113. To this end, Westergard writes, “we have been at least partially
conditioned by pre-Schoenberg pitch structures to hear pitch relationship mod. 12; i.e. we can be expected to hear a
family resemblance in the opening interval of any P pitch set be it up a semitone, up thirteen semitones, or down
eleven semitones. But have we been even partially conditioned by pre-Babbitt rhythmic structure to hear durational
relationships mod. 12; i.e. can we be expected to hear a family resemblance between a dotted quarter-note followed
by a sixteenth note (the opening ‘interval’ of duration set P6 and an eighth note followed by a dotted eighth notes
(the opening ‘interval of duration set P2? The perceptual problems outlined above are further intensified by problems
of performance. It would be difficult enough to differentiate between durations of ten and eleven sixteenth notes
defined by pairs of attacks controlled by one player on one instrument. But the attacks which define the durations
sets of Composition for Twelve Instruments may come from as many as twelve different players, each playing an
instrument with a different response time.”

247
cheat on the score, and simply have those values which are closest to reasonable values,” since

the rhythms that he constructed in the tape would have looked “terrifying” if notated precisely,

and he likely “could not have even read [the rhythms] [him]self.”7 He goes on to reveal that two

years later, when writing Philomel, he forwent the approximations, perceiving from past

comments given by singer Bethany Beardsley while performing Vision and Prayer that the

approximations did not help the performer as he thought they would. Consequently, in Philomel,

Babbitt wrote precise values for the tape.8 In the manuscript’s preface of Philomel, Babbitt even

assures the reader of the accuracy of his notation, barring a few unspecified exceptions:

Note: This “score” is a complete vocal part, and – with exceptions – a total
representation of the rhythmic and pitch content of the synthesized and recorded
accompaniment in all those sections of the work in which the singer participates. The
exceptions occur when, to avoid notational complexity, the rhythmic representation is
only closely approximate, and registral relations are simplified.9

Despite Babbitt’s best efforts, however, there are some mistakes in the notation of both Vision

and Prayer and Philomel; such inaccuracy of complex rhythmic notation points to some of the

confusion and limitation of tuplets in our current system.10

Mimmi Fulmer makes a comprehensive catalogue of the mistakes in Vision and Prayer as

part of her critical edition of the piece; however, there is no critical edition for Philomel, and

consequently there is no full catalogue of all the mistakes one may encounter in this substantial

7
Chasalow, “Milton Babbitt,” 14’10” – 15’00”. Babbitt does assure us, however, that all the music in the
acoustic voice part of Vision and Prayer is precisely notated, though the voice part for Vision and Prayer is far less
rhythmically complicated than the accompaniment. The same is true for Philomel.
8
Ibid., 16’40” – 17’00”.
9
Babbitt, Philomel, front matter (non-paginated).
10
Babbitt, Milton Babbitt's Vision and Prayer, vii. Mimmi Fulmer notes in her critical edition of Babbitt’s
Vision and Prayer that “[t]here are some mistakes in the score. Many of these are missing triplet signs or rhythms
that don't match up. One that caused me great confusion is in measure 27: until I realized that there is an extra eighth
rest in the tape pan, I couldn't figure out how to place ‘1’ in relation to the F in the tape.”

248
work. While the purview of this brief overview of Babbitt’s contribution to the diversification of

rhythm during the late 20th century is not to uncover all mistakes in Babbitt’s score of Philomel,

a few will be noted as a pointer to problems in contemporary use of tuplets as part of exceptional

rhythmic practice. Some of these mistakes are magnified by their inconsistent application

between Babbitt’s manuscript score and Norton’s anthological score of Philomel’s first part.11

There are, in some places, a few possible solutions to seeming notational rhythmic errors.

Any rhythmic analysis of Vision and Prayer is made even more onerous by Babbitt’s

compromises in the rhythmic notation of his synthesized accompaniment; it is now perhaps

impossible to know what manner of approximation he employed and where he used it.12 Thus,

any attempt to reverse engineer his compositional process is challenged by uncertainty. In

Philomel, however, we have a nearly accurate representation in the pitch and rhythmic

dimensions of his synthesized music. Thus, it affords us a reasonable opportunity to accurately

study his rhythmic structures and the efficacy of their notation.

While arguably arbitrary to the ultimate audiation, Babbitt’s bar lines do allow one to

easily parse his method of event-construction according to his time-point serialization technique.

Without any 3/4 bar line, rhythmic analysis according to the precepts of Babbitt’s method would

be quite difficult to undertake or understand, especially if not taken from the first event. A time-

point “analysis” for the opening of Babbitt’s Philomel is provided in Example 5-1. In the

opening, Babbitt’s large rhythmic structures begin at time-points according to the 8th

11
namely, it is sometimes difficult to ascertain whether a mistake lies with Babbitt initially, Norton
secondly, or both, since the mistake is not specifically noted by Norton in their edition, and it is unknown if Norton
consulted Babbitt on the creation of this edition, though we may assume they did. Perhaps, Norton’s “corrections,”
may themselves not be the “correct” corrections, especially if the emendations were done without Babbitt’s
consultation.
12
Babbitt died in 2011 and did not leave any record of the precise method of his approximations.

249
 I0 I11 I1 I9 I4 I6 I3 I2 I7 I8 I5 I10
P0 4 3 5 1 8 10 7 6 11 0 9 2 R
P1 5 4 1
P11 3 4 11
P1 7 4 3
P8 0 4 8
P6 10 4 6
P9 1 4 9
P10 2 4 10
P5 9 4 5
P4 8 4
P7 11 7 4
P2 6 2 4
RI
Figure 5-1. P0, I0, and I8 for the tone matrix of Babbitt’s Philomel.

transposition of his inverted prime row (I8) of his 12-tone matrix.13 A skeletal tone-row matrix is

provided in Figure 5-1 above; recall the numbers 0 through 11 correspond to the first through

twelfth 16th-notes of a 3/4 measure.

A critical question must be asked. Which is more important in a score: an aid to the

analyst in reverse engineering the logic of the composer’s method of organization, or an aid to

the performer in parsing a musical cypher to accurately receive, digest, and reproduce a desired

sound?14 Obviously one might wish to adeptly accomplish both ends in their work, but under a

13
While the piece begins almost exclusively with the use of prime tone-rows (namely P0, P1, P11, and P3),
the rhythmic aspect begins seemingly obscurely on the minor-sixth transposition of the inversion. Why? It seems
likely that Babbitt might have not wanted to begin the piece on, for example, the fourth 16th-notes of the first
measure as would be required by P0, since this would not be perceptible to anyone other than those reading a score.
Rather, he likely wished to begin on the first beat, thus necessitating his use of a row beginning on “C,” or 0, rather
than one like P0, which beings on “E,” or 4. While beginning on the pitch “E” seems artistically savvy given
Hollander’s text beginning on the sound “Eee,” beginning on this same beat – namely, the fourth subdivision of the
first beat – might seem a little absurd given the piece’s ultimate audiation: the first event will undoubtedly sound
like a downbeat, not like the fourth subdivision of the first beat, since there was essentially no “first beat.” Thus, the
perhaps ultimate conclusion – reductio ad absurdum – of this method of event-organization, particularly in pieces
without a conductor, is that all time-point serialized works must start with some row that begins on 0. To do
otherwise would be imperceptible to the listener, and thus any hope of actually perceiving the time-point row would
be impossible, since the listener would have no way of knowing whether the piece began on point-0 or point-11 or
any point between. A tone-row analysis of this work is nothing new. For a tone-row analysis of this opening see
Taruskin, "Chapter 4: The Third Revolution,” 203.

250
Example 5-1. Norton Edition of Milton Babbitt’s Philomel, Part 1. Overlaid in the lowest two
staves are the time-point events. Babbitt is using the I8 row of his tone matrix for time points.
In Figure 5-1 showing P0, I0, and I8, I have provided the numerical equivalents of pitches.

complicated notation like Babbitt’s Philomel, Babbitt himself seems to favor the former rather

than the latter. Naturally, a theorist or performer can navigate any difficulties with effort and

persistence, and so Babbitt need not attempt a happy medium to best serve both concurrently;

14
Needless to say, such a contentious and difficultly philosophical question is not the purview of this essay.
There is no implication that I am suggesting an answer to this question. However, such a question and its
ramification should be at least considered.

251
however, what might his music look like if the theory was obscured for the sake of clarifying the

ultimate audiation, which might not reflect the method that obtained it. To return to Taruskin’s

analogy in Renaissance music, one could arbitrarily bar a piece from the Renaissance under a 4/4

time-signature. While this might in some way reflect the mensuration under which the music was

composed, the time-signature might not ultimately reflect the sounding result of the composer’s

work, which was in all likelihood not composed under any bar line.

To this end, I have attempted a re-notation of the first minute of Philomel (bars 1-18 in

Babbitt’s original manuscript). My re-notation accords not to the mod-12 system of time-points

themselves, but rather to the actual time-point events which Babbitt has established. Rather than

leave the method intact – namely, the 3/4 bar line used to determine the arrival of the events – I

have re-barred the score to situate down beats at the beginning of these time-points and avoid

any tuplets carrying over bar lines. The re-notated score can be seen in Appendix B at the end of

this dissertation. While my edition has complicated the “beat” of the piece, it arguably has made

the “beat” more meaningful in accordance with the audiation of Philomel.

To demonstrate the metronomical effect of both the 3/4 time-signature and my own

edition, I have supplied links15 to four audio files: the first (Audio Example 5-1a) with a 64bpm

to the quarter-note metronome overlaying the Bethany Beardsley recording of Philomel, the

second (Audio Example 5-2a) with an irregular metronomic beat (still with 64bpm to the quarter-

note) according to my own edition with groupings of two and three 16th-notes. You might find,

as I do, that the 3/4 metronome sounds somewhat meaningless to its surrounding music; this may

be because most musical events in Philomel do not occur on a strong 3/4 beat, and even less

commonly on metrical downbeats. Under my edition, far more events occur on beats, and almost

15
Follow this link to hear the audio examples for this essay: [Link]
babbitt-audio-examples

252
every complex rhythmic event occurs on a downbeat. The metronome on Audio Example 5-2a is

more meaningfully tied to the surrounding music (albeit the understandable minute imprecision

from the performer). To further demonstrate Babbitt’s notation’s metronomic qualities versus my

own edition’s, I have provided a third and fourth example (Audio Examples 5.1b and 5.2b)

which are synthesized versions of Philomel, such that all parts, including the voice part, are as

precise as can be (obviously, the sounds in this example are rhythmically representative of

Babbitt’s intent, but not timbrally). Audio Example 5-1b uses the 3/4 metronome, while Audio

Example 5-2b uses my irregular metronome.

While this re-notation adds complexity to the beat of the music, it makes the metrical

structure more representative of the final audiation of the piece. Benefits to re-notation also

include less mathematical hurtles in calculating the metric positioning and duration of complex

tuplet structures. However, this new version does obscure the method of serializing musical

events, making a dialogue with the theorist more difficult. Below, I have excerpted a few

instances where the rhythmic notation has been starkly clarified by this metrical restructuring.

One interesting simplification can be seen between mm. 15-16 of the Norton/Babbitt

editions (Example 5-2 and 5.3 respectively) and my own edition in m. 22 (Example 5-4) all

shown below. Take note of the tuplets carried over the bar line in the Norton/Babbitt scores.

Some questions arise regarding the necessity of the tuplet notation in the synthesizer and its

execution in the acoustic voice. For instance, why must the tuplet be as such in the synthesizer

when its secondary measure’s component is simply a rest? Could not this tuplet be confined to

m. 15 exclusively, making the notation appear less daunting than it already does. There is no

necessity to write in this way. Furthermore, a question regarding audiation of the voice is worth

posing: what is Babbitt’s idea concerning the nature of staccato and staccatissimo markings,

253
namely what is the difference between an 8th-note with a staccato and a 16th-note? This measure

would look less daunting, if simply notated with 16th-notes or 8th-notes in the voice with either

staccatos or staccatissimos respectively. Perhaps Babbitt imagines a difference in the sound of a

16th-note with staccato and a 32nd-note, but he does not say as such, so we cannot know and

should perhaps default to these marking’s standard understanding, which would allow for re-

notation without theoretically changing the nature of the resultant sound.

Example 5-2. mm 15-16, Norton Edition of Milton Babbitt’s Philomel.16

16
Burkholder, "Milton Babbitt,” 1264.

254
Example 5-3. mm. 15-16, Babbitt Manuscript of Milton Babbitt’s Philomel.17

Example 5-4. mm. 19-22, Key Edition of Milton Babbitt’s Philomel.

17
Babbitt, Philomel, 3.

255
 My edition demonstrates notation changes to each of these issues. In Babbitt’s version, the

tuplet in the synthesizer in question begins with 5 more 16th-notes needed in m. 15. The event

then extends 4 16th-notes (or 1 quarter-note) into the next measure; this quarter-note is simply a

rest; so, why have this tuplet go over the bar line, since it need not be so complicated? We can

confine the tuplet entirely it to the prior measure and begin m. 16 with a typical quarter-note rest.

Furthermore, the tuplet in the synthesized sound need not be under a tuplet at all if metered under

a 9/16 measure, as in my own edition. This rhythmic event encompasses 9 16th-notes; thus, we

could notate this gesture within one measure of 9/16 rather than carry it somewhat unnecessarily

over the bar line of an arbitrary 3/4 measure. Since this tuplet is really expressing 12:9 16th-

notes, and since the 12 16th-notes are grouped by Babbitt into sets of three quarter-notes, we can

divide our rhythmic ratio by 4, giving 3 quarter-notes : 9 16th-notes. We can easily divide 9 16th-

notes into groupings of 3 16th-notes by the use of a dot on an 8th-note, giving 3 dotted 8th-notes in

this moment. Now, we essentially have 3:3. To check our work, 3 dotted 8 th-notes totals a half-

note plus 1 16th-note (as Babbitt’s tuplet originally encompassed). Thus, this tuplet can be

simplified under a 9/16 measure by dividing the measure into three groups of dotted 8th-notes

and subdividing each dotted 8th-note into two dotted 16th-notes. Then, the 8th-note in Babbitt’s

original tuplet simply becomes the first half of the second beat of a compounded 9/16 measure as

seen in my edition (m. 22). Furthermore, if we accept the proposal regarding staccatos and

staccatissimos outlined above, the voice part can be more clearly and cleanly notated without

excessive and obfuscating rests, since either 16th-notes with staccatos or 8th-notes with

staccatissimos will sound as 32nd-notes (seen in my edition excerpted above on the ossia staves

above the original tuplet notation in the voice).18

18
Naturally, some might find this problematic, but I am not sure what measurable difference there might be
in the audiation of the ossia staff versions in either m. 20 or 22 of my edition and Babbitt’s original tuplet with

256
 It is also important to note a nagging problem remaining at this juncture in Philomel: this

tuplet is not notated in Babbitt's original manuscript and is notated in Norton’s edition. The

tuplet is mathematically necessary, though I am not sure if Babbitt or some authoritative source,

like Bethany Beardsley, was consulted on Norton’s inclusion the tuplet. Norton’s solution is the

most obvious, but Babbitt might have intended something else. What is certain is that Babbitt’s

manuscript is missing something necessary at the end of measure 15 or the beginning of measure

16. Such errors appear frequently throughout the piece (see other noted errors Appendix B).

While the time-point serialization is obscured in my re-notation of Philomel, what my

notational rewriting does demonstrate is Babbitt’s “block structuring” of his rhythmic

complexes. The 3/4 barring belies the lack of overarching fluidity to the gestures in the first part

of Philomel (each gesture appearing on the score as fluidly crossing metrical delineations).

However, throughout most of Philomel, and especially prevalent in the first part, all complex

rhythmic events begin and end together, structured as blocks of rhythmic gestures; this is clearly

demonstrated by my edition’s new bar lines. Interestingly, as the piece progresses Babbitt

gradually applies more fluidity to these blocks, allowing these complex rhythmic gestures to

dovetail and interweave.19

numerous rests, as seen on the primary staff below the alternative renditions. Any empirical test of this hypothesis is
not the goal herein, but we should, at least, pose this question, since it merits consideration. Unfortunately, Babbitt is
no longer able to offer any explanation for his choice of rests vs. staccatos here.
19
The choppiness demonstrated by the rhythmic blocking throughout most of Philomel’s first part is
perhaps analogous to the state of Philomel’s character at the outset of this work, unable to speak beyond furtive and
meaningless utterances, having had her tongue cut out by her rapist assailant. As the piece progresses, Philomel is
granted her ability to speak and sing again. throughout this transformation, Babbitt continually increases the
interwoven fluidity of his rhythmic gestures, perhaps reflecting not only the reemergence of Philomel’s voice, but
her persistent tapestry “weaving” to visually communicate her injustice. Perhaps the initial exclusive use of block
gestures, one cutting directly into the next, also reflects the severing of Philomel from her powers of speech: the
cutting of her tongue. This a symbolic aspect of Babbitt’s Philomel begs further hermeneutical exploration. In
Ovid’s version of this story, she regains her communicative powers through the domestic skill of weaving – a central
of part of traditional roles of women in Ancient Greece. At the end of the tale, she regains her ability to produce
meaningful vocalization by the power of the gods, who turn Philomel into a Nightingale. There are numerous

257
 There is much that needs further investigation in Philomel; while much has already been

done to explain the pitch ordering and structuring in the work (perhaps the most surface level

exploration of the piece’s contents) as well as the hermeneutics of the work,20 little has really

been said in regard to its notation and rhythmic structuring, though these are perhaps the most

important aspects to consider when approaching this piece as a performer, let alone theorist or

devoted composer.21

In his interview with Eric Gross, Milton Babbitt commented on the personal meaning of

Philomel within his oeuvre. Gross asked, “How do you think Philomel fits into your work

overall?” Babbitt, perhaps with a hint a wit, responded, “That's a good question. It doesn't.”22 He

variants of this story, however, one in which the gods actually turn Philomel into a songless bird – perhaps a less
“happy” ending for Philomel.
20
For example, see Emily Adamowicz, "Subjectivity and Structure in Milton Babbitt’s Philomel" (Music
Theory Online. 17 (2), 2011); Bethany Beardslee and Minna Proctor, I Sang the Unsingable: My Life in Twentieth-
Century Music (2017); Richard Crawford, An Introduction to America's Music (New York: Norton, 2001). Mary
Lee Greitzer, Tormented Voices (PhD diss., Harvard University, Ann Arbor, MI: UMI/ProQuest, 2007); David
Lewin, Studies in Music with Text (Oxford: Oxford University Press, 2006); Christopher Macklin, "Musical
Irrationality in the Shadow of Pythagoras" in Contemporary Music Review, 29, no. 4: 387-393 (2010); Andrew
Mead, “‘One Man’s Signal is Another Man’s Noise’: Personal Encounters with Post-Tonal Music” in The Pleasure
of Modernist Music: Listening, Meaning, Intention, Ideology, ed. Arved Ashby. Rochester, New York: University of
Rochester Press (2004); Richard Swift, “Some Aspects of Aggregate Composition” in Perspectives of New
Music 14, no. 2: 236–248 (1976); Glenn Watkins, "Chapter 25: The United States of America" in Soundings: Music
in the Twentieth Century, 528-41 (New York: Schirmer Books, 1988);
21
How was Babbitt formulating the length of his events? For example, consider the first E in the highest
register of the synthesizer; it lasts 86 16th-notes. Why? 86 is 2 in a mod-12 system; is that meaningful in any way?
The beginnings of many of his rhythmic complexes seems all too clear, perfectly following his time-point method;
however, the design of the endings of these events is less clear. Sometimes the ends of events coincide with the
beginning of the next event, in which case the ends are derivative of the duration-interval aspects of Babbitt’s time-
point method. However, this is not always the case. This requires further investigation. Furthermore, how does an
event that begins within a tuplet, rather than the tuplet’s beginning, such as seen in measure 16 of my edition, fit into
Babbitt’s 16th-notes oriented time-point method. In this example, does the lowest voice’s rhythm actually operate as
part of the whole synthesizer’s rhythmic structure in this measure, meaning that the high-F# is, in fact, the beginning
time-point origin of all this material and only notated as it is for convenience. If this is the case, then we can explain
away any confusion; however, if this is not the case, what was Babbitt thinking in terms of his time-point method –
how does the second 16th-notes of a 10:7 tuplet fit into the twelve equally spaced time-point system?
22
Gross, “Milton Babbitt,” non-paginated.

258
goes on to explain to Gross why he ultimately turned to the “electronic medium;” for Babbitt,

computers were not meaningful in offering “new sounds.” Babbitt explained

[Computerized music] wasn't for the superficial titillation of sounds. It was for, above
all, music[al] time – the way you can control time. There's such a difference between
being able to produce a sound as a performer – being able to strike the keyboard; it's
automatic. To produce a duration, it's totally different. Teaching a child to imagine
rhythm, a succession of durations, is so much more difficult than teaching someone how
to put their finger down in the right place on an instrument. Time has always created
problems with contemporary music; that's why the music wasn't performed and when it
was performed; it was done sloppily. We were tired of this. The idea was that we could
control time as we wanted.23

Perhaps then, Philomel represents for Babbitt the categorical aim of his method of time-control.

Because of the computer’s exceeding ability to precisely control time, tempo, rate, measure, etc.,

Babbitt saw the digital performer as the definitive tool to unquestionably realize his temporal

ambitions, similar to Nancarrow’s use of mechanical devices like the player piano. There is no

question that Babbitt’s work with synthesizers in the 1960s and early 1970s heavily influenced

how he wrote for instruments for the remainder of his life;24 for this reason alone, a closer

examination of Philomel and its contemporary sisters (Composition for Synthesizer, 1961; Vision

and Prayer, 1961; Phonemena, 1970; and Reflections for Piano and Tape, 1974) is paramount to

the understanding of Babbitt’s development of his rhythmic theory and compositional methods.

Given their synthesized dimension, these pieces are perhaps where Babbitt came closest to an

uncompromised distillation of his ideas.

23
Ibid..
24
Ibid.. In this interview, Babbitt says that his experience with Philomel and the other works composed in
the 1960s for synthesizer (both purely electronic and electro-acoustic) “had an effect on what I write now for
instruments. Sometimes I write rhythms that are considered very difficult to play. They can be achieved, and they
have been achieved by performers willing to do it. There was a time that I would write out on the synthesizer and
let them hear what I would want them to be producing on their instruments. It has changed the way I wrote.”

259
 Generally, there are three challenges presented in the hyper-rhythmicity of Babbitt’s music,

which will persist with composers working in hyper-rhythmicity throughout the 20th century.

These challenges include:

1. Tuplets that begin off the typical beat of the measure, occurring rather on some subdivision
of a beat or on some weak beat of a measure.
2. Tuplets that span multiple measures.
3. Tuplets that contain within themselves unusual partitions of the notes within the tuplet; this
may include embedded tuplets.

It is worth examining a few examples of hyper-rhythmic music from the late-20th century to see

how this practice has evolved from Ives’ and even Babbitt by the 1980’s and into the early-21st

century.

Much of the discussion on rhythmic complexity during the late 1980s and early 1990s

stems from rhythmic developments (both dyadically and non-dyadically oriented) persisting and

evolving from the 1960s; in particular, much of the literature orbits the work of Brian

Ferneyhough, whose scores are famously replete with intricately woven nested rhythmic strata

along with seemingly countless other non-rhythmic musical dimensions. Ferneyhough’s

meticulous and hyper-rhythmic approach to notation and compositions first reached a wider

audience of younger composers and received international renown when Ferneyhough helped

coordinate Darmstadt’s Ferienkurse during the 1980s. Composers including Richard Barrett,

Aaron Cassidy, Frank Cox, James Clarke, Chris Dench, James Dillon, James Erber, Michael

Finnissy, Klaus Hübler, Claus-Steffen Mahnkopf and Roger Redgate ultimately responded to one

or more of Ferneyhough’s underlying aesthetics, even while collectively sharing a distrust for

what notation could, or should, represent.25 Ferneyhough’s In Nomine a 3 and Second String

Quartet offer brief summaries of his different approaches to hyper-rhythmicity.

25
See Duncan, The Concept of New Complexity, 4 for a discussion of composers of New Complexity
distrust for the ability of music notation to communicate effectively to the performer.

260
 In his Second String Quartet, Ferneyhough uses concatenated pan-rational tuplets to

create notated tempo rubatoes, accelerations, and decelerations. For example, in the first violin in

the second system given in Example 5-5, we can see the discontinuous transition from 5:2, 4:2,

3:2, 7:2, and 11:2, progressively modulating the speed of the violin from faster to slower and

then again to fast. All four instruments herein follow different trajectories of tempo fluctuation

through the application of various tuplets. Here, Ferneyhough has taken Carter’s idea of

continuous tempo transformation notated through the use of tempo markings and made the

process discrete through relocating the nexus of tempo to the rhythmic subdivision of macro-beat

units.

In his In Nomine a 3, Ferneyhough employed more frequent use of large-scale complex

stratifications of non-divisible rhythmic hierarchies along with nested tuplets. In the first system

of Example 5-6, Ferneyhough presents a tripartite proportion among the three instruments:

[Link] transitioning to [Link] and then [Link]. Within many of these strata appear nested

tuplets, requiring even more complex subdivisions of the given hierarchy. For example, consider

the oboe in measures 8-10. Ferneyhough has taken the FTU of each measure and divided it into a

prime number of equal parts (11, 13, and 31 respectively); then, within this hendectadic,

tridectadic, and 31-division26 hierarchy, he has prescribed variable further subdivisions: 11:7,

27:18 (3:2), 10:8 (5:4), 9:5, and 19:16, each creating a unique multiplicative micro-subdivision

unit of rhythmic time.

26
Technically “triacontahentadic” hierarchy.

261
Example 5-5. Brian Ferneyhough, Second String Quartet, bars 122-125.27

27
Brian Ferneyhough, Second string quartet (London: Peters, 1981).

262
Example 5-6a/b. A) Ferneyhough, In Nomine, bars 8-13;28 B) rhythmic reduction of oboe, bars 8-
10 (note, in 5-6b all values have been doubled for ease of reading and calculation).

28
Brian Ferneyhough, In nomine a 3: for piccolo, oboe and clarinet, [2001] (London: Peters, 2009).

263
Example 5-7. Rhythmic reduction, simplification, and metrical clarification of Ferneyhough’s
oboe part in measure 9 of In nominee a 3 (note, all values have been doubled for ease of reading
and calculation).

Let us consider the instance of the 4/8 measure (refer to Example 5-7 for a simple

rhythmic distillation of the oboe’s passage for ease of calculation): the 13:8 16th-notes can be

simplified to 13:8 8th-notes under a 4/4 measure (standard, unpartitioned FTU), consequently

giving us our first subdivision as 13th-notes (seen as 8th-notes in Ferneyhough’s score); this

means that Ferneyhough’s 16th-notes are technically 26th-notes, which under the 27:18

proportion are subdivided again into three equal parts, making a 78th note the smallest

subdivision of this measure in this instrument. Given the relative indivisibility of the measure

proportion, one can imagine how microscopic the common multiple subdivisions are when all

parts are combined. The level of precision demanded by such a rhythmicity is astronomical by

any human-based processor and ostensibly impossible to perform with complete accuracy.

264
 However, it is worth noting that Ferneyhough’s notation, while abundantly prescriptive

and seemingly meticulous lacks much in regard to rhythmic clarity within this non-dyadic

proportional system. Take for example the proportion 27:18. This is needlessly and illogically

information-dense. Why did Ferneyhough not simply notate 3:2? Furthermore, when one takes

some time to reduce Ferneyhough’s notation based on a 3:2 proportion, we find that there is

much less complexity in this instance than 27:18 would lead one to believe. I have reduced the

oboe’s part in this measure in Example 5-7.

Not only are many of the notes that once were under an embedded tuplet relocated to the

primary rhythmic tree (tridectadic), but we see a much simpler proportion emerge on just a few

notes. Furthermore, one can also see that Ferneyhough’s notation could be much improved by a

more considered approach to rhythmic beaning of his values into more sensible macro-units, as I

have done. It turns out, as with Babbitt, that Ferneyhough’s In Nomine as well as much of his

oeuvre possess many infelicities in rhythmic notation, either with needlessly over-complicated

tuplets or poor execution of rhythmic notation on a fundamental level, such as with illogical

beaming. Even in complex non-dyadic rhythmic structures, as Ferneyhough often presents, there

are still basic rhythmic notation principles at work (beaming, grouping, proportional

simplification when possible), which Ferneyhough often overlooks or else illogically ignores.29

While Ferneyhough’s original rhythmic notation contains the fewest notational

components, it is perhaps the most obfuscating to the gesture’s rhythmo-metric design. Though a

complete re-notation of Ferneyhough’s music is not the prerogative of this document, it is worth

noting that hyper-rhythmic writing like Babbitt’s and Ferneyhough’s tends to regularly lead to

29
It does make one wonder how much care and consideration actually went into Ferneyhough’s rhythmic
designs here and elsewhere.

265
inoptimal communication of ideas through a lack of careful calculation of hyper-rhythmic

proportions and consideration for the conventions of the system as given.

Regardless of any notational fallacies or infelicities, the rhythmic prescription in both

Babbitt and Ferneyhough is demanding, representing perhaps the extreme limit of human-

processable notated rubato and discrete tempo fluctuation. While we do not intend an exhaustive

survey of such hyper-rhythmic use of tuplets, it is worth pointing to a few more examples to

demonstrate differing uses of extreme tuplets and to see the persistence and modification of this

practice into the early-21st century.

Along with Brian Ferneyhough, British composers Michael Finnissy (b. 1946) and Roger

Redgate (b. 1958) also display extreme use of tuplets, often for solo polyphonic performance.

From Finnissy’s oeuvre, English Country Tunes for solo piano is perhaps the most famous

example of Finnissy’s use of hyper-rhythmicity. In Example 5-8 below, one can see Finnissy’s

use of indivisible rhythmic stratification with close, yet indivisible, polyrhythms reminiscent of

Nancarrow’s 60:61 Study for Player Piano. In Example 5-8 one sees close indivisible ratios

Example 5-8. Michael Finnissy, English Country Tunes. Excerpt, page 6.30

30
Michael Finnissy, English Country Tunes (London: United Music Publishers, 1986), 6.

266
Example 5-9. Michael Finnissy, English Country Tunes. Excerpt, page 50.31

Example 5-10. Roger Redgate’s Genoi Hoios Essi measures 69-71. 32

like 20:21, [Link]; also, in this example we see Finnissy’s use of tuplets as discrete

acceleration: in the lower stave we see the increasing rhythmic ratios 15:14, 21:14, and 23:14. In

Example 5-9 we see one of the unique challenges of Finnissy’s hyper-rhythmicity: his use of

complex off-set tuplets. Typically, composers of hyper-rhythmicity, writing for human

performance, do not off-set tuplets as done here, particularly for execution by one performer.

31
Michael Finnissy, English country-tunes (London: United Music Publishers, 1986), 50.
32
Roger Redgate, Genoi hoios essi: pour piano (Paris: H. Lemoine, 1988).

267
Rather than have the tuplets either begin or end together, or have one occur within the other, this

section features complex tuplets that begin and end at unrelated times, creating even more

complex indivisible rhythmic structures with microscopic common multiple subdivisions.

Example 5-10 above, taken from the first section of Roger Redgate’s Genoi Hoios Essi,

presents another example from a hyper-rhythmic composer composing frequently for solo

polyphonic instruments.33 Herein there are four independent rhythmic strands spanning the

register of the piano. While Redgate’s proportions are often simpler that Finnissy’s, and while

Redgate’s are more metrically organized than Finnissy’s, Redgate is known for employing

numerous (frequently more than two) indivisible poly-rhythmic strands, often requiring a soloist

to execute four to six rhythmic strata at once, whereas Finnissy’s music often only present two or

three strata.

The slightly younger American composer Jonathan Dawe (b. 1965) presents hyper-

rhythmicity in a Neo-Classical guise, often employing complex poly-rhythmic strata in a modal,

tonal, or poly-tonal context. His hyper-rhythmicity is also often derivative of regularized

polyrhythms and embedded tuplets. Compared to Ferneyhough or Finnissy, Dawe’s music often

presents a clearly perceivable and regulated macro-metrical structure and often only employs

tuplets with divisions less than 11. His use of embedded tuplets is often also regulated, rarely

using irregular subdivisions of a tuplet, rather equally dividing the tuplet into symmetrical parts

as seen below in Dawe’s Liber de arte contapuncti from 2001 (see Example 5-11), in which the

top staff uses large triplet values which are equally subdivided into regularly occurring 18 th-

33
Composed in 1981, Genoi is Redgate’s first published work and the first of several piano works
comprising a significant proportion of his oeuvre. Later works include Eidos (1985), Pas au-deld (1989), Beuys
(1992), trace (1995), arc (1997), ecart (2003) and Monk (2007).

268
notes. Many of Dawe’s rhythmic conceits seem derivative of rhythmically complex proportional

music of the Renaissance, especially given the pitch space in which he often works.

Example 5-11. Jonathan Dawe, Liber de arte contapuncti 1477/2001, bars 73-76.34

34
Jonathan Dawe, Liber de arte contrapuncti 1477/2001: for string quartet (2001).

269
 Much research has gone into the performability of hyper-rhythmic music (especially that

by composers labeled under New Complexity). Scholars and computers analyzing performances

alongside scores, quantifying how “accurate” the most authoritative recordings and virtuosic

performers actually are, have ultimately concluded that an “accurate” performance of music like

Ferneyhough’s is nigh impossible, if not completely so under human processing.35 However, we

should wonder at the hurdles placed before such “impossible” tasks when the composer themself

fails to optimally, if not considerately or accurately, notate rhythm. In comparing Ferneyhough’s

score to the recording of Ferneyhough’s Second String Quartet, Marsh argues that performance

approximation of Ferneyhough’s music, which seems to be the norm in his oeuvre by necessity,

is required to simplify “rhythmic absurdities,” which, “stem from the difficulties associated with

multiple instruments performing the same complex rhythms.”36 Of course, the “absurdities” of

these rhythms might not merely be from their purely complex nature, but rather from their poorly

calculated, if not ill-conceived, designs, both in Babbitt and Ferneyhough.37 Julia Silverman’s

35
I am thinking here particularly of the research done by Marsh, Smalley, and Heaton .Through Marsh’s
examination of recordings of Ferneyhough’s Intermedia alia Ciacona and Second String Quartet and quantification
of the accuracy of these “approved” (by Ferneyhough) performances, which Marsh shows are riddled with errors.
Marsh, by showing Arditti Quartet’s performance of Ferneyhough’s Second String Quartet rarely matches the music
as notated, concludes that “the performance is vastly different from the score,” resulting in “an approximation.”
Marsh cautiously notes that “there are occasions... when performer [approximation] does appear to come perilously
close to changing the music into something which the composer almost certainly did not intend or predict.” See
Duncan, The Concept of New Complexity, 33. See also Roger Marsh, “Heroic Motives. Roger Marsh Considers the
Relation between Sign and Sound in ‘Complex’ Music,” The Musical Times 135, no. 1812 (February, 1994): 84.
36
Roger Marsh, “Heroic Motives. Roger Marsh Considers the Relation between Sign and Sound in
‘Complex’ Music,” The Musical Times 135, no. 1812 (February, 1994): 84.
37
Of course, Milton Babbitt aimed for a “perfect performance” (as near as possible) of his designs; such is
the requirement of serialized music: one must perform the process accurately to have the authentic audiation of the
process, otherwise the audiation is not representative of the process which the composer hopes to hear. Most
scholars of New Complexity, looking at the works of Ferneyhough, have recognized that this is not the aim of New
Complexity or Ferneyhough. Stuart Duncan writes that in the works of Ferneyhough “notation can never present an
exact encoding of the aural experience; notation is the beginning of a process, not the end.” Ferneyhough himself
explains that “the criteria for aesthetically adequate performances lie in the extent to which the performer is
technically and spiritually able to recognize and embody the demands of fidelity (NOT ‘exactitude’!). It is not a
question of 20% or 99% ‘of the notes’” (see Ferneyhough, Collected Writings, 71). Consequently, Duncan asserts
that “a direct link between the performer and the notation is rejected by Ferneyhough.” To some extent, Duncan, and

270
controversial and polemic view, challenged by proponents of New Complexity, on the

disjuncture between sign and sound in New Complexity gains some reasonable credence in light

of such notational issues; Silverman questions “So is it all about nothing? ... do they have any

particular sounds in mind, or doesn’t it matter? There must be something very interesting for

them in what they do, it is just not clear whether it is the music [they notated].”38

Scores like Finnissy’s present interpretative problems since there is rarely a clear macro-

metrical structure. Consequently, given continuous tuplet use, their frequent notation without

fundamental note of division, and the implementation of overlaying and catawampus

interlockings, the music of Finnissy is very difficult to precisely mathematical parse in any one

moment (granted, Finnissy’s music likely possesses some of the same expectation as

Ferneyhough’s in the exactitude of execution). While the question of performability is not

important for our discussion herein, what is important is aiding the best possible performance

through the clearest possible encoding of information. In the case of hyper-rhythmicity, there is a

dense amount of information that needs to be communicated in the rhythmic dimension, which is

perhaps the most mathematical dimension of music notation. Consequently, this information

perhaps Ferneyhough himself asserts, that on some level the precise prescriptions he, Ferneyhough himself, has
prescribed do not actually matter, or at least not matter as much as such prescriptions might have mattered for
Milton Babbitt a generation before; Ferneyhough sees his aesthetic not as a result of hyper-complexity in ultra-
modernist serialism, but rather as an outgrowth of indeterminacy: Ferneyhough writes that previous music was
“graspable” and “invariant,” “something that can be directly transmitted, [but] this is no longer the case… since
indeterminacy assumed the mantle of progress” (see Ferneyhough, Collected Writings, 5). If some kind of
indeterminacy is the conclusion of Ferneyhough’s prescriptive practice, and his notation is far more needlessly
complicated in places that it need be (as shown above), then might all that “ink of the page” be utterly pointless
beyond mere visual artistry? One might be well justified in questioning whether there is any point at all to
prescriptive notation in Ferneyhough’s music, or if there might be a better path to produce the seemingly
“indeterminate” results he seems to find satisfactory, if not “correct,” without the passive requirement (given via the
presentation of seemingly meticulous notation) of running performers, theorists, and musicologists through
needlessly confused communication that ultimately says “play something sort of like this thing here.” Indeed, is
Ferneyhough’s music more akin to Augenmusik in its use of “complexity,” aimed to use the artifice of complexity on
its surface to illicit a reaction from an emotional processor, rather than attempt complexity to the end of its pure
musical results?
38
Silverman, review of “Aspects of Complexity,” 34.

271
matrix needs precision and careful delineation for processing ease; care must be taken in

information-dense genres to find the optimal, efficacious, and most coherent manner to notate

these hyper-rhythmic designs.

Aggregate Tuplets and Stratified Mensurations in the Late-20th Century

The stratified, aggregate polyrhythms heard in Cowell’s Quartet Romantic and Fabric as

well as many of Conlon Nancarrow’s rational proportional canons also inspired many composers

of the late 20th century to employ tuplets to a different extreme that the hyper-rhythmicists of

New Complexity and similarly rhythmically multifarious genres. Inspired by the “harmonic”

progression and construction of rhythm as outlined by Cowell in New Musical Resources,

composers like Ben Johnston, Akira Nishimura, Michael Gordon, and Peter Alexander

Thoegersen among many others will frequently craft music that operates under a continual

stratification of indivisible polyrhythms.

Ben Johnston (1926 - 2019) was a microtonal composer, often writing in just

intonation and incorporating harmonically organized polyrhythmic stratifications. Like his

contemporary Conlon Nancarrow, Johnston was influenced by Ives and Cowell, though

Johnston was introduced to this preceding generation of experimentalists by composer Harry

Partch. Johnston understood any sense of tonality as the organization of pitches by ratio.39

Similarly, Johnston saw tempo as the organization of rate of speed by ratio; consequently,

Johnston understood pitch as one form of micro tempo and tonality as merely a system of

proportional tempi. Simply, a succession of pitches is merely a fluctuation of very rapid

tempi, wherein the specific pitches are ‘hyper-rhythms’ beating so rapidly that they

39
Ben Johnston, Maximum Clarity and Other Writings on Music (Urbana, Illinois: University of Illinois
Press, 2007).

272
modulated from rhythm to pitch. Like Cowell, Johnston understood all these phenomenon –

pitch, rhythm, and tempo – as aspects of the same singularity only operating at different

relative speeds. This philosophical-acoustic perspective of rhythm and tempo heavily

influenced the rhythmic construction of Johnston’s works, particularly guiding him to

frequently organize rhythm by polyrhythmic strata in a “harmonic” fashion.

Given both the “harmonic” design of many of these polyrhythmic strata and

considerations of human performability, Johnston along with other composers using such

indivisible rhythmic layers often confine themselves to relatively simple ratios, usually using

numbers less than 10. Johnston’s String Quartet No. 4 presents the many ways Johnston

typically uses polyrhythmic strata. Example 5-12 below shows a purely rhythmo-harmonic

conceit by Johnston in this work. Herein, one sees Johnston move from a four-parted

rhythmic ratio of [Link] to [Link] to [Link] to [Link] (via another iteration of [Link]).

Later on, Johnston provides more complex rhythmic stratification against a freely melodic

cello part (see Example 5-13): herein we see the ratios [Link] and [Link] amidst further

use of his simpler ratios from previously. Later still in Example 5-14, we see Johnston using

ratios seen before (namely, [Link]), however now the values within these tuplets are

frequently joined, subdivided, and tied over to other tuplets, providing a more complex and

less clearly delineated use of such rhythmic strata. Perhaps Johnston’s most extreme use of

indivisible polyrhythmic stratification comes from his Sextet, wherein he rapidly and

discretely moves from [Link] to [Link] to [Link] to [Link] over the

course of 14 measures (see Example 5-15 below for the rhythmic skeleton on this section of

Johnston’s Sextet).

273
 While Ben Johnston was one composer who used rhythmic stratification in the late-

20th century to a relatively human-performable extreme in his works, many composers have

Example 5-12. Ben Johnston, String Quartet 4, page 10 (1st-4th system).40

40
Ben Johnston, String quartet number 4: for the Fine Arts Quartet (Baltimore, MD: Smith Publications,
1990).

274
Example 5-13. Ben Johnston, String Quartet 4, page 15 (4th system) -16 (1st system).41

Example 5-14. Ben Johnston, String Quartet 4, page 16 (3rd-4th system).42

41
Ibid.

275
Example 5-15. Ben Johnston, Sextet, rhythmic skeleton of opening sections.

42
Ibid.

276
also used the “harmonic” notion of continuous (usually indivisible) polyrhythmic

stratification to craft their music. Two other composers of note, who frequently use such

rhythmic conceits are Michael Gordon and Akira Nishimura, though both composers

approach the effect to different ends. Gordon’s use of polyrhythmic stratification tends to be

similar to Cowell’s suggestion of “harmonic” progressions of rhythm, though Gordon often

operates in a relatively minimalist aesthetic compared to Cowell or Johnston, usually

employing such polyrhythms under simple and static pitch modalities. Gordon’s Four Kings

Fight Five presents a stereotypical example of his use of polyrhythmic stratification.

Example 5-16 displays [Link] (simplified as [Link]); usually Gordon’s rhythmic

strata are relatively simple, though here I have presented the most complex in this work.

Typically, Gordon’s strata are organized around various multiples and subdivisions of 2 and

3, usually only requiring the creative use of the augmentation dot in compound meters.

Contrastingly, Nishimura typically uses polyrhythmic stratification to the ends of

mensural heterophony. Clearly interested in loose and strict imitation on differing time scales,

Nishimura usually employs layered tuplets to create undulating rhythmic realizations of the same

or very similar pitch material in different voices, similar to a mensuration/prolation canon though

less strictly applied. Furthermore, Nishimura typically obscures his polyrhythmic strata by

passing each instrument freely between the strata, essentially disjoining the musical dimension of

instrumentation from rhythmic design, something less frequently done in composers using

continuous polyrhythmic stratification techniques. In Example 5-17, excerpted from Nishimura’s

String Quartet No. 4, we see the polyrhythm [Link] in any one measure, though taking the system

as a whole we see that each instrument moves freely between these strata, no particular

instrument occupying one rhythmic division continuously, though the rhythmic complex itself is

277
continuously present. Furthermore, one can see Nishimura’s use of pseudo-mensuration canon in

these voices, the violins and viola each starting with similar intervallic and contour content,

though moving at fluctuating rates depending on their respective pathway through the three

rhythmic strata. Later in the same quartet, we see a similar use of disjoined instrumentation and

polyrhythmic strata to the ends of fluctuating heterophony under the ratio [Link] (see Example

5-18).

Example 5-16. Michael Gordon, Four Kings Fight Five, bars. 231-234.43

43
Gordon, Michael. Four kings fight five: for nine instruments (1988). New York, NY: Red Poppy Music,
2020.

278
Example 5-17. Akira Nishimura, String Quartet No. 4, mvt. 3, rehearsal 102.44

Example 5-18. Akira Nishimura, String Quartet No. 4, mvt. 3, rehearsal 111-112..45

44
Akira Nishimura, String quartet no. 4: "Nrsimha" = Gengaku shijūsoukyoku dai-4-ban : Nurushinha
(Tokyo: Zen-On, 2009).
45
Ibid.

279
 Ultimately, the continuous use of indivisible polyrhythmic strata would inspire notation

shorthands, avoiding the use of perhaps needless tuplet signs over every subsequent measure,

particularly in works using continuous and instrumentally fixed rhythmic stratification. Example

5-19, Karen Khachaturian’s46 (1920-2011) Trio for Strings from 1984, demonstrates his use of

dyadic-time signatures to signal the indivisible yet metrically coincident polyrhythm [Link]. In

this example, Khachaturian uses the time signatures 9/8, 7/8, and 5/8 in metrical alignment by

the FTU of a 3/4 partitioned whole-note to signal what would typically be notated as measure-

long tuplets of 9:6 in the violin, 7:6 in the viola, and 5:3 in the cello.

Example 5-19. Karen Khachaturian, Trio for Strings, rehearsal 7.47

46
The son of the more famous Aram Khachaturian.
47
Karen Surenovich Khachaturian, Trio, fuer Violine, Viola und Violoncello (Wien: Universal Edition,
1990).

280
 Ben Johnston was also using such dyadic shorthands his String Quartet No. 5 from 1979

to signal similar continuous non-dyadic subdivisions of an FTU. Example 5-20 shows Johnston’s

whole-note subdivision into 9, 5, and 4 equal parts against the top voice (violin) moving in a 4/4-

time signature 33% faster than the cello’s 4/4 time. The second Violin, Viola, and Cello are all

metrically coincident by the whole note at 30 bpm, but through the use of a 9/8-, 5/4-, and 4/4-

time signature in each respective voice, Johnston has, like Khachaturian a few years later,

sidestepped the use of continuous measure-long, stratified tuplets.48 Even as late as 1995,

Johnston is still using dyadic short-hands; however, in this example he uses this shorthand to

avoid cumbersome ties under a pentadic partition in his String Quartet No. 10 (see Example 5-

21). Though here Johnston reverts to the use of continuous measure-long, stratified tuplets in the

viola and cello (6:5 and 7:5 respectively), Johnston simplifies his first violin part after its first

measure by giving a time signature of 4/4 against the time signature of 5/4 in the rest of the

voices, ultimately avoiding the typical necessity of tying under a pentadic partition.

The creative shorthand for tuplets through the use of metrically normalized dyadic-time

signatures embodies a re-emergent proportional interpretation of time signature by the end of the

20th century; namely, Khachaturian’s and Johnston’s use of time signature here is highly similar

to the Tincto-Gaffurian understanding of mensural proportions in the late 15th and early 16th

century. However, Khachaturian and Johnston were not the first to reimagine the potential

meaning and use of time signature as handed down from the late 17th century. This simple

proportional notation, by use of a mensurally understood signature, points to a rationalized

48
Worth noting is that when considered all together with the first violin part moving 33% faster than the
cello part, the whole section embodies a proportion of [Link]. The notated polytope (given simply as an
alternative to the dyadic time signature shorthand standing in for tuplets), namely [Link] bpm structurally
stands for the following rhythmo-harmonic chord: tonic doubled, just-intoned major third, and a Pythagorean major
second.

281
reclaiming of global numerical operators in music to signal macro-structure proportional changes

in rhythmic hierarchy that would ultimately carry forward into the 21st century.

Example 5-20. Ben Johnston, String Quartet No. 5, page 15-16.49

49
Ben Johnston, String quartet no. 5 (Baltimore, MD: Smith Publications, 1992), 15-16.

282
Example 5-21. Ben Johnston, String Quartet 10, mvt. III, bars 1-4.50

A New Look at Time Signature and Tuplets in the Late-20th and Early-21st Century

Fascinatingly preceding Henry Cowell’s 1930 exhortation from New Musical Resources

to no longer name non-dyadic rhythmic values by their dyadic neighbors (e.g. 8th-note triplet) but

rather by their actual fractional value (e.g. 12th-note), recognizing the reality of 3rd notes, 5th

notes, etc. Mexican composer Carlos Chavez explicitly calls for a singular 12th-note in the final

movement of his Piano Sonata No. 3 from 1928; however, Chavez notates this incomplete triplet

50
Ben Johnston, String quartet no. 10: 1995 (Baltimore, MD: Smith, 1999).

283
value not through some arbitrary note-shape like Cowell would suggest, but rather designates

this value through the use of a non-dyadic time signature: 1/12 (see Example 5-22).51

Example 5-22. Carlos Chavez, Piano Sonata No. 3, page 19, final system.52

Cowell did suggest, just a few years after Chavez’ third piano sonata, such a reimagining

of time signature’s purview in his New Music Resources,53 wherein Cowell demonstrated how

the bottom number in every time signature is, in essence, the division of the whole-note.

Sporadically throughout the 20th century, musical works emerged that employed these kinds of

non-dyadic time signatures.54 Despite pan-rational time signatures’ occasional reemergence

following Chavez and Cowell, it would take nearly half a century before this new avenue of time

signature would be theoretically considered and nearly a complete century before such ideas

would be more widely and consistently applied to music.

The first to theoretically attempt to codify the non-dyadic use of time signature was

composer Paul Creston (1906-1985) in 1979, who outlined in his book Rational Metric Notation

the possibility of notating time signatures in a similarly non-dyadic fashion to simplify the

51
If Chavez was indeed inspired by Cowell in this prescription, he must have been familiar with Cowell’s
earlier pan-rational rhythmic system from Fabric or Quartet Romantic/Euphometric from around 1917.
52
Carlos Chávez, Tercera sonata para piano (México, D.F: Ediciones Mexicanas de Música, 1989), 19.
53
See Cowell, New Musical Resources, 85-89.
54
See Read, Modern Rhythmic Notation, 97.

284
notation of continuous non-dyadic subdivisions or partitions of a FTU, recognizing the

independence of non-dyadic rhythmic subdivision from dyadic hierarchies. Rather than do as

Karen Khachaturian in his Trio and clumsily use metrically coincident dyadic-time signatures,

one could simply notate non-dyadic time signatures, as Carlos Chavez, to signal non-dyadic

subdivisions of an FTU along with incomplete partitions of those non-dyadic subdivisions.

In his book on proper rhythmic notation, Creston polemically notes the persistent and

illogical problem of contemporary notation and its perpetuation by teachers for generations:

The foregoing is the Binary system of measurement. In every book on music theory, [the
binary system of rhythmic notation] is, despite the existence in practice of the Ternary
system, the only [system] presented. It is incomplete, imprecise and unrealistic. It is
incomplete and imprecise because (1) it ignores the Ternary meters, (2) there is no
Metric Signature indicated, and (3) all the symbols are not shown—only note-heads,
stems and beams—while there are 3 other symbols for note-values: the dot, the tie and
the numeral. The unrealistic aspect is in the disregard of rhythmic practice since the 18th
century.55 An examination of just a small segment of the music of Western civilization will
reveal the use of 1/3 notes, 1/6s, 1/12s, 1/24s, etc. In fact, almost every fraction of a
whole note from 1/64 to 63/64 has been employed, not to mention such fractions as 3/20s,
5/12s and 1/28s. The misconception has been perpetuated thru loose and inaccurate
terminology past on from generation to generation, and unquestioned by students and
teachers. Why should a 3/4 note be named a ‘dotted-half,’ and a 3/8 note, a ‘dotted-
quarter’? Higher mathematics is not involved in these calculations—merely simple
arithmetic.56

Of course, Creston was not the first to note this logical flaw in Western music notation, Beswick,

Cowell, and Bartolozzi having addressed this previously (to little avail). However, Creston

moves his reader towards not merely a recognition of non-dyadic rhythmic values, but also their

hierarchical prescription through non-dyadic time signatures, which he interestingly (and

probably unknowingly) couches in a Tincto-Gaffurian proportional manner, questioning the

modern notion that time signatures are not fractions. To this end Creston passionately writes:

55
Of course, as shown herein, this problem stretches back at least to the 17th century, if not to the 16th.
56
Creston, Rational Metric Notation, 36. Note that non-standard spellings appear as in the source.

285
 It has been stated too often that Metric Signatures are not to be regarded as fractions.
Why? Because with traditional terminology, the fractional factor works only with Binary
meters— 2/4, 3/4, 4/4—but not with Ternary meters as traditionally notated—6/8, 9/8,
12/8. The fraction in the traditionally notated Ternary meters, however, is
mathematically incorrect…. In fact, traditional (12/8) can be notated in 4/4, thereby
justifying the proper signature of 12/12.”57

Creston then boldly disregards Cowell’s (and Beswick’s)58 system of note-shapes (though

without mentioning Cowell directly), claiming that with such proportional interpretations of time

signatures, capable of notating any fractional rhythmic value subdivision of the whole note, we

have “no need to invent new symbols for note-values.”59 The echo of Tinctoris from half a

millennium prior in 1472 is remarkable.

Furthermore, an invented system in music can rarely be as effective or as successful as

one that has evolved, one that has developed from the known to the unknown, or that is
the extension or clarification of a principle. (A whole thesis could be written on this very
subject based on the attempts that have been, and are being, made in proposing new
symbols for note-values; but it would not serve the purpose of this book.)60 Every
irregular or unusual note-value can be calculated by a simple mathematical formula and
sometimes by a musical formula. If it is a fraction of a whole note in 4/4, either method is
practical; but if it is a fraction of any note other than a whole, half, or quarter note, then
the mathematical method is simpler. To illustrate: if there are 5 equal notes in a measure
of 4/4— [quintuplet quarter-notes] —each note is 1/5 X (of) 4/4 = 4/20 = 1/5. The
musical method is more involved: 61

57
Creston, Rational Metric Notation, 37.
58
Though I seriously doubt Creston was familiar with Beswick’s system for a variety of reasons.
59
Creston, Rational Metric Notation, 38.
60
The situational irony of Creston’s statement presented herein is not lost on the author of this present
document.
61
Creston, Rational Metric Notation, 38.

286
Creston’s “musico-mathematical” illustration of his method of non-dyadic time signature

prescription demonstrates the notational independence and existence of 5 th-notes and 20th-notes,

and their ability to be structurally regrouped contrary to traditional tuplet necessities, essentially

creating what we would classify as a dectadic hierarchy (see Chapter 1), with 5th-notes

dyadically subdivisible (and potentially pentadically subdivisible, though Creston’s theory of

pan-rational rhythmic notation does not extend so far). Creston’s reasoning is thus: if there are 5

notes to a 4th-note in 4/4 time, there would be 20 equal subdivisions of the whole note in the

same time as there traditionally would be 4; therefore the 5 units of the 4th-note are 20th-notes.62

Creston also brings up the issue of the quarter-note’s unclear pointer reallocation to

different rhythmic values. For example, Creston gives the example of a quarter-note quintuplet

under a 3/4 time signature; Creston explains that through the arithmetical logical of

understanding each of the 5th-notes as 1/5 of 3/4 (namely, (1/5)(3/4) = 3/20), each seeming

quintuplet quarter-note (or 5th-note) is in fact a 3/20th-note:

This brings up a point which proves the imprecision of traditional terminology: When is a
quarter note NOT a quarter note? None of the written quarter notes in [the below
examples] [are] of quarter-note value. The written quarter notes of [in the examples
below] are … 1/6 notes;… 3/16 notes; and … 1/5 notes [respectively].63

62
Creston, Rational Metric Notation, 39.
63
Creston, Rational Metric Notation, 39.

287
 To simplify the prescription of non-dyadic tuplets and dyadic-time signatures, Creston

suggests that we construct time signatures that reduce to a common fraction of the product of the

tuplet’s ratio and the dyadic-time signature’s ratio. With such a pan-rational time signature, the

divisor represents the number of “primary units” and “extrametrical units to the pulse.”64 In

essence, Creston’s pan-rational time signatures help to designate the first- and second-order

subdivision of the FTU (“primary units” and “extrametrical units” respectively), but no further;

any further subdivision we must assume is dyadic. Creston gives the below examples of his

system of pan-rational time signature (see Figure 5-2). Thus, according to Creston, a time

signature of 15/20 designates that the “primary unit” is 3 equal divisions of the FTU, since 15/20

simplifies to 3/4, and the “extrametrical unit” is 5 equal divisions of the “primary unit.”

Strangely and some what contradictorily, however, Creston still defines all this according to a

partition of a dyadic hierarchy (3/4), and so the values of the “primary” and “extrametrical” units

are not quite precise: by Creston’s definition these notes are quarter-notes and 20th-notes

respectively; however, they should technically be 3rd-notes and 15th-notes if considered apart

from dyadicism.

To mend the problems Creston sees in contemporary rhythmic notation, he proposes a

“revised paradigm of common note-values” and gives the table excerpted below as his proposed

system (see Figure 5-3). Significantly, this table includes non-dyadic rhythmic subdivisions and

time signatures; however, Creston’s table is woefully incomplete, only giving notational

standards for dyadic, triadic, and pentadic hierarchies along with the various multiples formed by

them together. Furthermore, Creston seems to necessitate tuplets and ties where they seem

unnecessary (see respectively the 1/5 and 1/3 row in Figure 5-3). Last, Creston fails to give a

64
Creston, Rational Metric Notation, 79.

288
Figure 5-2. Creston’s non-dyadic time signatures with “primary” and “extrametrical” units.65

systematic way by which to interpret and know the implied subdivisions of any hierarchical level

of rhythmic subdivision. Ultimately, while his system points to a needed revolution in the

refashioning of time signature proportionally, his system is far too underdeveloped, cumbersome,

and unsystematic to succeed. However, his suggestion of non-dyadic time signature to freely

prescribe non-dyadic subdivisions of FTUs is significant.

Naturally, Creston’s idea did not receive wide acclaim, since throughout the final two

decades of the 20th century composers were doing as Khachaturian and Ben Johnston and

employing dyadic-short hands for essentially non-dyadic rhythmic constructs. One composer,

beginning in the 1990s would bring to the foreground of contemporary musical discourse the use

of pan-rational time signatures: Thomas Adès.

The first composer to frequently use Creston’s non-dyadic time signatures was British

composer Thomas Adès (b. 1971). Though Adès has never reported that Creston’s notational

system provided him with the impetus to use “irrational time signatures,” Adès does cite as

“great” American artists “an insurance millionaire and a man who lived in a garage in Mexico,”

65
Creston, Rational Metric Notation, 79.

289
Figure 5-3: Paul Creston’s “Table of Common Note Values” from Rational Metric Notation.66

66
Creston, Rational Metric Notation, 42.

290
referring to Charles Ives and Conlon Nancarrow in a New Yorker article from 1998,67 both of

whom certainly would have led Adès to Cowell and the idea of non-dyadic time signatures.

Adès’ first use of non-dyadic time signatures occurs in his work for solo piano from the mid-

1990s, Traced Overhead, wherein he regularly calls for a partitioning of hexadic time (2/6),

essentially prescribing two classically notated quarter-note triplets without the necessity of the

tuplet or its completion (see Example 5-23 below).

By the outset of the 21st century, Adès was employing polyphonically layered non-dyadic

time signatures, essentially combining the practices of indivisible polyrhythmic stratification and

the use of incomplete tuplet aggregates. Adès Piano Quintet from 2001, is to date his most

extreme use of layered and independent non-dyadic time signatures, standing as perhaps the

present, singular paragon in the exploration pan-rational rhythmic hierarchies, incorporating

complex hyper-rhythmicity, layered indivisible polyrhythms, and incomplete tuplet aggregates

under the renewed system of proportional time signature.

Example 5-23. Thomas Adès, Traced Overhead, II Aetheria.68

67
Alex Ross, "Roll Over, Beethoven," New Yorker, October 26, 1998, 116.

291
 Example 5-24 from Adès’ Piano Quintet displays the composer’s use of off-set and

layered non-dyadic time signatures against tuplets in dyadic time signatures. Adès also uses these

time signatures in concatenation, to create a discrete yet constantly fluctuating sense of tempo

without the necessity of tuplet aggregates (see Example 5-25). In the reduction from Example 5-

25, the metronome markings (not present in the original score) are calculated according to the

opening tempo; these variable tempos underscore the metric fluctuations implicit in the pan-

rational time signatures. Belling writes that “[i]t is in these instances where Adès's music most

resembles the new complexity of some of his fellow countrymen.” While Belling admits that

Adès’ notation is optimized to achieve the rhythmic ends it prescribes, he concedes that “the act

of achieving a correct and even tempo which relates to the strings, as instructed by the vertices of

alignment, is at best very difficult.”69 Belling goes on to question, like many have questioned

regarding previous hyper-rhythmic music, “does the effect warrant the expense… of the

complexity?” Belling points to an answer to his question in “closely and critically examining”

the bar in the piano which immediately follows the system given in Example 5-26 below (see

Example 5-27).70

Though Adès’ uses only a simple dyadic time signature here, this measure is nevertheless

encumbered by no fewer than three levels of embedded tuplets. Belling offers in his thesis an

alternative notation, shown in the Example 5-28 below, which he claims “achieves a very close

simulacrum of the aural effect” only requiring one level of subdivision (or no subdivision if the

68
Adès, Thomas. Traced Overhead. Faber Music Ltd, 1997.
69
Belling, “Thinking Irrational,” 35.
70
Belling, “Thinking Irrational,” 36.

292
Example 5-24. Thomas Adès, Piano Quintet.71

Example 5-25. Reduction of Rehearsal Mark 4, Adès, Piano Quintet, 5-7.72

71
Thomas Adès, Piano Quintet (London: Faber Music, 2017).
72
Metronome marks in this figure were added and do not appear in the score. Reduction found in Belling,
“Thinking Irrational,” 26.

293
Example 5-26. Thomas Adès, Piano Quintet, rehearsal mark 3.

Example 5-27: Thomas Adès, Piano Quintet, excerpt from the Piano

Example 5-28. Belling’s alternative piano part, 5 after [3] in Thomas Adès’ Piano Quintet.

294
bar were made compound). Belling presents them with relative spacing to demonstrate visually

the temporal closeness of Adès’ original and his alternative.73

Belling’s conclusions to his notational questions regarding his reworkings of Adès’

perhaps overly complex rhythmic prescription question what, if any, material difference there

would be in the performance of either of the above notations? Would either be performed

accurately at all? Belling wonders whether a similar compromise could also be found for the

music from Example 5-26. However, Belling ultimately concludes that

… Adès has determined, in this instance, that [this passage’s] difficulty does not
supersede the value of the rhythm [Adès] has specifically conceived. This could be
related to the fact that he played that very part himself at the premiere, and therefore had
already internalized the rhythms…. Here is a composer who is meticulous with rhythm,
and in a piece of rhythmic extremes, the detail, he says, is worth it.74

Adès himself in later works acknowledges the unusual notational challenges of rhythmic

complexity, particularly under pan-rationalism, through providing his own alternative notations.

Example 5-29. Adès, Violin Concerto, violin solo, rehearsal 14, two notations by Adès.75
73
Belling, “Thinking Irrational,” 36.
74
Belling, “Thinking Irrational,” 37.
75
As taken from Hue Belling, Thinking Irrational, Thomas Adès and New Rhythms (Unpublished master's
thesis, Royal College of Music, London, United Kingdom, 2010), 41; see also Thomas Adès, Violin Concerto:
Concentric Paths : Op. 23 (2005) (London: Faber Music, 2010).

295
Through notating by the lowest common rhythmic denominator (namely, one sixth of a quarter-

note) Adès is able to present the violin solo in Example 5.29 in a more conventional, albeit busy,

notation. As Belling notes, at the relatively slow tempo indicated it is “entirely possible to

imagine the performer using [sextuplet quarter] notes (1/24th notes, in [non-dyadic] meter

parlance) as the base counting units.”76 Belling offers an alternative in-line with Adès’ use of

non-dyadic time signatures (see Example 5-30); however, Belling does admit that “[i]n this

alternative, this superfluity of information would hardly serve to reduce confusion.”77

Example 5-30. Adès, Violin Concerto, violin solo, Huw Belling’s second alternative notation.78

Following the successful performance of his violin concerto, Concentric Paths, Adès was

encouraged to revisit pan-rationalisms through broken tuplets and non-dyadic time signatures in

other orchestral works, including Tevot in 2007 and Totentanz in 2013 (see Examples 5.31 and

5.32 respectively). Though with a touch of concern, Billing notes that in the case of Tevot that

“[a] mere two years after experimenting with broken triplets, Adès now trusts… the rank-and-file

76
Belling, “Thinking Irrational,” 42.
77
Belling, “Thinking Irrational,” 43.
78
As taken from Belling, Thinking Irrational, 43; see also Adès, Violin Concerto.

296
Example 5-31. Thomas Adès, Tevot, after rehearsal mark F.79

Example 5-32. Thomas Adès, Totentanz, after rehearsal mark 2.80

79
Thomas Adès, Tevot: for large orchestra (London: Faber Music, 2007).
80
Thomas Adès, Totentanz: For Mezzo-Soprano, Baritone and Orchestra (London: Faber, 2013).

297
orchestral musician with these broken rhythms [, since] [t]here is no alternative notation” to

these rather complex rhythmic conceits.81

Beginning with Traced Overhead, culminating in Piano Quintet, and refined and

simplified in Concentric Path and later orchestral works, Adès’ rhythmic pan-rationalism finally

realizes one of Cowell's fundamental rhythmic observations from the 1930s: base-two rhythmic

subdivisions and hierarchies are not the necessary truth to discrete musical time.

Conclusion to Chapter 5

Up to the present, Cowell’s and Creston’s system of pan-rational time signature is still

receiving attention from and use by composers, albeit exceptionally. Videos on YouTube have

begun featuring “irrational time signature” explanations and discussions, and websites are

devoted to offering tutorials for the performance of these operators. If not Creston, Adès’ use of

these pan-rational signatures has spurred enough interest that finding the occasional use of non-

dyadic time signatures in music, such as in Peter Thoegersen’s Hypercube from 2012 (see

Example 5-33), is becoming more and more frequent.

While the late 20th and early 21st century bear witness to an extreme diversification of

rhythm to the point of hyper-rhythmicity, there is still an abiding problem with Western music

rhythm as notated and, consequently, practiced: the reliance on and implicit supremacy of

dyadicism. Even in the most complex hyper-rhythmic tuplets, stratified indivisible

polyrhythmics, or non-dyadic time signature arrays there is still the underlying fundamental

divisibility by two implied. When Finnissy prescribes a 31:17 tuplet each part of that tuplet is

still necessarily divisible by two according to the standard interpretation of our basic set of notes.

In complex note-shape systems like Henry Cowell’s from New Musical Resources each non-

81
Belling, “Thinking Irrational,” 44-45.

298
dyadic value is, after the first subdivision of the FTU, an intrinsically dyadic hierarchy. Even in

the non-dyadic time signatures of Cowell, Creston, and Adès the non-dyadic subdivision of the

whole note is, again, intrinsically dyadic after this first subdivision.

Example 5-33. Peter Alexander Thoegersen, Hypercube, first page; see viola and cello in second
system for 20/20 and 6/6 respectively.82

82
Peter Alexander Thoegersen, Hypercube: a polytempic polymicrotonal work for string quartet (2012).

299
 Without the tuplet, there is no way to indicate multi-level non-dyadic divisions of some

rhythmic value. Being consequently reliant on the tuplet, at least on some level of rhythmic

prescription, we are also bound to the completion of its aggregate; consequently, though

Cowell’s system and Creston’s system allow for the intermingling of incomplete tuplets, they

cannot allow for the intermingling of non-dyadic values either outside the given note-shapes

(Cowell) or beyond the “primary” and “extrametrical” subdivision levels. Furthermore, the

system of pan-rational time signatures at present does not allow for the intermingling of dyadic

and non-dyadic note values within a signature’s measure; one must always segregate “irrational”

values from “rational” values by the bar line. While more freedom is given by foregoing the

tuplet bracket in pan-rational time signatures, composers are still bound by the bar line and must

isolate their rhythmic hierarchies. Though the joining of various hierarchies within a measure is

allowed by Cowell’s note shapes, we must remember that these shapes neglect many, rather

simple, rhythmic values and are still essentially dyadic.

Thus, after a century of efforts to attempt a refashioning of Western rhythmic notation,

we are still rather confined in our ability to notate rhythms precisely and freely. Even after over

half a millennium, should we search for answers in past systems, we still have no better

solutions. Every answer offered by the 20th and 21st century thus far has only returned us to some

variation of 14th or 15th century practice (note-shapes or proportion signs). The only innovation

truly “new” from the 20th century (as far as prescriptive rhythmic notation is concerned) is the

ability to notate incomplete proportions (namely, incomplete tuplets). Even under the Tincto-

Gaffurian system, the proportions given always required the aggregate of the proportion;

however, the combination of the partitioning power of modern time signature with the free

proportionality of 15th century numerical mensuration ratios allowed for the most powerful

300
aspect of each operator to emerge. However, as noted previously, this “new” time signature tells

us nothing about further levels of subdivision, which one must assume is dyadic unless a tuplet is

given. But why cannot we have a time signature that prescribes subdivisions on all levels into

five equal parts – begetting 5th, 25th, and 125th notes – or into three equal parts – begetting 3rd,

9th, 27th, and 81st notes. Or even more interestingly, why cannot we even designate multipath

division trees (similar to modern “compound” time signature with augmentation dots) that allow

for choice on each level? For example, rather than two or three division on any level as found in

6/8 (3/4) time, why not allow for 3 or 5 division on any level – begetting 3rd, 5th, 9th, 15th, 25th,

27th, 45th, 75th, 81st, and 125th notes?

Does such a system need note-shapes in proliferation, like Cowell’s, or is there a better

optimization of this problem? How many note-shapes are necessary to make a flexible, yet

human processible system? How can Cowell’s and Creston’s system of pan-rational time

signatures be better interpreted to signal not merely division of the whole-note, but divisions of

any note under that global operator? To answer such questions and design a better version of the

rhythmic notation system we currently use, first we must ask ourselves what makes a “good”

system of notation, particularly given the interpreters and translators of that system.

Meta-Synthesis: What Makes a Good System of Notation?

“No notation, of whatever iconically representational state,” writes Brian Ferneyhough,

“can presume to record information encompassing all aspects of the sonic phenomenon for

which it stands.”83 Indeed, this is likely true; no system of notation can presume to be

comprehensive in its capability to encode and communicate information, nor should a notation

83
Ferneyhough, Collected Writings, 3.

301
necessarily strive for such completeness. It would be hubristic to assume that all cases can be

accounted for or all phenomena have been addressed. Furthermore, it is clear that many such

phenomenon might be unnecessary to account for given the goals of a particular notation.84

Certainly, our present system of musical notation does not claim “completeness;” however, few

demand the ‘incompletenesses’ addressed or mended. For example, in the 391 “new” musical

notation systems from the past three centuries outlined by Gardner Read in his book Source Book

of Proposed Music Notation Reforms only two proposed any intrinsic change to our method of

prescribing rhythm – Cowell and Bartolozzi – and both of these were still reliant on dyadic

hierarchies. To this end, Read writes that

Most inventors of unorthodox notational systems retained common-practice methods in

depicting rhythmic and durational values, concentrating their reform efforts on pitch
clarification. Only when the procedure that individualized pitch made time values
unclear did the reformers search for new approaches to indication of duration.85

However, all such “new” approaches essentially added nothing intrinsically new to the discourse

of rhythm in Western music. Furthermore, none of these 391 proposals of notational reform held

any meaningful traction, all falling into obscurity as soon as they were invented. Nothing

between the 17th and 20th century was as successful as Franco’s innovation in the 13th century.

Why was this seemingly perpetual failure of new systems the case? Of course, the momentum of

84
For example, the English language alphabet currently uses 26 letters. There are, however, far more signs
available, even in the Latin alphabet, that other language systems use. Though, since English does not frequently or
at all call upon the aural phenomenon encoded in those characters absent from its alphabet, there is no necessity to
include such needless characters for the system by which English is abstractly encoded. However, there are obvious
limitations to English’s abstract notation: not all the characters accurately or consistently communicate the sounds
they intend to stand for, English frequently grafts foreign words into its lexicon and so might ultimately necessitate
the grafting of new letters. Furthermore, there are limitations to the evolution of English (or any language) by what
is not available in its system of encoding. The person can make a nearly infinite array of sounds, most of which are
not well defined under a system of notation; it is possible that if such sounds were given notational pointers, then
such sounds might become more readily use in that written language.
85
Gardner Read, Source Book of Proposed Music Notation Reforms (New York: Greenwood Press, 1987),
135. For a complete discussion of attempted modern note-shape designs see pages 135-281. Nearly every proposed
system either applies note-shape design only to pitch, many such systems ignoring the durational dimension entirely
or else applying only a system that encapsulated rhythm as typical.

302
history and tradition is a strong force to redirect, no matter the merit of a different direction; Kurt

Stone poetically acknowledges this fact when he gives his “rather prosaic reason” for the failure

of these many better, clearer, freer, more logical, more pedagogical, more intuitive, or whatever

better qualitative measure these “innovators” might have believed evident in these new systems

outlined by Read:

Most musicians, after spending years since childhood to become experts in dealing with
the complexities and shortcomings of traditional notation, are quite unwilling to learn
new, unproven additional systems, especially since common-practice notation has
enabled them to perform a vast range of music perfectly well. Let’s face it: Notational
innovations become desirable, or in fact necessary, only when music becomes so much
more complex or so completely different in approach and technique that traditional
notation can no longer cope with it, which has been the case since the early 1950s when
music required new notational symbols and systems. As a result, there has been an
overproduction of new devices, including countless duplications or different
interpretations of the same signs, with the invention everywhere of novel notation for
three decades. Fortunately, a trend toward regaining standards, where appropriate, is
underway.86

Composers like Karlheinz Stockhausen, Aaron Cassidy, and Jason Thorpe Buchanan

throughout the late-20th and early-21st century have developed new tablatures, graphics, and

symbols to represent myriad new musical sounds, techniques, gestures, and dimensions; of

course, with each new notation and with each step further from traditional tablatures and systems

is an additional burden on the performer’s processing capacity, much of which has of course

been devoted to the interpretation and translation of perhaps wholly different systems. Thus, as

Read points out, “far from improving the notation standardized at the time of each proposed

reform or substituting a superior new method,” most notational innovations “succeeded only in

creating more problems for the performer than they solved.”87 Consequently, we should ask

86
Ibid., x (in the forward by Kurt Stone).
87
Ibid., 1.

303
whether there is a compromise available when a new notation is perhaps demanded by the sonic

result desired.

As evidenced by our historical survey, one sees that, more often than not, notational

reforms happen not through overhaul but by emendation, redaction, reallocation, and

implantation. With very few exceptions, any enduring change to a given system of information

encoding tends to conform to that system rather than supplant it. While a nuanced gestural

requirement of a work by Aaron Cassidy might not necessitate a careful reworking of the

common system of music notation, a change or new demand in such a fundamental dimension of

music as rhythm might suggest itself as more broadly applicable and essential. Thus, in

considering a new system of rhythmic notation, it is likely best not to offer something wholly

different from the present method of rhythmic encoding but rather offer a new method of

interpreting its current elements and operators with perhaps just a few, relatively simple

additions.

Based on the consistent failure of previous proposals, rhythmic or otherwise, in the

development of a “better” system of musical notation, most of these “revisionists,” according to

Read, must have been unaware that “certain basic conditions must be met before any proposed

reform can succeed.”88 One of the first attempts to codify these conditions occurred in 1847

under the auspices of the Belgian Royal Academy of the Sciences department of Fine Arts. The

academy, in an effort to better understand the present problems and possible solutions to the

deficiencies of music notation at that time, collated and outlined all the known notational reforms

they could uncover up to 1847; they concentrated primarily on those reforms based on the use of

numbers, letters, and stenographic symbols. Through this historical survey they sought to test the

88
Ibid., 4

304
strengths and weaknesses of each proposal through applying the new system’s governing

principles to every current dimension of music notation to ultimately form a consensus on a

singular, new and acceptable method of music notation. Nine years after the Belgian conclave,

Joseph Raymondi, one of the survey’s organizers, published his Examen critique des notations

musicales89 wherein he strongly advocated an improvement to and simplification of music

notation. Raymondi recognized that for at least two centuries reformers had regularly attempted

to locate and fix imperfections in conventional notation; however, being mostly or wholly

uninformed of proposals prior to their time, these reformers were tragically ignorant and

myopically convinced they alone could devise a better method from first principles.90 Raymondi

ultimately failed in reforming music notation, but did significantly recognize the importance of

historical perspective and deferment to tradition in fashioning solutions to notational

deficiencies, wisdom we have attempted to heed herein.

As a consequence of his studies, Raymondi formulated his ten primary criteria for any

notational reform:

1. To make clearer the function of the staff lines and spaces.

2. To obviate the need for ledger lines, thus keeping all the notes on the staff.
3. To make all octaves have the same relationship to the staff.
4. To abolish the profusion of clef signs.
5. To simplify, or abolish, the accidental signs.
6. To improve the visual aspects of duration.
7. To regulate the indications of measure.
8. To improve the notation of irregular rhythmic values.
9. To facilitate the writing down of music.
10. To make musical typography easier to print and to read.91

89
Joseph Raymondi, Examen critique des notations musicales proposée depuis deux siècles (Paris, Libraire
Encyclopedique de Roret, 1856).
90
Ibid.
91
Ibid.

305
Of course, any new system need not aim to change all these dimensions; we are, of course, only

concerned with the 6th through 10th items on Raymondi’s list (primarily the 8th), items which

would be mostly ignored following Raumondi for the next century until Cowell’s widely

acclaimed publication of New Musical Resources (Beswick’s reform being altogether unknown).

Item 9 on this list is also something we should carefully consider, for it is not the goal of a new

system of rhythmic notation to make the literal inscribing of music easier, but rather the goal in

making rhythm more diverse and freely transformable under a modified notation is to make the

writing down of musical ideas that are intrinsically rhythmically unavailable in the current

system available, facilitating a composer to write down whatever rhythmic complex they desire

rather than a limited subset of discrete rhythmic patterns.

Closer to the present and perhaps better articulated is Erhard Karkoschka’s list of

standards for any addition to or overhaul of music notation. In his Notation in New Music from

the early 1970s, Karkoschka proposes that any change to music notation should:

1. possess all the possibilities of traditional notation.

2. not go against tradition without good reason.
3. have enough new technical possibilities to be able to represent the present stage of
musical thinking.
4. be capable of presenting complicated structures in a simpler form than does present-day
notation.
5. have a broad, neutral basis and avoid, if possible, representing any particular style.
6. make easier the transition to individual notation forms, e.g., approximate values and
musical graphics.
7. have no difficulty in being able to represent more than twelve values [intervals] in an
octave.
8. possess a clear correlation between what the eye sees and the ear consequently hears,
taking into consideration two basic characteristics:
a. The visual event must be apparent as the direct translation of the auditory event,
requiring as few additional thought processes as possible and
b. The individual symbols and the totality of symbols must be formed on an optical
basis; they must be "correct” in the visual-psychological sense.92

92
Ibid., 5. See also Erhard Karkoschka, Notation in New Music (New York: Praeger Publishers, Inc., 1972).

306
Significantly, Karkoschka recognizes the importance and efficacy of allying notational reforms

to “traditional notation,” allowing for all the possibilities already inherent and commonly

understood in that notation to remain, ultimately cautioning against ‘reinventing the wheel’ by

going against tradition recklessly and perhaps needlessly. Karkoschka’s suggestion that a reform

be able to represent complicated structures in a simpler form than the current prevailing system

is also significant and a balance that must be carefully considered. What is needlessly or

hopelessly complex and how can it be simplified? Is the simplified version truly simpler to read

and comprehend or a mere substitution of one complexity for another equal complexity?

Furthermore, demanding a modification (especially on a fundamental level) be capable of

representing any currently known style of music is paramount, for if a change is only applicable

to a particular genre, composer, or piece, that emendation is doomed to live and die with that

genre, composer, or singular piece in all likelihood. Providing a simpler, widely applicable

method of notation, rhythmic or otherwise, is not only efficacious but more apt to endure.

Though Karkoschka’s criteria have been disputed,93 his updating of Raymondi’s precepts

remains reasonably valid regardless of any reservations one might have with one of his

expectations. According to Read, Karkoschka’s suggestion that “meaningful notational reform…

not stray too far from the time-tested values of traditional notation” is a “point wisely taken” by

any notational innovator. Karkoschka’s rationale for adherence to our present notation holds that

[this system] has survived centuries of attempts at reform, and justifiably. In spite of
certain inadequacies, it has supported a many-sided musical culture, especially because
it presented pitches and durations to the eye in such a way that the sound corresponded
directly with the visual event even if the notation was not completely perfect.94
93
British musicologist Hugo Cole in his Sounds and Signs disputed Karkoschka’s criterium regarding
graphic appeal. Cole believed that it was fallacy to think that graphic appeal to the eye is essential for efficacious
notation. Moreover, simplification, according to Cole, does not necessarily beget better methods of information
encoding. As an example, Cole pointed out that many languages (English being his primary example) are
abundantly and perhaps needlessly complex and inconsistent but are completely efficient once mastered. See Hugo
Cole, Sounds and Signs (London: Oxford University Press, 1974).

307
By Read’s estimation, it is precisely because the vast majority of reforms fail to achieve “this

essential synthesis” that they have also failed “to make a lasting impact on the development of

notation.”95 Read emphasizes the continual and problematic failure of notational reformers’

“fuzzy thinking” and inability to be subservient to established semiotics:

Time and time again one sees imperfections of one element of conventional notation
overcome at the expense of another crucial aspect of that notation. If the inventor has
succeeded in eliminating the need for accidental signs, let us say, he has at the same time
increased the difficulty of distinguishing one notehead from another. If he has devised a
numerical or alphabetical method of identifying the various octaves of the piano
keyboard, replacing the standard clef signs and 8va indicators, he may have confused the
performer by an illogical sequence of those numerals or letters. If he has worked out a
clever new manner of indicating durational values, he has more often than not made
those distinctions too minute to be easily perceived. If he has invented an ingenious
means of outlining one notational parameter, he has unaccountably neglected to consider
an equally vital factor — pitch, but not duration; an unorthodox staff, but no indication
of register; individual note duration, but no rests. The list of such inconsistencies could
be extended ad infinitum.96

Our present system of non-mensural, prescriptive rhythmic notation is notoriously

inadequate in expressing complex and irregular, particularly non-dyadic, rhythmic structures.

The question is begged as to why composers and theorists from the past 500 years were nearly

wholly uninterested in eradicating this notational incompleteness. Though tuning, tonality, and

pitch notation has undergone many successful and failed reforms since 1500, rhythm since the

time of Tinctoris and Gaffurius has had nothing intrinsically new added to its lexicon, barring the

relatively isolated experiments of Cowell and Nancarrow. Nearly all the tuplets used by the

hyper-rhythmic composers of the 20th century, from Ives to Ferneyhough and beyond, were

theoretically available under the Tincto-Gaffurian system of the 15th century. The one substantial

94
Karkoschka, Notation in New Music, 15.
95
Read, Source Book, 5.
96
Ibid., 6.

308
rhythmic “innovation” the West has seen in the past half millennium and which is now embraced

by composers the world over is the collapse of the mensural system into a fixed-value note

system, wherein the note elements (quarter-note, 8th-note, 16th-note, etc.) have definite, discrete

values which do not change their meaning except by local operators (i.e., tuplets). Perhaps it is

time to return to a freer, yet still prescriptive, use of our present note-shapes, allowing their

meaning to pivot based on global rather than merely local operators.

However, in proposing yet another system of rhythmic diversification and yet another

appendage to music notation as presently given, we ought to be careful to offer a system which

follows from our conclusions laid out in Synthesis I, II, and III, which themselves proceed from

our historical survey (as advised by Raymondi), as well as keep in mind the general suggestions

and observations given by Read, Stone, Raymondi, and Karkoschka.

Our present notation concurs with Conclusion S1.01a and S1.01b as well as Conclusion

S1.03,97 since our current system has both generative and derivative elements that are

systematically derived and are, barring trivial ad infinitum derivation, rather limited in their

proliferation. For example, though a consistent pattern cannot be drawn between the whole-note

and 8th-note derivation, a pattern does begin on the quarter-note, which is consequently a

generative, rather than derivative, element along with the whole-note and half-note, which do not

demonstrate a consistent pattern among the other rhythmic values. Thus, for fundamental

97
For ease of reference, Conclusion S1.01 from Synthesis I states that “The success of a system is
dependent on the set of generative elements and the process by which derivative elements are generated. Such
generative and derivative elements a) need to have some logical and consistent origin for ease of interpretation and
translation, and consequently b) cannot be based on seemingly arbitrary and endlessly proliferating forms lest the
processor fail to interpret and translate due to a dwindling processing capability.” Conclusion S1.03 states that
“Recognizing the limited nature of human perception as a discrete processor, as noted in Observation S1.01, the
elements of the system must not only be logically derivative from each other but must also have some reasonable
limit to variation or invention, disregarding trivial ad infinitium derivation as in Observation S1.02.”

309
rhythmic note-elements, there are three: the whole-note, half-note, and quarter-note; the quarter-

note may have a process applied to it – the addition of flags to the stem – to derive the other,

smaller rhythmic subdivisions. Similarly, for fundamental rhythmic rest-elements, there are four:

the whole-rest, half-rest, quarter-rest, and 8th-rest; the 8th-rest may have a process applied to it –

the addition of flags to the stem and singular first flag – to derive the other, smaller rhythmic rest

subdivisions.

It is worth pausing a moment to problematize the fact that the pattern to derive the 8th-

note from the whole-note requires not a repeating pattern but a series of seemingly arbitrary

processes which never repeats again: whole-note gets a stem to become half-note, half-note gets

colored in to become a quarter-note, and the quarter-note gets a flag to become the 8th-note. Only

following the 8th-note does a pattern emerge. Moving upwards from the whole-note there is also

no discernable pattern to derive the double-whole-note nor the quadruple-whole-note. This need

not necessarily be the case; one could devise a system by which all notes follow logically from a

singular pattern beginning with the whole-note. However, this infelicity is only a minor

inconvenience, given the general efficacy of the present system and its wide proliferation. Thus,

though there is perhaps something to be desired for addressing the seeming arbitrariness of some

of the fundamental yet non-generative elements (e.g. whole-note), the system is simple enough

and, of course, widely known already such that any minor inefficacy as this can be overlooked.

For rhythmic diversification Conclusion S1.01a/b also holds along with Conclusion

S2.04,98 since there are a limited number of operators; among the global operators are the time

signature and (perhaps depending on use) the tempo marking, while the local operators are the

98
Conclusion S2.04 states that “Recognizing the limited nature of human perception as a discrete
processor, as noted in Observation S1.01, the operators of the system must not only be necessarily unique, and/or
logically derivative from each other but must also have some reasonable limit to variation or invention, disregarding
trivial ad infinitium variation as in Observation S1.02.”

310
augmentation dot(s), tie(s), and tuplet(s). Furthermore, all variations of these operators are

systematically derivative, tuplets logically variable according to their expressed mathematical

ratio and augmentations dots systematically additive by a pattern according to the inverse powers

of 2.

However, it might be worth considering how inefficacious our present system is and how

it has been optimized, perhaps overly so, to accord to a particular aesthetic of musical discourse,

namely the Common Practice Period of the 17th through 19th century. While a purely efficacious

system is neither an appropriate goal or theoretically possible, our present system likely does

need a better balance between efficacy and inefficacy as well as optimization and specialization

within quantization as suggested in Conclusion S3.03a/b. Having been crafted towards a

particular aesthetic for centuries, our present system might work well under particular types of

quantizations, but it has been clearly over-optimized for such particular quantizations and so is

often incoherent when attempts to encode new or foreign aesthetics within a system well-honed

to other, perhaps antithetical, ends.

There are a few conclusions from our prior syntheses that our present system does not

perfectly reflect, most significantly Conclusion S1.04.99 While our present notational system is

limited, in the dimension of rhythm it lacks some significant flexibility as soon as one wishes to

depart from reliance on dyadic hierarchies and tuplet operators. Consequently, we see the

“unsystematic grafting” suggested in Conclusion S1.04, which inevitably leads to a system

99
Conclusion S4.04 states that “A system needs to be limited (as in Conclusion S1.03) but flexible, able to
be fluently interpreted broadly while also capable of communicating nuance when and where needed. If a system is
too limited and allows for no flexibility, flexibility could become unsystematically grafted onto the system when the
system has excluded or cannot allow for some unforeseen necessary element. Such a grafting will possibly cause the
system to become unsystematic, which, by Axiom S1.01, will cause the collapse of the system or the ultimate
redaction of such an emendation to re-stabilize the system. Thus, a system must either be amendable within
reasonable limits by some logical process, or, better yet, have the inbuilt capability to carry multiple
interpretations/translations per pointer, which, given some limiting context, cannot be misinterpreted.”

311
becoming unsystematic and consequently collapsing back to its prior state of stability (namely,

the ultimate failure of the proposed reform and a return to status quo). Our system lacks some

“inbuilt capability to carry multiple interpretations/translations per pointer,” particularly beyond

dyadic hierarchies (i.e. a quarter-note is always a quarter-note unless under a tuplet, and then it is

still intrinsically dyadic nonetheless). While our present system allows for the pivoting of

rhythmic pointers, such pivoting is extremely confined, requiring a tuplet, the aggregate of that

tuplet, and further tuplets should further non-dyadic subdivisions of that pivoted element be

required. Additionally, by Conclusion S1.05,100 it is better to allow for more agile pivoting of

elements rather than create wholly new elements for similar types of phenomenon (i.e. rhythm).

How then do we take this somewhat inefficacious and singularly optimized system and

make it more efficacious and allow for a modicum of specialization when exceptional

information needs to be encoded? First, in creating a new system or modifying a present one, it is

advisable to not only work within the present system of notation widely used, but also to make

sure through appending or modifying any dimension of it (rhythm in our case) we do not cause it

to become unsystematic, needlessly cumbersome, or beyond the capabilities of human

processing according to Conclusion S1.03.101 Should we choose to add new elements or

operators, it would be best, if at all possible, to have them be derivative of current elements and

100
Conclusion S1.05 states that “given Conclusions S1.03 and S1.04, it seems advisable to have a system
with hierarchical limiters that allow for the clear pointer pivoting within an element rather than create wholly new
elements for similar types of pointers. Such pivoting itself will need regulating according to the processor, which we
assume is human.
101
Conclusion S1.03 states that “Recognizing the limited nature of human perception as a discrete
processor, as noted in Observation S1.01, the elements of the system must not only be logically derivative from each
other but must also have some reasonable limit to variation or invention, disregarding trivial ad infinitium derivation
as in Observation S1.02.”

312
operators and make sure that any new additions are wholly unique and necessary, according to

Conclusion S1.02a, S1.02b, S2.03, and S2.04.102

Furthermore, through any addition, our present system should still be “limited” but

become more flexible as per Conclusion S1.04. By Conclusion S2.01,103 it is perhaps better to

amend the system not by addition of numerous new elements, but rather through the careful

application of new operators, which by Conclusions S1.05104 and S2.06105 efficaciously allow

for hierarchical limiters for pointer pivoting within elements. Thus, in regard to expanding

rhythmic notation within the present system, it seems more advisable to amend our

understanding and use of time signature (operator), as done in Creston and Adès, rather than

append a large or endless system of note shapes (elements) as attempted under the ars subtilior

or with Henry Cowell. While new elements might become necessary to distinguish greater and

more complex arrays of pan-rational rhythmic values, it is best if any addition of elements is

102
Conclusion S1.02 states that “If one wishes to append a new generative element to an extant system,
such a new element needs to either a) be derivative and follow, via some process, from the elements already present
(as per our first conclusion), or b) be justifiably and wholly different from all other forms, not generatable from
some process and semiologically distinguishable, being itself uniquely distinguished by its intended interpretation
and ultimate concrete translation. Conclusion S2.03 states that “If a particular system and a set of operators is widely
adopted for interpretation and translation, it is advisable to derive any further operators from the operators given, or
else new operators need to be justifiably and wholly different from all other forms, not generatable from some
process and semiologically distinguishable, being itself uniquely distinguished by its assigned logical modifier.
Conclusion S2.04 states that “Recognizing the limited nature of human perception as a discrete processor, as noted
in Observation S1.01, the operators of the system must not only be necessarily unique, and/or logically derivative
from each other but must also have some reasonable limit to variation or invention, disregarding trivial ad infinitium
variation as in Observation S1.02.”
103
Conclusion S2.01 states that “The potential of a system is only realized through the operators applied to
it. Since the application of operators to a system’s elements, in part or whole, is quantization, a system is only
meaningfully interpretable and translatable under quantization.”
104
Conclusion S1.05 states that “Given Conclusions S1.03 and S1.04, it seems advisable to have a system
with hierarchical limiters that allow for the clear pointer pivoting within an element rather than create wholly new
elements for similar types of pointers. Such pivoting itself will need regulating according to the processor, which we
assume is human.”
105
Conclusion S2.06 states that “Operators allow for “pointer pivoting” within a system, as concluded
necessary in Conclusion S1.05 and appended corollaries.”

313
taken systematically and under very limited and necessary cases. However, Creston’s and Adès’

use of non-dyadic time signatures is not truly “pan-rational” and lacks clear systemization. Thus,

it is necessary, as already begun in the first chapter, to give a clearer, more comprehensive, and

flexible definition of time signature should we plan to use it as a global operator towards

rhythmic diversification.

In the next and final chapter, we will see how the first chapter, our historical survey, and

our synthesis observations on notational systems coalesce into a new, more flexible, and

efficacious system of pan-rational rhythmic notation according to pan-rational time signatures

and multipath rhythmic hierarchies designed for human processing.

314
 CHAPTER 6
NEW RHYTHMIC RESOURCES:
A PERFORMATIVE SYSTEM OF PAN-RATIONAL RHYTHMIC PRESCRIPTION

Adoptions, Abandonments, Adaptations, and Additions

Let us first begin by establishing that this “new” notation is not a replacement for

rhythmic notations as presently established. Rather, it is an alternative to basic rhythmic

paradigms when a composer wishes to express rhythmic structures unavailable in standard

notation. I do not suggest that this notation be employed if there is no intrinsic need to do so (i.e.

if the music is in a typical dyadic rhythmic hierarchy). Furthermore, I do not propose that this

system removes or replaces any element or operator in rhythmic notation. To this end, it is first

important to delineate that which will be maintained from tradition.

Adoptions

1. Note and rest shapes will be maintained, though, as we will see, modifications to
their interpretations will be given as well as some additional transformations to
them.
2. The dyadic-rational rhythmic hierarchy tree will be available as it traditionally
appears.
3. Augmentation dots may be used in their traditional fashion, especially in dyadic-
rational hierarchies; however, as we will see, sometimes the augmentation dot can
be forgone should one wish to do so.
4. Tuplets may still be used, though they can be forgone under some circumstances
depending on the desire of the composer. However, there will be instances where
tuplets are more advisable than pan-rational rhythmic notation.

For the sake of traditional continuity, typical note rests and shapes will be employed in

both their traditional fashion and, through the application of global operator pivots, as pan-

rational rhythmic symbols. Since most music, past and present, operates under dyadic rhythmic

hierarchies, the dyadic-rational hierarchy will be maintained as typically given according to the

symbols maintained under Adoption-1. Local operators like augmentation dots and tuplets

continue to find use within pan-rational rhythmic structures, and so they will be maintained by

315
their traditional definition, however transformable under global operators (i.e. an augmentation

dot supplying an additional half of its appended note value will add half of the note according to

the pan-rational time signature).

Consequent of these “adoptions” is the capability of this extended rhythmic notation

system to encode any form of rhythmically prescriptive music previously encodable. Rather than

supplant present notation, this emendation adopts nearly all previous norms and capabilities of

the old system. However, with these continuities, we will abandon four fundamental precepts of

the old system of rhythmic notation:

Abandonments

1. The standard definition of time signature (i.e. top number designating number of
beats per measure with bottom number designating the value which takes the
beat)
2. Obligation to notate under dyadic hierarchies (explicit or implicit)
3. Necessity or de facto use of unidimensional hierarchy trees.
4. Assumption that note-shapes necessarily correspond to dyadic hierarchies.

As already shown in Chapter 1, the traditional understanding of the time signature is drastically

limited and myopic. Thus, while the operator will be maintained, its definition and the definition

of its constituent parts will be discarded and supplanted with our definition as given in Chapter 1

and following. Of course, under pan-rational rhythmic notation there is no longer the obligation

to notate music within a dyadic hierarchy and, by our system outlined below, no necessity of

unidimensional hierarchies, rather allowing for two (or even three) dimensional hierarchies.

Finally, though the traditional note shapes will be maintained and though they may still stand for

their traditional rhythmic values, they may also, under new rhythmic operators, pivot their value

to other non-dyadic hierarchies.

Consequent of these abandoned notions, we will supply two essential modifications, the

first of which was outlined in Chapter 1:

316
 Modifications

1. The meaning of time signature and its constituent parts along with the
classification of time signatures (i.e. perfect regular, imperfect regular, irregular)
2. The meaning of traditional dyadic-rhythmic note values, allowing for pan-rational
pivoting of these values according to the time signature operator.

As already shown in Chapter 1, the more liberal definition of time signature according to pan-

rationalism, allowing for non-dyadic FTU subdivisions, allows for not only perfect regular non-

dyadic hierarchies but also mixed hierarchies under imperfect intermingling of hierarchies

through the combination of prime factors into branching rhythmic trees. In so doing, we must

require that standard note values pivot to signal either dyadic or non-dyadic hierarchies

depending on the time signature operator.

Last, we will create one new set of generative elements and their derivatives, which will

complement our new understanding of time signature as outlined in Chapter 1 and subsequently

in this chapter.

Additions

1. Twofold set of note-shapes and rest-shapes for imperfect-regular rhythmic

hierarchies.

In appending one set of derivations of the generative note shapes already given, we allow for the

concurrent intermingling of two rhythmic trees, allowing for imperfect regular rhythmic

hierarchies under pan-rational time signatures. By this, we provide the clearer delineation of

larger, more complex rhythmic structures in notated music than permitted by time signatures that

are exclusively perfectly regular or those provided by Paul Creston and used by Thomas Adès

(see Chapter 5).

317
 Notation of Pan-Rational Time Signatures and Their Rhythmic Hierarchies

As outlined in Chapter 1, we understand time signature to represent two global rhythmic

operators on some generative Fundamental Time Unit – subdivisional/combinative

(denominator) and partitional (numerator) structure – both of which are independent of the other.

Partitioning may be designated by any positive integer number,1 while subdivision is only

represented in rational rhythmic hierarchies by some integer greater than or equal to 1, though

we can consider 1 a trivial case, since it lacks all hierarchical implications.2

To the operator or partitioning, we aim to add nothing intrinsically new, so here we will

primarily focus on the semiotics of subdivision and combination according to pan-rational time

signature. Recall that a rational time signature may either be regular or irregular; first, let us

consider regular rational rhythmic hierarchies, which themselves come in two varieties: perfect

and imperfect. As the simplest case, let us examine perfect-regular rational rhythmic hierarchies.

Perfect-Regular Rational Rhythmic Hierarchies

As demonstrated in Chapter 1, perfect-regular hierarchies, since they represent a

subdivision tree with only one branch, use time signatures with a prime denominator in the form

𝑘
, where k is generally some integer and p is a prime (e.g. 2/2, 5/3, 3/5, 8/7, etc.). While one can
𝑝

certainly notate a perfect-regular time signature with a prime to some power (e.g. 1/2n, 1/3n, etc.),

1
Zero would constitute a music apart from time (e.g. zero beats per measure), which contradicts our
definition of musical time as given in Chapter 1. Negative time signatures were discussed in a footnote in Chapter 1
and discounted as semiotically trivial, only (by popular definition of the object) denoting a reversal of conventional
accenting structure of its corresponding “positive” meter, which can be easily notated with other local operators.
Using a non-integer (fractional value) would alter the denominator of the time signatures; thus, if one presents a
fractional partition in a time signature, it might be advisable to carry the fractional part to the denominator in most
cases, though this is not necessary always.
2
Note that providing a fraction as the denominator itself will simply alter the partitioning given, since
dividing by a fraction is merely multiplication by the inverse of that fraction and so carries the fractional part of the
denominator to the numerator when the fraction is simplified.

318
such as 4/4, such a time signature is sometimes needlessly complicated, since 4/4 is more simply

notated as 2/2.3 Thus, let us go ahead and establish a suggestion:

Suggestion 1: Any pan-rational time signature used to the ends of pan-rationalism (rather

than conventional ends) should be written in simplest form with all prime factors to a

power of 1 unless there is a clearly good reason to do otherwise. For example, the
4 2
imperfect-regular time signature 4/12 ((22 )(31)) is better written as 2/6 ((21 )(31)), since the

redundant prime factor of two provides no essential information not already given in the

simplified form. Contrastingly, the perfect-regular time signature 2/9 is advisable, since it
𝑘
cannot be simplified, though it is related to the triadic hierarchy 3.

Since we allow standard note elements to pivot their value depending on the time signature

global operator, no new note elements are required when using a perfect-regular time signature.

Simply, what is a 4th-note under k/2 would be a 9th-note under k/3 or a 25th-note under k/5. This

single rhythmic tree is given below in Figure 6-1.

𝑘 𝑝 𝑝2 𝑝3 𝑝4
𝑝

𝑘
Figure 6-1. Time signature tree for signatures in the form of , where k is some integer
𝑝
partitioning of the FTU, and 𝑝 is the only prime factor of the FTU where 2 ≤ 𝑝 .

3
Note that since we have redefined the time signature to not express “what note gets the beat,” but rather
“how many equal subdivisions of the FTU are given and into how many equal parts do each successive subdivision
contain,” there is no difference between 4/4 and 2/2, since any subdivision into 4 equal parts will contain within
itself the compatibility of dyadic combination into two equal parts. No longer does the time signature indicate what
note value in a measure gets a tactus; such a musical dimension as tactus can easily enough be determined by
convention or compositional demarcation via beaming, accenting, slurring, etc..

319
Thus, by merely changing the time signature, the implied value of these note-shapes can change,

though the FTU remains constant (barring some change in tempo). This appears similar to

Creston’s and Adès use of non-dyadic time signatures; however, this use supplies all further

levels of subdivision beyond the first, creating complete hierarchies based on the prime number

in the denominator of the time signature. Of course, as noted previously, any subdivision value

outside of the prescribed hierarchy can be indicated by use of a traditional tuplet (though

aggregate completion of that tuplet is required as typical). See Example 6-1 for three variations

of a melody written using dyadic-, triadic-, and pentadic-rational time (note the use of tuplets in

the examples).

While these time signatures may be used independently, it is perhaps more musically

interesting to employ them together, since the above melodies are relatively easy to accomplish

using tuplets. Example 6-2a takes the three variations and joins them, measure by measure; this

feat, as shown in Example 6-2b/c, is rather cumbersome to execute via standard notation, and

perhaps impossible given more complex musical requirements.

Example 6-1. Melodic variations in dyadic-, triadic-, and pentadic-rational time.

320
Example 6-2. Comparative example of A) Key’s pan-rational rhythmic notation with perfect-
regular time signatures, B) by use of tempo changes with tuplets, and C) by use of tuplets
exclusively.

While the sound of 6-2a is achievable in classical notation either by use of tempo changes

with tuplets or by tuplets alone, the notation of 6-2a is remarkably cleaner. Furthermore, and

importantly for a composer, 6-2a is far more intuitive to arrive at given an expanded definition of

time signature than either 6-2b/c, neither of which are necessarily obvious solutions to realize 6-

2a. In fact, it should be noted that 6-2c is only reasonable to notate if one realizes the

simplifications achieved by writing this under a classical “compound” time signature rather than

2/2, as originally given. Notating this melody under a classical “simple” time signature would be

exorbitantly cumbersome (see Example 6-3. Furthermore, should more music follow what is

given here, 6-2c will necessarily require one to begin tieing from within quintuplets, which will

only compound the complexity of the notation as the piece moves forward.

Example 6-3. Melody from Example 6-02a/b/c given in traditional “simple” time signature.

321
While arguably any rational rhythmic structure is accomplishable under tuplets and tempo

changes (either both or exclusively one or the other), structures like Example 6-3 are

astronomically difficult for a human processor to comprehend, let alone perform. While Example

6-2a is not necessarily easy to execute, the notation makes the interpretation and translation of

this quantization far more approachable, if not executable, merely due to the fewer operators that

need to be processed in any one moment in Example 6-2a as compared to either 6-2b, c, or

especially d.

Imperfect-Regular Rational Rhythmic Hierarchies

Perhaps of even greater interest are the imperfect-regular rational rhythmic hierarchies,

which allow one to subdivide on any order by two prime factors. As demonstrated in Chapter 1,

imperfect-regular hierarchies, since they represent a subdivision tree with continuously

multiplying branching networks, use time signatures with a denominator containing two unique

𝑘
prime factors in the form 𝑝𝑞, where k is some integer and p and q are the unique prime factors

(e.g. 2/6, 5/10, 3/14, etc.).

For theoretical purposes, we can allow for a imperfectly-regular time signature to contain

𝑘 𝑘
as many unique prime factors as one might wish (i.e. 𝑝𝑞𝑟 , 𝑝𝑞𝑟𝑠 , etc.); however, if we are

considering human processors, such time signatures are highly unadvisable, since the first order

subdivision will necessarily be very small and any further subdivisions will likely be utterly

unperformable at an reasonable tempo even if the first order is performable. Even if such a time

signature was only used to give contrast to a more typical time (e.g. 7/42 contrasting 2/2), the

note given in first subdivision of the exceptional time signature would still be extraordinarily fast

and small in comparison to the more typical value or else the typical value would have to be so

322
slow that it could not be easily performed for its extraordinary length. To see the calculation of

tempo by note value for the smallest and second-smallest possible tripartite-imperfect-regular

time signature see Table 6-1 below.

Table 6-1. Tempo values by first subdivision on two smallest pqr-time signatures.
FTU 2/2 𝑘 𝑘 𝑘
Smallest 𝑝𝑞𝑟 = 30 Next Smallest 𝑝𝑞𝑟 =
p = 2, q = 3, r = 5 𝑘
42
p = 2, q = 3, r = 7

Tempo 15 bpm 4th-note = 60 bpm 30th-note = 450 bpm 42nd-note = 630 bpm
32nd-note = 480
bpm

Even if one limited their use of tripartite imperfect-regular time signatures to the smallest two

available, the differences in the basic subdivisions would be so minute, even at a relatively slow

tempo, that the difference between a typical value like the 32 nd-note and the more exceptional

30th- or 42nd-note would be difficult to accurately execute by a human, likely resulting in merely

a wash of very fast notes, one collection perhaps slightly faster or slower than the other, but

likely without precision, especially if odd groupings of these very fast notes were employed.

Thus, follows another suggestion, which the composer attempting to use this system under

human processing should consider:

Suggestion 2: The only hyper-bipartite imperfect-regular rational time signature that one

should use is that with the lowest three primes. Otherwise, the value of the first order

subdivision will be too fast to accurately execute or the relative value difference between

the pqr note and some other non-pqr note will be so slight or nuanced that calculation

323
 will be extremely difficult for a human to accurately produce. Ultimately, one should not

use any hyper-bipartite regular rational time signatures, since any such micro-rhythmic

values will likely be inexecutable by human performance (see Table 6-2).

Table 6-2. Table of first-order subdivision values of lowest six pqr-time signatures and lowest
three pqrs-time signatures.
Time Sig. Prime Factors Closest Comparable Tempo for First-
Denominator Dyadic Value Order Note Value
(First-Order Note (FTU = 15bpm)
Value)
30 p = 2, q = 3, r = 5 32nd-note 450 bpm
42 p = 2, q = 3, r = 7 32nd-note 630 bpm
th
66 p = 2, q = 3, r = 11 64 -note 990 bpm
78 p = 2, q = 3, r = 13 64th-note 1170 bpm
102 p = 2, q = 3, r = 17 128th-note 1530 bpm
105 p = 3, q = 5, r = 7 256th-note 1575 bpm
th
210 p = 2, q = 3, r = 5, s = 256 -note 3150 bpm
7
330 p = 2, q = 3, r = 5, s = 256th-note 4950 bpm
11
390 p = 2, q = 3, r = 5, s = 256th-note 5850 bpm
13

Table 6-2 alone is enough to caution against the use of such time signatures for human

performance since one rapidly moves to close approximations of extreme dyadic values (i.e.

64th-, 128th-, and 256th-notes) even with very low prime factors. Note that the lowest possible

pqrs-time signature is within the realm of audible hertz at 52.5 beats per second and arguably

more a pitch than a rhythm. Thus, we will not consider pqr- or pqrs-time signatures further; these

hyper-dimensional4 rhythmic networks are purely theoretical.

4
By hyper-dimensional we mean that pqr-, pqrs-, pqrst-time signatures (and so on) represent rhythmic
hierarchy networks with three or more perpendicular pathways in any one moment, depending on the number of
unique prime factors. A pqr-time signature could be represented by a matrix of cubes, wherein one moves from the
FTU-cube in one corner diagonally across the cube matrix by adjacent cubes, with each adjacent cube representing
the subdivision (or combination) of the value from which one is moving according to the prime value’s dimension in
which one moves (the dimensions being p, q, r, s, and so on). Similarly, a pqrs-hierarchy could be represented by a
4th-dimensional cube (of course only abstractly) in a similar manner. However, a pq-time signature is merely a two-

324
 Thus, let us just consider imperfect-regular time signatures in the form k/pq, accepting

that p and q will necessarily need to be relatively low prime values for human processors for

similar reason as outlined above for pqr- and pqrs-time signatures. Consequently, we are dealing

with a two-dimensional rhythmic hierarchy with subdivision/combination paths in either the p or

q dimension on any order level. We can then visualize this hierarchy by a grid with p divisions

along the vertical axis and q divisions along the horizontal axis with the region between the

vertical and horizontal extremes containing rhythmic values with prime factorization

subdivisions of both p and q (see Figure 6-2).

Since the first-order subdivision of any pq-time signature is the pq-note, the “pq” region

of this chart is the primary concern. From the pq-note one may subdivide in either the q-direction

(moving right) or the p-direction (down); when one subdivides by both p and q together, one

moves diagonally (both over and down) from the note being subdivided. Herein, to allow for p-

notes, q-notes, and pq-notes, we will need to add three generative elements to our standard set:

1. the p-hierarchy is represented by traditional notes shapes with the addition of a forward

slash through the notehead.

2. the q-hierarchy is represented by traditional note shapes with the addition of a backwards

slash through the notehead.

3. pq-notes apply flags for each additional p or q subdivision with p subdivisions adding a

traditional flag and q subdivisions adding a flag with a vertical line.

dimensional matrix network of value points, with one moving through the hierarchy either in the p or q direction in
any one moment; this is clearly representable by a square. The perfect-regular rational time signatures with only one
possible branch are merely a line, which is a one-dimensional object. The trivial case of a time signature with a
denominator of 1, lacking any hierarchical network, is the zeroth-dimensional point. By this logic, we can also see
how a fraction (or time signature) with zero in the denominator is undefined, since there is no dimension less than
the zeroth-dimension by conventional spatial-dimensional definitions.

325
 𝑘 q note q2 note q3 note q4 note
𝑝𝑞

FTU

p note pq note pq2 note pq3 note pq4 note

p2 note p2q note p2q2 note p2q3 note p2q4 note

p3 note p3q note p3q2 note p3q3 note p3q4 note

p4 note p4q note p4q2 note p4q3 note p4q4 note

𝑘
Figure 6-2. Rhythmic hierarchy for signatures in the form of 𝑝𝑞 , where k is some integer
partition of the FTU, and 𝑝 and 𝑞 are the distinct prime factors of the FTU where 2 ≤ 𝑝 < 𝑞 .
Note that this hierarchy theoretically extends to infinitesimal values.

326
 While this system appears to add many new note-shapes to the musical lexicon, it

essentially adds three to five due to human processing considerations. Just as with extreme pqr-

and pqrs- time signatures, pq-notes that have high p or q exponents will necessarily be very

small rhythmic values and consequently inadvisable to use. For example, the p 4q4 note, even

with the lowest prime factors possible (p=2, q=3), will be a 1296 th-note, which is unquestionably

unperformable by humans. Thus, many of these notes are necessarily theoretical. Consequently,

we have supplied colorized regions to this chart to suggest the most advisable values to use under

any reasonable pq-time signature. The colors denote:

• GREEN: reasonable note values under performable pq-time signatures:

• WHITE: use these values with extreme caution, based on tempo, rhythmic structuring, and
other compositional parameters.5
• GRAY: Essentially theoretical notes, unadvisable to use. Any further extension of this
chart in the direction of subdivision will be gray.
• PURPLE: easily performable notes but use seldomly. If used frequently, it is recommended
simply to change the operator to the corresponding perfect-regular rational time signature
(linear hierarchy with one prime factor).
• RED: reasonably performable, but, if used, one should likely change the operator to the
corresponding perfect-regular rational time signature.

Thus, let us supply a suggestion:

Suggestion 3: except with good cause under careful consideration, only use the note

values in pq-time signatures given in the green region of Figure 6-2 (i.e. p, q, pq, pq2, pq3,

p2q, p3q, p2q2-notes)

5
The FTU is in the white region, since under some pqi-time signatures with relatively high prime factors
(e.g. 7 and 11), the FTU could be an extremely long note due to the necessary adjustments of tempo to make 7 th-
notes (p-notes), 11th-notes (q-notes), 77th-notes (pq-notes) and their subdivisions reasonably performable. Of course,
such excessive lengths might not be difficult depending on the instrument playing these long values or the
performer’s ability to circular breathe.

327
 Consequent of Suggestion 3, only three to five note shapes are new to this system (p, q,

pq2, pq3, p2q2-notes), but note that only three new generative elements are added (the forward-

slash for p-notes, the backwards-slash for q-notes, and the vertical line for pq-notes with

additional q parts), and these generative elements only apply when one uses a pq-time signature.

For the completion of the system, we also need to consider rest designations and

combinative notes greater than either the p-note, q-note, or FTU. For perfect-regular time

signatures, the rests correspond with the notes given in Figure 6-1 (traditional rests with pivoting

values); for pq-time signatures, the system of rests is given in Figure 6-3 below. Note the rule of

forward-slash and back-slash for p and q noteheads is congruous with p and q rests. For pq-rests,

the vertical line generative element is exchanged with a reversed rest-flag; similarly, just as with

pq-notes requiring a vertical line for each additional q part greater than the first q subdivision,

the pq-rest requires one reversed flag for each additional q subdivision greater than the first. The

colored regions of Figure 6-3 are congruent with Figure 6-2.

Just as one can subdivide notes, one can also multiply them. For example, within the

traditional system of rhythmic notation, one can have a whole-note, double-whole-note, and

quadruple-whole note;6 similarly, one can also append part of a rhythmic value to itself via

augmentation dots. Naturally, any combination of rhythmic values can be accomplished through

the application of ties; however, sometimes it is efficacious to avoid numerous cumbersome ties

with the use of singular note signs. To such ends, an extension of the imperfect-regular rhythmic

hierarchy is provided in Figure 6-4 below. Note that all of these values can be achieved by tying

the basic set of p-, q-, and pq-notes together or, in some cases, by the application of an

augmentation dot (which still operates by its classical definition) to some p-, q-, or pq-note.

6
There is no standard notation with any multiple greater than quadruple-whole-note, barring the addition of
half of the quadruple-whole-note via the augmentation dot.

328
 𝑞 𝑞2 𝑞3 𝑞4
𝑘
𝑝𝑞

𝑝 𝑝𝑞 𝑝𝑞2 𝑝𝑞3 𝑝𝑞4

𝑝2 𝑝2 𝑞 𝑝2 𝑞2 𝑝2 𝑞3 𝑝2 𝑞4

𝑝3 𝑝3 𝑞 𝑝3 𝑞2 𝑝3 𝑞3 𝑝3 𝑞4

𝑝4 𝑝4 𝑞 𝑝4 𝑞2 𝑝4 𝑞3 𝑝4 𝑞4

𝑘
Figure 6-3. Time signature tree of corresponding rest values for signatures in the form of ,
𝑝𝑞
where k is some integer partitioning of the FTU, and 𝑝 and 𝑞 are the distinct prime factors of the
FTU where 2 ≤ 𝑝 < 𝑞

However, as already noted, there might be instances where a singular shape is more desirable for

aesthetic or communicative purposes; thus, these extensions are available if needed. Also note

that this chart does extend theoretically in all directions from the FTU.

329
 Though these notes are inadvisable to use due to their extremely long durations and easier

notation via ties, let us briefly explain the generative process applied in these extended regions,

since the process is slightly different than that for subdivisions. First, all notes in either the

extended p-region (bottom left) or extended q-region (top right) use the slash according to the

region (p or q). Next, the generative element for both regions are progressive vertical lines

through the noteheads. These lines are not applied according to horizontal or vertical

directionality of combination, but rather collective vertical/horizontal (e.g. diagonal) movement.

Thus, for each progressive additional p-1q+1 or p+1q-1 value multiplied, a similar note-shape will

acquire a vertical line (e.g. q-note to p-1q2-note to p-2q3-note and so on). Consequently, we see a

diagonal pattern of similar note-shapes across the hierarchy, each with its variation according to

its distance from its corresponding central note-shape. Note that vertical lines only begin to be

applied outside the standard pq-region enclosed in red. Any rests in these extended regions

would follow the same pattern of slash plus progressive vertical line application based on the

region, p or q.

The region enclosed in red in Figure 6-04 is the basic pq-hierarchy region as given in

Figure 6-2. The region enclosed in black represents the aggregate of the green region of Figure

6-2 and the region of extended values which is advisable to use; all those values outside the

black region (either enclosed in red or otherwise) are not advisable. For values that lie outside

the black region, it would likely be easier to communicate such rhythmic values by use of ties

rather than these exceptional note shapes even though such values might require numerous ties.

Thus, let us supply this suggestion:

330
 p-hierarchy

p-2 note p-2q note p-2q2 note p-2q3 note p-2q4 note

p-1q-1 note p-1 note p-1q note p-1q2 note p-1q3 note p-1q4 note

FTU2

q-2 note q-1 note q note q2 note q3 note q4 note

𝒌
𝒑𝒒
q-hierarchy

FTU
(p0q0 note)

pq-2 note pq-1 note p note pq note pq2 note pq3 note pq4 note

p2q-2 note p2q-1 note p2 note p2q note p2q2 note p2q3 note p2q4 note

p3q-2 note p3q-1 note p3 note p3q note p3q2 note p3q3 note p3q4 note

p4q-2 note p4q-1 note p4 note p4q note p4q2 note p4q3 note p4q4 note

Figure 6-4. Expanded pan-rational rhythmic hierarchy tree, with combinatorial values.

331
 Suggestion 4: for multiplicative/combinative rhythmic values greater than the FTU or

outside the standard pq-hierarchy, use ties or augmentation dots unless some well-

considered reason dictates otherwise, in which case you may use one of the extended

values. The only extended values that are common enough and easy enough to read as

single note-shapes and thus reasonable to employ are the p-1q-1 (FTU2), p-1, q-1, pq-1, and

p-1q note.

Consequent of Suggestion 4, confining oneself to only the black-enclosed region of Figure 6-4

requires no additional elements in the system, since the combinative values therein do not require

the vertical line.

Now, let us consider an example of an actual imperfect-regular time signature and its

two-dimensional hierarchy as notated in our grid and in proportional form. Let us take k/15 as

our example; since partitioning has no effect on the calculation of rhythmic values, we need not

consider any particular partitioning. First, we determine that our two primes factor are p = 3 and

q = 5, giving us a pentadectadic hierarchy, which we can abbreviate as “pentadecaplex.” The grid

in Figure 6-5 provides us with the precise numerical value of each subdivision according to the

pq-product in each order. Note that we have not supplied the calculation of rhythmic subdivision

value for notes in the white or gray regions, since they are excessively small and essentially

unusable.7

Given pentadecaplex, one can easily see how notes beyond the green region are highly

inadvisable; even notes on the smaller end of the green region are already unlikely performable

with these slightly larger prime factors. Thus, we offer another suggestion:

7
e.g. the p4q4 note is 50,625 equal divisions of the FTU, which under any typical tempo, let us say FTU =
15 bpm, is a pitch rather than a rhythm: specifically 843.75 Hz.

332
 Suggestion 5: It should be noted that the larger the prime factors one chooses to use in

their pq-time signature, the necessarily smaller the green region of the pq-hierarchy grid

becomes, unless one proportionally adjusts tempo. However, even adjustments of tempo

have their performable and perceptible limitations.

5th note 25th note 125th note 625th note

𝑘
15

3rd note 15th note 75th note 375th note pq4 note

9th note 45th note 225th note p2q3 note p2q4 note

27th note 135th note p3q2 note p3q3 note p3q4 note

81st note p4q note p4q2 note p4q3 note p4q4 note

Figure 6-5. Rhythmic hierarchy for pentadectadic time signature (p = 3 and q = 5)

333
 Let us consider one final abstract example with proportional notation appended to our

grid hierarchy. Let us use the dectadic time signature k/10 with p = 2 and q = 5. This imperfect-

regular rational time signature will give the hierarchy in Figure 6-6 below.

q note q2 note q3 note q4 note

𝑘 5th note 25th note 125th note 625th note
10

p note pq note pq2 note pq3 note pq4 note

2nd (half) note 10th note 50th note 250th note

p2 note p2q note p2q2 note p2q3 note p2q4 note

th
4 (quarter) note 20th note 100th note

p3 note p3q note p3q2 note p3q3 note p3q4 note

8th note 40th note

p4 note p4q note p4q2 note p4q3 note p4q4 note

16th note

Figure 6-6. Rhythmic hierarchy for dectadic time signature (p = 2 and q = 5)

Now, for the sake of notational interest, let us observe the hierarchy in linear, proportional music

notation, as given in Figure 6-7 below.

334
Figure 6-7. Dectadic Rational Rhythmic Hierarchy (“Decaplex”)

335
Figure 6-07 (continued).

336
There are two observations one should take from Figure 6-7. First, the method by which to notate

pq-notes with fractional q-parts greater than the first q-subdivision (i.e. notes with a vertical line

through one or more of their appended flags) when they appear in concatenation under beaming.

Second, the rapid proliferation of extreme fractional values that appear under such

multidimensional hierarchies by the third and fourth order subdivision, even under relatively

fundamental rational times like decaplex (the second smallest product of two primes possible).

As suggested previously, such values should be avoided or only exceptionally used; one should

probably limit themselves to the first system of Figure 6-7 (i.e. first and second order

subdivisions).

Applications of the Pan-Rational Rhythmic Notation System

Let us now consider some practical applications and examples of regular-rational

rhythmic hierarchies. Ultimately, the use of pan-rational rhythmic notation adeptly facilitates two

kinds of rhythmic notation: 1) an extension of a rhythmic practice already available, though

limited, and 2) a practice unique to pan-rationalism. The first of these notations is metric

modulation (discussed in Chapter 4); the second is the direct notation of pan-rational rhythmic

structures, particularly those that intermingle across different rhythmic hierarchies consecutively

and concurrently. Let us first examine metric modulation under pan-rationalism.

Since pan-rationalism works essentially to facilitate proportional pivoting of rhythmic

values, the first application of pan-rational time signatures to music as presently written is to

more adeptly notate particular kinds of metric modulations. While most metric modulations are

perfectly accomplishable under traditional metric modulation conventions, we will see examples

below where it is arguably clearer to use pan-rational time signatures.

337
 Metric modulation happens either through mensural modulation or partitioned

modulation. While this practice is present rudimentarily throughout Western music history, the

wide and exceptional use of this rhythmic type is particularly reliant upon the work of Elliott

Carter (see Chapter 4), though metric modulation has migrated into relatively common practice

in the late 20th and early 21st century.

It should be noted that metric modulations, either mensural or partitioned, come in two

types: prepared or unprepared. Prepared metric modulations, which are overwhelmingly the most

common, establish the tactus, in some fashion (either via multi-order partitioning or mensural

borrowing), of the metric to which one is modulating prior to the modulation itself; in essence,

preparation creates some tactus upon which we may metrically pivot. Unprepared metric

modulations, which are less common, simply change our rhythmic hierarchy without any given

pivot; arguably, these are not technically “modulatory,” since there is no given link between the

two meters and the shift is made perhaps arbitrarily. Herein, we will primarily consider prepared

metric modulations.

Mensural modulation is where the overriding meter is modified by requiring a foreign

subdivision within some order of the current FTU become the de facto subdivision in the same

order of some different FTU. The order in which this new subdivision is established is called the

order of mensural pivot. This modulation often requires the new FTU be partitioned via the

partitioning of the order of mensural pivot. Demonstration 6.1 gives a classic example of this

process.

338
Demonstration 6.1

Let us first consider a mensural modulation via a triplex subdivision in the third order.

Consider Example 6-8 below.

Figure 6-8. Imperfect Triplex Mensural Modulation via a third order pivot.

Herein, we begin with a fully dyadic rhythmic tree, but by the end of measure two we have

introduced a foreign subdivision into the third order; this new subdivision becomes the de facto

third order in a new rhythmic tree, here a new dyadic tree via the partitioning of the implied

triadic tree from the end of measure two.

If the modulation was a perfect mensural modulation, then measure three would begin

with a new time signature that would maintain a triplex subdivision in the third order, either by

convention with a dyadic-rational time signature, k/p, where k is a multiple of 3 and p is a power

of 2, likely 23 (e.g. 6/8 or 9/8), or by a non-dyadic rational time signature, k/pq, where p = 2 and

q = 3. See Figure 6-9 below.1 However, measure three maintains the same dyadic time signature

from before with no implicit triplex division in any order. Thus, we say this is an imperfect

mensural modulation, since we are in essence (as implied by the notated beat groupings)

partitioning each order subdivision by 2/3’s (or 4/3’s), rather than maintaining a triplex tactus.

1
Though possible to notate in this fashion, I do not recommend resorting to 6.10’s methodology when
conventional notation accomplishes the same ends easily.

339
Figure 6-9. Perfect Triplex Mensural Modulation via a third order pivot.

An additional example of perfect mensural modulation via third order pentaplex can be seen in

Figure 6-10; here is an example wherein a non-dyadic rational time signature might have some

notational benefits over conventional methods given not only the more intuitive grouping of

rhythms but also the lack of obfuscating ties on pentadic values.

Figure 6-10. Perfect Pentaplex Mensural Modulation via a third order pivot.

Partitioned modulation is where the overriding meter is modified by partitioning some

subdivision order of the FTU (addition/subtraction) and then requiring that this newly partitioned

subdivision become a (typically) un-partitioned subdivision of some different FTU.

Demonstration 6.02

For example, let us consider a 2/2 time signature with third-order subdivision of the 8th

note. By convention or by recognizing that 2/2 is a regularly-perfect, dyadically rational

340
rhythmic hierarchy, we know that the fourth-order subdivision will require duplex subdivision of

the 8th note into two 16th notes. We will now assume that we wish to add one 4th-order

subdivision to one 3rd-order subdivision (8th+16th note). This operation gives us the classic

“dotted” note in the 3rd-order. In partitioned modulation, we would use this “dotted” 3rd-order

subdivision and make it the new 3rd-order subdivision of some new FTU. Classically, this could

be expressed either way given below in Figure 6-11.

Figure 6-11. Partitioned Modulation by partitioned addition to the 3rd-order (equally, subtraction
from the 2nd-order).

Herein, the FTU of the first measure in 2/2 is 15 bpm; however, the FTU in the second measure

changes to 10 bpm for the first version and does not change for the second (15 bpm). This

change in the FTU is the first difference in these two different methods of partitioned

modulation. Second, it is important to recognize the differences in the implied subdivisions

(without employing any tuplets). Since we are essentially modulating by dotted-eighth note, it

might be important to either maintain the triplex subdivision of it (3 16 th-notes) or force a new

and previously inexplicit subdivision of it into two parts. The first method naturally implies a

new duplex subdivision of the quarter note value, while the second preserves the triplex

subdivision implicit in the dotted note while also still providing for a duplex subdivision without

the use of tuplets. However, note that to use any triplex subdivision in the first method, one must

then employ triplets. The method a composer might choose will totally depend on their further

intentions.

341
 Alternatively, there is a third method of partitioned modulation, which uses non-dyadic

rational time signatures, which will preserve both duplex and triplex subdivisions while also

fluidly allowing for their intermixing (see Figure 6-12). Here, the dotted-eighth note from the

first measure becomes the 1/12th note (p2q note) of the second measure. Equally though less

obviously, the quarter note of the first measure becomes the 1/9th note (q2 note) of the second

measure. The FTU changes from 15 bpm to 6.67 bpm through this non-dyadic rational metric

modulation. This method preserves both triplex and duplex subdivision in all orders as well as

the provides hexaplex subdivision through the pq-subdivision path. Granted, this method in this

simple case might not be the optimal choice for the composer; however, recognizing this

possibility might allow for a better method in more complex cases, which we will examine later

in this chapter. For a complete comparative chart of these three methodologies, see Figure 6-13

below.

Figure 6-12 Partitioned Modulation by partitioned addition to the 3rd-order via non-dyadic
rational time signature.

Demonstration 6.03

Let us consider a further yet related example with partitioned subtraction. Again, we are

given a 2/2 time signature with 1st-order subdivision of the 2nd (half) note. By convention or by

recognizing that 2/2 is a regularly perfect, dyadically rational rhythmic hierarchy, we know that

the 2nd-order subdivision will require duplex subdivision of the 4th (quarter) note into two 8th

notes (3rd order). We will now assume that we wish to subtract one 3rd-order subdivision from

one 1st-order subdivision (2nd-8th note). This operation gives us the classic “dotted” quarter note

342
in the 2nd-order. In partitioned modulation, we would use this “dotted” 2nd-order subdivision and

make it the new 2nd-order subdivision of some new FTU. Classically, this could be expressed

either way given below in Figure 6-14.

Figure 6-13. Partitioned Modulation methodologies with non-dyadic rational rhythmic hierarchy
for k/6 time signatures.

343
Figure 6-14. Partitioned Modulation by partitioned subtraction to the 1st order (equally, addition
to the 2nd order).

Here, we have the same case of implied duplex and triplex subdivision as above in Figure

6-13: the first method preserves the implied duplex subdivision from 2/2, while the second

method transforms the implied subdivision into triplex, albeit with the easy possibility of a

duplex alternative. The FTUs transform in both by the same ratio as they did in Figure 6-9.

Similarly, we could also accomplish this same partitioned modulation by means of a non-dyadic

rational time signature as in Figure 6-15 below. Here, the dotted quarter from the second

measure becomes the 6th note (pq-note) in the third measure. This time signature allows for both

the triplex and duplex subdivision. Again, this method in this simple case might not be the

optimal choice for the composer but is illustrative here. The hierarchical tree will be the same as

in Figure 6-13.

Figure 6-15. Partitioned Modulation by partitioned subtraction to the 1st order via non-dyadic
rational time signature.

Demonstration 6.04

As a final example of partitioned modulation, let us consider a slightly more complex

example, which requires addition from both the 3rd and 4th order (8th and 16th note respectively)

to the 2nd order (quarter note) subdivision in 3/4 time. This modulation will establish the double-

344
Figure 6-16. Partitioned Modulation by partitioned addition to the 2nd order (equally, subtraction
to the 1st order).

dotted quarter in 3/4 as the new beat within a new time signature. Above in Figure 6-16, you can

see the two classical methods to accomplish this end.

The first method reestablishes a heptaplex subdivision as a new beat with only duplex

subdivision; to recall heptaplex beat groupings, we require a tuple. The second method preserves

heptaplex subdivision while also allowing for the duplex subdivision by the continued use of

double-dotting. In the first case, the FTU change from 25 bpm to 14.286, while in the second

case the FTU remains 25 bpm.

Similarly, we could also accomplish this partitioned modulation by means of a non-

dyadic rational time signature as given in Figure 6-17. Here, the double-dotted quarter from the

third measure becomes the 1/14th note (pq-note) in the third measure. Here, the FTU changes

from 25 bpm to 4.08 bpm. This new time signature allows for both heptaplex and duplex

subdivision. Note that the notated metric modulation written above the staff is essentially

unnecessary given the time signature operator; we only supply it here for clarity.

Figure 6-17. Partitioned modulation by partitioned addition to the 2nd order via non-dyadic
rational time signature.

345
 Practical Examples of Pan-Rationalism2

Examples of such metric modulation via pan-rational rhythmic operators can be seen in a

handful of my own works composed early in the stages of this notation’s development, at which

time Creston’s and Adès’ use of “irrational time signatures” heavily influenced the

implementation of such signs in my own work. Consequently, these operators were not used

precisely as outlined above; however, they function as brief pivots to rhythmic hierarchies

outside the overriding meter by calling upon a common or present non-dyadic subdivision in the

work and presenting some partition of it apart from a tuplet. Consequently, they will function

well here to demonstrate practical examples of pan-rational operators functioning as metric

modulators. We can see this first in my Viola Sonata No. 1, Ceol Mor,3 wherein decaplex and

hexaplex shift the metric hierarchy continuously outside of the overriding dyadic hierarchy, the

total rhythmic effect attempting to imitate traditional un-notated rhythmic structures found in

classical Scottish bagpipe music, piobaireachd or “ceol mor.”

Herein we can see the basic use of these time signatures to execute metric modulations,

both prepared (Example 6-6 and 6-7) and unprepared (Example 6-4 and 6-5), albeit the slightly

inaccurate use according to our definitions and suggestion from above. For example, 6-4’s use of

7/12 time signature is confusing since 12 has three prime factors with only two that are unique: 3

and 2. The redundancy of a prime factor is unnecessary and perhaps confusing when creating a

hierarchy; thus, this operator is slightly inefficacious. Though, considered purely by definition,

we are told that the whole-note is subdivided into 12 equal parts (essentially four traditional

2
Note that all works by Jordan Alexander Key referenced in this chapter can be found in their complete
form in Appendix C.
3
Award recipient of the American Viola Society Maurice Gardner Competition for Composers 2020. For
the complete work, see Appendix C of selected pan-rational works by Jordan Alexander Key.

346
triplets), with only 7 of those 12 parts sounding under partition. Thus, this time signature simply

indicates 2 and one-third triplets occupy this measure, which is well communicated by basic

definition of the pan-rational time signature; however, when one considers the given pq-

hierarchy chart above, the pq-note (12th-note) should appear like the quarter-note, but in this

example the 12th-note is actually that with a single flag while the flagless note is the 6 th-note.

More correctly, this notation should appear as hexaplex with a 3.5 partition (3.5/6 or 7/12); with

this simple change the notation is correct by our established definitions. This same explanation

and problem holds true for Example 6-5; however, the non-dyadic time signatures in 6.03b

should appear as 2.5/6 and 3.5/6. Examples 6-6 and 6-7 present prepared pentaplex metric

modulations with the correct two-factor time signatures. Both examples demonstrate the same

problem as Examples 6-4 and 6-5, presenting the p2q-note value as the pq-note. This could be

simply fixed by changing the partition of the measure by half (4.5/10 and 3.5/10 respectively).

Example 6-4. Key’s Viola Sonata No. 1, Ceol Mor, mm. 8-11 showing hexaplex.4

Example 6-5. Key’s Viola Sonata No. 1, Ceol Mor, mm. 26-31 showing hexaplex.

4
Note that the variation in note head size indicates open strings on the viola and is consequently irrelevant
to our present discussion.

347
Example 6-6. Key’s Viola Sonata No. 1, Ceol Mor, mm. 16-19 showing decaplex.

Example 6-7. Key’s Viola Sonata No. 1, Ceol Mor, mm. 70-76 showing decaplex.

Viola Sonata No. 1, beside its prototypical notational inconsistencies, presents an

additional composition problem: while the composer is freed from the tuplet, there is still an

obligation to segregate rhythmic hierarchies by the bar line. It is impossible under this early

version of the pan-rational system as given by Creston, to fluidly intermingle pan-rational

rhythmic values without bar lines, which might obfuscate essential compositional rhythmic

348
constructs. A rhythmic structure as simple as a 5th-note followed by a 3rd-note cannot happen

without the presentation of two distinct measures, which disallows the clear presentation of pan-

rational rhythmic motives. This problem from Viola Sonata No. 1 first led to establishing

branching hierarchies of at least two values via prime factorization of time signature

denominators. The first work to present this possibility was Key’s Toccata ad Sancta Caecilia

for solo organ, written as a pedagogical piece for pan-rationalism, specifically tetradecaplex.

However, Toccata immediately presents the second major problem not yet realized in the

formulation of pan-rationalism: obligation of dyadic subdivision after first order subdivision. In

Examples 6-8, 6-9, and 6-10, one can see the prototype of p and q notes under pq-time

signatures. Here, note shapes, as somewhat suggested by Henry Cowell, were used, albeit with

only two sets of shapes: normal for p-notes and diamond for q-notes, which we should note are

only dyadically divisible (see Examples 6-9/10 for the 14th-note in diamond shape with flags).

What Toccata does effectively achieve is the ability to present two hierarchies of notes in

one measure so as to better develop rhythmic groupings and essential compositional motives as

well as present pan-rationalism in an accessible and performable context, giving these pan-

rational rhythms in repetitive, recognizable, and performable linear packets. While arguably any

complex rhythmic structure can be devised under pan-rationalism, this piece works towards a

performable, “classical” couching of this extension to an essentially “classical” system.

In Key’s Octet: Threnody on the Death of Children we can see the first use of pan-

rationalism consistent with our definitions found herein. Examples 6-11, 6-12, and 6-13 presents

three correct implementations of partitioned decaplex, which is interestingly set against a

septuplet. Nothing further need be said on the use of decaplex, since it follows definitions found

349
Example 6-8. Key’s Toccata ad Sancta Caecilia for solo organ, mm. 1-9 showing tetradecaplex.

Example 6-9. Key’s Toccata ad Sancta Caecilia for solo organ, mm. 10-25 showing
tetradecaplex.

350
Example 6-10. Key’s Toccata ad Sancta Caecilia for solo organ, mm. 62-72 showing
tetradecaplex.

herein; however, it is worth noting that the septuplet in the vibraphone could be rendered using a

pqr-time signature with a common multiple of 2(5)(7) = 70. Here, it is obvious that a time

signature with a denominator of 70 would be quite obfuscating if not intimidating, let alone the

additional note-shapes that would be required to distinguish r-notes from p-notes and q-notes.

Thus, I have opted to notate the seven equal subdivisions in the time of these 6 equal

subdivisions of 3/5ths of the FTU by a tuplet. Again, it is worth remembering that this system

does not attempt to remove tuplets from the musical lexicon, for they are still often quite

efficacious compared to the pan-rational alternative.

351
Example 6-11. Key’s Octet, page 10, demonstrating the use of partitioned decaplex.

Example 6-12. Key’s Octet, page 14, demonstrating the use of partitioned decaplex.

352
Example 6-13. Key’s Octet, page 24, demonstrating two uses of partitioned decaplex.

Example 6-13 presents a complete continuous metric modulation via decaplex. While

Example 6-11 and 6-12 only metrically modulates for the measure enclosed by the decaplex

operator, Example 6-13 (by means of an explicit metrical modulation) designates the 10 th-note

from measure 98 as the 4th-note in measure 99 and following. This particular metric modulation

is unique in that it presents an incomplete aggregate of a traditional tuplet (dectuplet) as the pivot

value for the modulation. Any such metric modulation by an incomplete tuplet was and is

unavailable in any metric modulation in traditional dyadic notation.

Example 6-145 presents the most continuous use of pan-rationalism in Octet. In this

section, nearly the entire green region of the hexaplex grid is employed. Note the additional use

5
We will return to discuss the first three measures of this excerpt (and the measures previous to it) in a later
section of this chapter on irregular mensurations.

353
of septuplets and quintuplets within this hierarchy. Since values divisible by two or three are the

most common in this section, hexaplex is the most appropriate hierarchy to set as the metric

operator, even with the frequent use of septuplets and quintuplets. Ultimately, this entire section

could be notated using tuplets under a classical 3/4 time signature, but for purposes of

demonstrating the execution of the system of pan-rational notation, this section has been notated

with hexaplex.

It is worth considering the utility of pan-rationalism when a rhythmic structure can be just

as easily notated using classical rhythmic devices. The section given in Example 6-14 is, in fact,

not how the score notated this music originally; I do not think this the most optimal method of

notation, since classical notation easily achieves the same ends. I have only provided Example 6-

14 as a demonstration. There are endless examples of complex rhythmic notation, which are not

only better notated via classical notation but perhaps exorbitantly cumbersome (if not nearly

impossible) to notate under pan-rationalism (we will broach this topic in greater depth later in

this chapter). For example, Example 6-15 presents a highly stratified polyrhythmic section of

Octet, which could be notated using some pan-rational rhythmic operator, but to do so would be

even more cumbersome than Example 6-14 due to the regular exchange of strata between the

instruments, the various hierarchies simultaneously employed, and the extensive subdivision

within each hierarchy. This leads us to a suggestion:

Suggestion 6: Generally, music that has successfully been notated using classical

notation, is likely not in need of translation to pan-rational notation, since very little is

gained through such a translation if the original is reasonably comprehensible.

354
 .

Example 6-14. Key’s Octet, page 38-40, demonstrating extensive use of partitioned hexaplex

355
Example 6-14 (continued).

Example 6-15. Key’s Octet, page 30, example of pan-rationalism w/o pan-rational signatures.

356
Example 6-15 (continued).

Generally, there are three alternatives available to execute traditional hyper-rhythmic music,

each of which must be carefully considered by the composer in regard to which is the most

optimal form of communication. These three methods are:

1. Proportional tuplet notation (like the hyper-rhythmists)

2. Proportional tempo notation (like Nancarrow)
3. Pan-rationalism

To demonstrate these three potential possibilities, I have provided three notations of the first

movement of my hyper-rhythmic work Bicinia in Appendix C. Below I have provided an excerpt

of these three versions to demonstrate the translation potential between these three

methodologies. A brief glance at each version of this piece is enough to see the benefits and

problems with each realization.

357
 First, the tuplet realization (Example 6-16) is beneficial in that it clearly communicates

the proportional subdivisions on every level and demarcates the relationship between the

polyphonic voices via the discrete tuplet operators. However, such use of tuplets and embedded

tuplets presents the most operators in any one moment, which can possibly overload a processor,

especially one not accustomed to complex tuplet notation.

Second, the proportional tempo realization (Example 6-17) is beneficial in that it does not

require excessive operators, substituting the many local operators for the singular global operator

of the tempo. This can aid a processor not comfortable or familiar with complex tuplet notation,

presenting the music as apparently rudimentary rhythmic values in each voice independently.

However, this proportional tempo version obfuscates the discrete relationship between the voices

in any one moment, necessitating the processor to maintain a consistent tempo throughout

without any reference to polyrhythmic hierarchical structure between the voices. This

obfuscation, however, can be mediated by the use of irregular bar lines according to each voice’s

rhythmic phrasing according to its time signature (traditional in this example) and tempo. Such

irregular placement of bar lines is a substitute for the discrete delineation of the tuplet bracket, a

substitution which is perhaps less cumbersome than the tuplet operator itself.

Third, the pan-rational realization (Example 6-18) is beneficial in that it does not require

the performer to have an accurate library of tempos, which can be difficult since tempo is

intrinsically a category of scientific time (dependent on the measurement of seconds and

minutes) rather than experiential time (time as subjectively experience by humans). 6 While not

requiring the regular coordination of FTU values between polyphonic voices, pan-rationalism is

well-organized according to FTUs; consequently, the pan-rational realization restructures its

6
See Chapter 1 for the discussion of different categories of time.

358
Example 6-16. Key’s Bicinia, first movement, second-third page, proportional tuplet realization

359
Example 6-17. Key’s Bicinia, first movement, second-third page, proportional tempos.

360
Example 6-18. Key’s Bicinia, first movement, second-third page, pan-rational realization.

361
metric structure to accord to regular FTUs rather than double-whole-notes, dotted-whole-notes,

half-notes, or other irregular values. Thus, processors can more easily coordinate the common

subdivisions and multiples of their rhythmic hierarchies according to the regulated pulse of

FTUs. Pan-rationalism also has the same benefit as the proportional tempo realization in that it

requires no tuplets beyond purely irregular values outside the operating hierarchy. Thus, the pan-

rational realization has the fewest operators, though it does have slightly more elements,

occasionally calling for p- or q- note values within a pq-hierarchy, as well as pq-notes with

additional q-components (flags with vertical lines). Thus, while one perhaps gains efficacy in the

loss of many operators and measurement by scientific time, one loses in the addition of a few

more elements (i.e. note-shapes), though these note shapes are generally few.

Thus, there are ultimately three methods of hyper-rhythmic notation, each presenting

benefits and problems, each of which will affect a processor’s translation and interpretation of

the quantization. In composing under such aesthetics, the composer should be aware of these

boons and limitations.

The fullest realization of pan-rational rhythmic notation’s potential yet in Key’s

repertoire is in Rithmomachia: Icosihexaplex for solo violin, the complete score of which can be

found in Appendix C. This work, as its name suggests, uses the imperfect-regular time signature

k/26 with prime factors 2 and 13. Herein, we find rhythmic structures which are essentially

unavailable in dyadic-rational hyper-rhythmicity without the use of unimaginably extreme

tuplets, far worse than those found in Example 6.3. Furthermore, this work attempts to structure

the intermingling dyadic- and tridectadic- hierarchy under icosihexaplex in such a way as to

create recognizable, uniquely pan-rational rhythmic motives that can be compositionally

transformed under traditional developmental stratagems. It also attempts to teach the performer

362
this particular pan-rational hierarchy – icosihexaplex – via a progressive migration away from

complete aggregates of 4th-notes and 13th-notes to various incomplete sets and juxtapositions of

smaller collections of dyadic- and tridectadic-notes as distinct rhythmic ideas.

Example 6-19 shows the opening of the work, wherein the complete aggregates of each

hierarchy in icosihexaplex are presented, orienting the performer to the basic value of each

subdivision. By measure 26 in Example 6-20 the performer encounters the first distinct pan-

rational rhythmic idea that progressively transforms throughout the work: two 4th-notes plus four

or two 13th-notes. By measure 40, the 13th-note component of this rhythmic motive has slowly

been subtracted to give the pan-rational rhythmic motive of two 4th-notes plus one 13th-note.

Measure 46 launches the piece into its next developmental section. It is worth stopping in this

measure to note the use of numerous ties to join seven 13th-notes together. Here is a practical

example of a place where the extended chart of pan-rational rhythmic values could be useful (see

Figure 6-21. Under the standard set of pan-rational note values, a grouping of seven 13th-notes is

unavailable without the use of ties; however, to accomplish a 7/13ths-note, we could simply move

up the q-hierarchy, combining q-notes by p-values (joining 13th-notes by pairs). In so doing, we

2^2
can derive a p-2q note (i.e. 13
= 4/13th-note). Furthermore, by dotting this value, we add to it half

of its value (i.e. an additional 2/13th-note), creating a 6/13th-note; if we then additionally apply a

second dot (i.e double-dot), we add half of that first additional half to the total (i.e. one half of

2/13th-note or 1/13th-note), giving us the total of 4/13th-note plus 2/13th-note plus 1/13th-note,

which is a 7/13th-note. This would be represented by with two dots. However, the ties are

perfectly clear and require far less translation and interpretation from the processor, and so are

probably a better communication method than the obfuscating sign given above.

363
Example 6-19. Key’s Rithmomachia: Icosihexaplex, mm 8 – 16.

Example 6-20. Key’s Rithmomachia: Icosihexaplex, mm 26 – 31.

Example 6-21. Key’s Rithmomachia: Icosihexaplex, mm 40 – 46.

364
 Over the remainder of the piece, the reverse of the initial subtraction process ensues,

progressively subtracting p-valued notes from the thematic rhythmic motive until measure 122.

By measure 122 in Example 6-22, an 8th-note is presented in juxtaposition with an 11th-note as a

new and final rhythmic motive for the climax and conclusion of the work. This stark

juxtaposition of 8th-notes and 11th-notes presents the performer and listener with a unique and

indivisible rhythmic motive in close succession, a rhythmic structure wholly unavailable in

traditional rhythmic notation and relatively cumbersome to compose under Creston’s/Adès’

system, which would require a new measure for every change of pan-rational note value. Herein,

we see the full monophonic potential of pan-rationalism to not only notate interesting and unique

rhythmic structures, but also present them in a compositional and thematic fashion as would be

classically done with dyadic-rhythms, intermingling any partition of any pan-rational subdivision

with any other.

In the conclusion to this chapter, we will discuss at greater length some of the general

problems and benefits of this notation. However, before our final conclusion, we should take a

moment to discuss irregular (indivisible) rational time signatures and mensurations in the context

of pan-rational rhythmic notation as well as irrational (incommensurate) mensurations.

Example 6-22. Key’s Rithmomachia: Icosihexaplex, mm 122 – 128.

365
 Notation of Indivisible and Irrational Mensurations Under Pan-Rationalism

Irregular (Indivisible) Rational Time Signatures/Mensurations

As outlined in Chapter 1, irregular time signatures do not express any patterned

hierarchy, since they are by definition irregularly subdivisible/combinable. Consequently, there

is no “time signature” by our definition herein that reasonably encapsulates irregular hierarchies.

Given indivisible subdivisions of some FTU, there are certainly common multiples of such

indivisible subdivisions that could allow for a time signature (especially of the hyper-pq

dimensional variety); however, such time signatures would both require extremely large

denominators and use of the exceptional extended rhythmic values as found in Figure 6-4, since

to notate the various indivisible values one would necessarily need to move upwards via

combination from the extremely small first order subdivision into the extended regions of the

hyperdimensional matrix. This is extremely unadvisable by prior suggestions. Thus, we offer a

suggestion:

Suggestion 7: It would be far better to notate an essentially irregular hierarchy via an

irregular mensuration according to each level compositionally present in a piece rather

than devise some obtuse regular-rational time signature via a large common multiple.

Thus, one should notate non-hierarchical concurrently indivisible rhythmic strata via a

mensuration or classical tuplets rather than a single pan-rational time signature. To this end, there

are a few additional considerations to account for.

First, a mensuration should only be used for three or more indivisible rhythmic strata. If

one is only using two indivisible rhythmic strata, then a regular-rational time signature is

perfectly advisable, since two indivisible values that are reasonably performable by humans will

likely have a relatively low common multiple allowing for a reasonable pq-time signature

366
prescription. Otherwise, two indivisible rhythmic strata are also easily demarcated by use of

tuplets. Consequently, a mensuration, if one desires to use such a global operator rather than the

local operators of the tuplet, should list at least three distinct subdivisions of the FTU as its

operator to make the use of a mensuration at all efficacious. The only case where a bi-partite

mensuration is reasonable is when the rhythmic strata are incommensurate; we will discuss such

situations in the next section.

Second, mensurations, as defined in Chapter 1, cannot be partitioned and thus must

present the aggregate of its expressed indivisible subdivisions. If one wishes to express partitions

of some set of indivisible stratified polyrhythms, then one must use some regular-rational time

signature that allows for partitioning. Such an exceptional use of mensuration can be

accomplished by use of mensuration and regular-rational time signature concurrently. Similarly,

there is also the exceptional circumstance where one wants to express regular non-dyadic

subdivisions within irregular rhythmic strata under a mensuration. In such cases, one can of

course employ mensurations concurrently with independent regular-rational time signatures in

each voice as well. Such a music, of course, will be highly complex and difficult to execute, due

to the overarching indivisibility of the rhythmic strata and the various partitioning and

divisibility of those indivisible strata. Under such a rhythmically complex music, the only

organizing force might be the FTU, but even the FTU will not hold as an organizing pulse if the

differing indivisible rhythmic strata are partitioned, since such a partitioning will necessarily be

by units of the FTU that are of differing, indivisible lengths in each stratum. Thus, such a music

should be approached with compositional and mathematical care, though it is notationally

available via concurrent use of indivisible mensuration and divisible time signature.

367
 Let us consider some practical examples of music that could use indivisible mensurations

rather than tuplets. First, we will return to Karen Khachaturian’s Trio for Strings (see Example 6-

23),7 which already is using a kind of indivisible mensuration, though couched in traditional time

signature. Here, Khachaturian has effectively used traditional time signatures to signal the

indivisible poly-rhythmic stratification in this section. This indivisible polyrhythmic stratification

can be equally notated using tuplets in every voice, as we have also seen in many works in

Chapter 4 and 5, as well as proportional simultaneous tempos. Consequently, we see the same

three methods of notation for irregular hierarchies as we saw with regular ones:

• Proportional tuplet notation (like the hyper-rhythmists)

• Proportional tempo notation (like Nancarrow)
• Layered Time Signatures or Irregular-Rational Mensuration

Either the tuplet, tempo, or layered classical time signature methods are fine methods to denote

irregular mensurations, all of which have been employed by numerous 20th century composers as

discussed in Chapters 4 and 5 (see Charles Ives, Conlon Nancarrow, and Ben Johnston for

additional examples). I have even used the method of layered classical time signatures to

accomplish a clean notation of indivisible polyrhythmic strata without the use of tuplets in his

Saxophone Quartet No. 1 (see Example 6-24).

However, in accordance with our definitions laid out in Chapter 1, we offer an alternative

method of notating irregular rational rhythmic hierarchies, which we will demonstrate using my

Octet (see Example 6-25). In this section of Octet there are four indivisible rational rhythmic

hierarchies – 3, 4, 5, and 7 – along with an additional triadic subdivision of the triadic hierarchy

(9). Consequently, one can notate this by a mensuration proportion of [Link] or (for the sake of

clarity) [Link], since the additional 9 essentially changes nothing in the interpretation of the

7
see Chapter 5 for our prior discussion of this work

368
mensuration. The remainder of the music following the mensuration will be interpreted based on

the contents of the measure (whether that voice is clearly divided into 3, 4, 5, 7, or 9 equal parts).

Since we are providing an irregular mensuration, no further levels of the hierarchy are supplied,

and thus either there are none, or one might infer a traditional de facto dyadic subdivision as

might be necessary.

Example 6-23. Khachaturian’s Trio, indivisible polyrhythmic stratification, notated using layered
classical time signatures.

Example 6-24. Key’s Saxophone Quartet No. 1, indivisible polyrhythmic stratification, notated
using layered classical time signatures.

369
Example 6-24 (continued).

Example 6-25. Key’s Octet, indivisible polyrhythm, notated using pan-rational mensuration.

370
Example 6-25 (continued).

371
 This form of pan-rational notation – irregular rational mensurations – allows Octet’s

music (and potentially any of the above-mentioned composers’ indivisible polyrhythmic music)

to forgo numerous tuplets. However, the lack of tuplets and use of a mensuration operator does

not necessarily make the notation of indivisible rhythmic strata more understandable than other

methodologies. Thus, we see how mensuration is not necessarily more useful than traditional

notation, though it is an available consequence of pan-rational rhythmic notation systems.

Ultimately, indivisible mensurations are not as utilitarian as regular-rational time

signatures, since, in most cases, they only re-notate rhythmic structures already available in

classical rhythmic notation via tuplets. Only in exceptional circumstances does indivisible

mensuration allow new rhythmic structures, though such exceptional circumstances, as

mentioned before, beget extremely rhythmically complex music demanding exceptional, perhaps

super-human processors. One such example of super-human rhythmic structures made available

via mensuration are those under concurrent implementation of partitioned regular-rational time

signatures as well as those with incommensurate rhythmic strata. Consequently, irregular

mensuration also leads us to the notation of irrational rhythmic hierarchies, which are necessarily

mensurations since no incommensurate value has equal subdivision and is consequently

irregular. Thus, let us briefly discuss the notation of irrational rhythmic hierarchies by

mensuration.

Irrational (Incommensurate) Mensurations under Pan-Rationalism

We will only briefly discuss irrational mensurations, since they are essentially

unperformable by human processors (except by approximation), and so arguably need not be

notated since such rhythmic structures are only able to be meaningfully approximated by

372
computer aid, which itself does not need a form of music notation necessarily. However, such

rhythmic ideas are a natural outgrowth of pan-rationalism and so worth mentioning,

As mentioned in Chapter 4, Conlon Nancarrow was the first composer to pioneer

exploring mathematically irrational rhythmic structures in a few of his later studies for player

piano. In these studies, Nancarrow notates “time signatures” with either one irrational or two

irrational components (e.g. 2:√2 or 𝑒: 𝜋). Calling these operators “time signatures” is a

misnomer, since they operate according to neither the classical definition of time signature nor

the pan-rational definition supplied in Chapter 1. What Nancarrow communicates through his

irrational “time signatures” is rather a tempo proportion, two voices (often each a polyphonic

complex in itself) progressing at some setoff tempos that are in the prescribed proportion – e.g.

(60 bpm)2 : (60bpm)√2 .

While the above example (as well as Nancarrow’s other irrational time signatures) is

ultimately a conceit of tempo, the result of this tempo conceit is ultimately a polyrhythmic

complex that is the continuous stratification of two incommensurate rhythmic hierarchies, one

dyadic and the other irrational by comparison but essentially dyadic on its own. By itself, a piece

that is wholly in a tempo of (60bpm)√2 is not rhythmically irrational, since all rhythmic values

would essentially be divisible by √2 and consequently commensurate (e.g. recall from Chapter 1

that while √2 is irrational 7√2: 3√2 is commensurate since the irrational values divide out; this

is essentially the case for a piece wholly in an irrational tempo). However, as soon as a musical

structure that is irrationally couched (likely via tempo) is juxtaposed to another structure that is

not couched in that same irrational value or some rational multiple of it, some irrational

relationship will emerge through this juxtaposition.

373
 Irrational rhythmic relationships can be encapsulated by indivisible mensuration since

such mensuration proportions are, in essence, tempo proportions. For example, an indivisible

mensuration proportion of [Link], dictates two equivalent interpretations:

1. Tempo Interpretation: each rhythmic stratum will use the same rhythmic values but will
progress with tempos in these relative proportions (i.e. tempo times [Link] respectively)
2. Pan-Rational Interpretation: each rhythmic stratum will use the indicated equal
subdivision of the FTU (3, 5, 7, 11 equal subdivisions).

Though seemingly different in approach, the second case is still operating under tempo

relationships; however, the tempo in the pan-rational interpretation is determined by the FTU

itself, from which all the rhythmic values are derived, rather than the derivative elements

themselves, as would be the case in the traditional tempo interpretation.

Along with the examples of irrational approximations given by Nancarrow in the mid-

20th century player piano studies, I, in my works To Say Pi and Nachi No Taki, have composed

digital music wherein computers are used to approximate irrational rhythmic structures to

varying degrees, many more closely irrational than possible for Nancarrow on his piano rolls.

In To Say Pi, I superimpose various approximations of and recursive calculating

processes for pi onto a few musical dimensions, particularly rhythm, time, and harmony. Many

rational approximations of pi, from relatively precise to the grossly oversimplified, are employed

to generate harmonic as well as rhythmic material in this piece. Some of the rational

approximations one can hear in this work include 3/1, 13/4, 16/5, 19/6, 22/7, 44/14, 333/106,

355/113. The most accurate of these - 3/1, 22/7, 333/106, 355/113 - come from the continued

fraction approximation of pi. By the end of the piece, the next two approximations in the

continued fraction algorithm, 103993/33102 and 104348/33215, create a climatic metamorphosis

from the approximate irrational rhythmic structure of pi to the harmonic realization of the

number. Of all of these above rational approximations, 355/113 is the only fraction in the

374
sequence that gives more exact digits of pi (i.e. 7) than the number of digits needed to

approximate it (i.e. 6). Consequently, this ratio is the most important in the formal structure of

the music.

Throughout the piece, the music undulates between oversimplified approximations and

more precise approximations, giving the music a sense of acceleration and deceleration. On a

macro-scale the work contains four large sections, each with its own instrumental colors: first,

metallic sounds forming a digital gamelan; second, plucked sounds forming a digital orchestra;

third, more abstract tones and waves making a digital organ; fourth, all these ensembles

combined. The development of the materials in each section (lasting approximately 2.5 minutes

each) follows a figurative circular trajectory: building from simple structures, adding more

rhythmic layers, morphing harmonies from equal temperament to “pi-temperament,” and then

returning to simplicity with the diffusion of these materials. Within each of these large formal

circles are smaller formal circles which follow similar trajectories as the larger ones. The

temporal length of both these macro and micro formal circles follows the ratios given above: the

largest formal circles designed to encompass a polyrhythmic structure of 355/113 or 333/106 and

the smaller ones designed to encompass 22/7 and various multiples of it.

Ultimately, each of these rational ratios is an approximation. How far we take this

approximation is generally determined by how precise one needs to be. The choice is somewhat

arbitrary as to how far one will take the endless process of irrational computation. The

conclusion of this piece attempts to point to this possibly infinite process and the precision which

it demands. After the final “circle” section, the coda of the piece transforms all the polyrhythmic,

pi-approximating ratios, currently present in each of the three ensembles, from the realm of beat

and pulse into the realm of pitch and harmony by processing the data to transform the rhythmic

375
ratios 3/1, 13/4, 16/5, 19/6, 22/7, 44/14, etc. into 333/106, 355/113, 103993/33102 and

104348/33215. This process creates a different effect than simply generating a pitch at these

frequencies. Rather, this process takes the material (all pitches, rhythms, and harmonies) and

essentially accelerates it to such an extent that the music becomes only perceptible as pitch with

overtones. Essentially, the “musical data,” relatively unpacked for the listener in the first 10

minutes of the piece, becomes repacked into about 30 seconds of sound at the end. This sound

mass stands as another approximation of pi but also attempts to show the computational and

auditory challenge of fully realizing this number sonically and then processing it by the human

ear. In this last 30 seconds, you hear as many notes as you did in the prior 10 minutes, amounting

to more than one million polyrhythmic, micro-tuned musical data points.

Though we can approximate an irrational value as well as we may like, a complete,

precise realization of such a rhythmic structure is ultimately theoretical; while the speculation

and approximation of such a conceit is certainly worthwhile, any such realization necessarily sits

within pan-rationalism by the necessity of rational approximation. Thus, any meaningful

approach to irrational rhythmic structures is aided by a better system to notate, encode, and/or

compositionally manipulate pan-rational rhythmic values, either on paper or in computer data

banks.

Conclusion

Charles Francis Abdy Williams in his 1903 book The Story of Notation wryly remarks

that if one were to search the shelves of the various libraries of Europe, they would probably find

that “for some centuries a new notation has appeared about every three or four years, each of

which is called by its author ‘The’ new notation, for he fondly thinks that it will become

376
universal.”8 I do not feel so optimistic as to think that this proposed addition to modern music

notation will find any use beyond my own work; as noted previously, the momentum of tradition

is often a difficult force to overcome. Indeed, this fact of history is why, rather than propose a

wholly “new” system, I have attempted to only add a tool to composers’ present toolbox, a tool

which they may freely use or not use with conventional rhythmic notation. I do believe, based on

careful study that this system is the best yet to be devised for the free diversification of

prescribed rhythm as notated for human performance. However, this system is also capable of

notating any rational (and theoretically irrational) rhythm that is well beyond human means,

consequently giving the composer a cognitive tool to semiologically organize and encode

discrete rhythmic ideas to any level of accuracy desired.

However, this system does have some problems, which is why I do not suggest that it

replace all current rhythmic elements and operators. As demonstrated, tuplets, tempo markings,

and classical time signatures, while all theoretically replaceable, still have a place due to their

efficacy under certain – particularly dyadic – circumstances; furthermore, since most music

notated past, present, and likely in the future is intrinsically dyadic, these dyadic-oriented

operators and elements will continue to be more readily useful and communicative than non-

dyadic operators and elements. Only if and when the system of pan-rationalism becomes a

standard and well-understood part of music notation broadly might such classical dyadic

operators be overturned. Until then, there is no reason to forsake many parts of notational

tradition that still function coherently and efficaciously, especially in those cases where pan-

rationalism is not necessarily better and perhaps more obfuscating than a traditional equivalent.

8
See Williams, C. F. Abdy. 1978. The story of notation.

377
 Thus, in conclusion let me briefly list some problems the composer should be aware of,

especially if they should use pan-rationalism. Furthermore, let me clearly reiterate the benefits of

pan-rationalism as an appendage to modern prescriptive notation and give some final suggestions

to composers for this system’s use and further exploration.

Problems in Key’s System of Pan-Rational Rhythmic Notation

Let us recognize some problems that arise within this system of pan-rational rhythmic

notation. These are not problems of incompleteness or inconsistency; rather, these are problems

that might hinder human interpretation and translation of the notation.

First, as noted many times already, there are rhythmic structures in music which are far

easier to notate under traditional operators and elements than with pan-rational operators. Much

of the music of 20th and early 21st century hyper-rhythmicity is quite difficult to notate using

pan-rational time signatures. Depending on the complexity of the tuplets called upon, pan-

rationalism is more or less effective. If the various rhythmic subdivisions called up by a

composer share just a few prime factors, then perhaps pan-rationalism can be employed

effectively, but, as soon as three or more prime factors of any size greater than the smallest are

used, pan-rationalism becomes very cumbersome and difficult to interpret by a human processor.

In such cases of extremely diverse (usually dyadically couched) hyper-rhythmicity, as that found

in Ferneyhough, tuplet notation is probably the clearest solution, though the composer should be

careful to find the best rationalization of their rhythmic structures via tuplets (as addressed

during our discussion of Babbitt and Ferneyhough in Chapter 5). Tuplets are merely local pan-

rational operators, and so the same care of simplification, prime factorization, and metrical

structuring must be taken under the local operators of tuplets as done with the pan-rational global

operators of time signatures. Key’s Bicinia and its three translations gives a good example of the

378
comparative notation of pan-rational tuplets and time signatures (as well as tempos), which can

stand as a model of clearly notated, calculated, and delineated hyper-rhythmicity.9

Secondarily, similarly, and consequently, tuplets remain a necessary part of rhythmic

notation, for better or worse. I do not feel confident that a mission to remove the necessity of the

tuplet, as somewhat suggested by Cowell, is sensible. The tuplet is very useful under many

circumstances. Thus, pan-rationalism does not forsake the tuplet. This might be seen by some as

a problem of the system, though I would say that it is a benefit. The composer must in some

cases use tuplets for clearer communication, but by having both the tuplet and pan-rational time

signature (as well as mensuration), the composer has the freedom of choice and may or may not

obligate themselves to the “tyranny” of the tuplet aggregate. In some sense, pan-rationalism is

bound to the tuplet not because it is necessary to the system generally, but because it is

efficacious to the system when interpreted and translated by human processors. Should we

forsake human processors, then tuplets are wholly unnecessary to the execution of a pan-rational

system of musical notation. However, since humans have been trained to read tuplets and cannot

easily parse pan-rational time signatures with many prime factors (and consequently many note

shapes), we should limit the system to what is easily processed by humans. Thus, the remainder

of problems in pan-rationalism are really an artifact of human limitations, rather than limitations

of the system itself. However, if the system is to be optimized for human interpretation and

9
It is interesting to note that in the first performance of this work, I provided the two clarinetists (Amy
Humberd & Ian McIntyre) with both the tempo and tuplet version of the score; they each individually, without any
foreknowledge, learned their own part under a different translation, each having a different notational preference. I
had thought they would opt for the tempo score, since I thought that it would appear the least intimidating.
Ultimately, Amy found that the tempo score was easier to read, and opted to use the tempo score throughout, while
Ian found that the tuplet score was easier to comprehend, especially under considerations of ensemble coordination.
The pan-rational score was not available for them to consider using at the time. It might be advisable, when
composing rhythmically complex music, to provide such alternatives as they are sensible to notate, since performers
will have differing predilections in the interpretation and translation of music notation. Both Thomas Ades and I
have found this multiple notational approach successful, though it is abundantly time consuming.

379
translation, some of the strictures of the system of pan-rationalism as given might be inoptimal,

and one might wish to devise a better solution than the ones I have given.

Third, let us consider the pan-rational time signature operator itself. While when

presented with low p, q, or pq values, time signatures are reasonably interpretable, very rapidly

these time signatures can become quite cumbersome and likely intimidating to human processors

when slightly higher prime factors are used. Consider just briefly the music in Icosihexaplex,

most of which is given in time signatures over 26 often with partitioned values that are

themselves quite large and fractional. The interpretation of these time signatures can become

fluent with practice, but upon initial study are rather overwhelming. Thus, if we are subservient

to human processing, pan-rational time rhythmic operators will naturally be slightly more

cumbersome than their traditional counterparts; however, I propose that the benefits of truly new

rhythmic ideas and avenues for uncharted rhythmic exploration significantly outweigh the

cumbersomeness of the few performable large prime factor time signatures.

Fourth, let us consider the pan-rational rhythmic elements themselves (i.e. the additional

note shapes). Of course, any additional elements will encumber a system; however, our additions

herein are relatively minimized with a comparatively abundant addition of rhythmic

diversification. Thus, the addition of a few new elements is significantly outweighed by what is

gained. However, the relatively slight difference in appearance between p- and q-notes under a

pq-time signature (i.e. the forward and backward slash through the note head) will be difficult for

human processors to read until the notation is fluently learned. This style of notation was

selected for two reasons. First, most clearly different notehead shapes (diamond, triangle, square,

etc.) have already been given meanings under classical notation (e.g. diamond noteheads often

indicate harmonics, triangle noteheads have numerous meanings in percussion music as well as

380
the frequent indication of “highest or lowest” possible pitch in other instruments, etc.); thus, lest

this system confront inconsistency when combined with other, non-rhythmic techniques (i.e.

harmonics), recognizable notehead shapes cannot be used. Second, the choice of slash marks

allows for the use of standard noteheads, allowing all rhythmic values to be derived from

traditionally rhythmic elements in a logical and systematic fashion. Furthermore, since p- and q-

notes share a common origin in pq-time (namely, the un-slashed notehead), it becomes easier to

logically combine them into pq-note values by the redaction of the opposing slash marks. Thus,

while the slight difference between forward-slash and backward-slash will probably confuse

some performers initially, once learned, the logic of the choice will likely be apparent, and any

confusion will be readily overcome.

Fifth, as seen in Icosihexaplex, under pan-rational time signatures with relatively disparate

prime factors (e.g. 2 and 11 or 3 and 13), the interpretation of their corresponding p- and q-notes

is somewhat counterintuitive, since what appears as a version of the classical “half-note” (i.e. the

void note with a stem) is actually two relatively distinct rhythmic portions of the FTU, one

relatively large (half in the case of p = 2) and one relatively small (13 th in the case of q = 13).

Thus, a q-note under a hierarchy like icosihexaplex (26) or icosidiplex (22) will seem

counterintuitively small compared to the traditional expectation of a void notehead. Of course,

the logic of the system should outweigh any cognitive predilections of human traditionalisms,

but this is a stumbling block any interpreter of pan-rationalism should be cognizant of. To

attempt to mend this interpretive pitfall for human processors, it becomes imperative that the

composer proportionally organize their music; that is, give the note values via physical spacing

in the score as close as possible to their relative temporal spacing in audiation. Icosihexaplex

does this relative physical spacing of p-, q-, and pq-notes. In examining the score in Appendix C,

381
one will see that void q-valued note shapes occupy very little space as compared to the similar,

yet rhythmically distinct, void p-valued note shapes. Thus, this problem is easily navigated when

it arises, recognizing that such interpretative issues will arguably not be present under prime

factors that are adjacent or nearly adjacent (e.g. 3 and 5), which will be the more frequent case.

The last problem is not with the system itself, but with the perception of the music written by

it. While the “perception” of musical ideas should never necessarily preclude one from

incorporating such musical structures into a piece,10 the composer should note some of the

limitations of human cognition and execution of particular pan-rational rhythmic ideas. Many of

these have already been addressed and mitigated by our suggestions offered in this chapter.

However, it is worth reiterating the issue. A complete justification of the following phenomenon

is not the purview of this document, but we will propose its reality without question: the human

mind seems to have a proclivity to approximate reality, particularly according to the limits of its

perception (inborn or trained) and the expectations culturally inculcated in it. For example, most

people do not cringe at the radical tuning of equal temperament, likely because the mind can

happily roundoff the irrational micro-part of each musical interval. Similarly, few “average”

musical listeners will cringe at a slightly out of tune piano when listening to a familiar tune in a

piano bar, rather recognizing the familiar foreground tune and discarding the background noise

of notes which are slightly removed from standard tuning. Only when tuning becomes distinctly

removed from the expected does the “average” listener notice, perhaps with a cringe, that

something is not as expected. Obviously, the threshold of such roundoffs and when slight

10
For example, just because the average listener cannot hear 12-tone rows and chromatic aggregate
completion, or just because one cannot probably perceive the precise ratio of 60:61 as presented in a player piano
study by Nancarrow does not mean that such musical structures are not worth exploring and presenting to human
listeners. The ability of one to accurately perceive something should not be the rule to determine its worth.

382
differences in phenomenon migrate from the background of perception to the foreground will

differ from person to person by training, culture, and physical makeup.

This tendency of the human mind to approximate reality, overlooking granular details to

focus on general characteristics, is relevant to the composition of pan-rational rhythms. While

some rhythms are just arguably unperformable and imperceptible in a precise manner (e.g.

telling or performing with accuracy that a rhythm is 60:61 vs 61:63), there is a gray region where

perceptibility and performability are arguable. Such an example of this grayness is given in

Icosihexaplex with the juxtaposition of triplets (i.e. 12th-notes) and 13th-notes.11 Given the tempo

of the piece, the temporal difference between a 12th-note and a 13th-note is miniscule. Many

people, if presented with the two in concatenation by groups of three (i.e. three 12 th-notes and

three 13th-notes) will likely hear two sets of triplets. Since human performers often perform with

some rubato, intentional or unintentional, the slight difference between these 12 th-notes and 13th-

notes can easily be joined by roundoff in the mind as two triplets. Furthermore, when located

remotely from each other (i.e. a set of three 12th-notes in one measure and a set of three 13th-

notes some measures away with no juxtaposition between the 12 th- and 13th-notes), most human

listers will likely hear these rhythmic structures as congruent, even if played with absolute

accuracy by a computer. When such rhythmic structures of high temporal similarity are played

by humans, who are by nature even at their best granularly inaccurate and imperfect, it Is likely

that the difference between a 12th-note and a 13th-note will not even be executed accurately, the

phenomenon of roundoff nearly as common in performance as in listening.

Thus, with this final problem, the system is not challenged, but what is composed in the

system is challenged. A composer should not limit themselves from exploring exceptional

11
Another example would be the closeness of 11th-note to dotted 16th-note, since the dotted 16th-note is a
3/32 -note which is very close to a 3/33rd-note, which is a 1/11th-note by simplification.
nd

383
musical structure due to the limitations of human perception; however, should one be interested

in human perception and accurate performance, such considerations of this “grayness” in

rhythmic distinctiveness should be carefully considered. Are the notes one is presenting

meaningfully distinct under the approximating nature of the human mind and inaccurate nature

of human performance? If possibly indistinct, one should consider whether this lack of

perceptible rhythmic discreteness as well as the potential inability to perform and perceive it are

important to their conceit or not.

Benefits to Key’s System of Pan-Rational Rhythmic Notation

Let us in summary list the benefits of the adoption of pan-rationalism to modern rhythmic

notation, should the benefits not already be obvious. First, as said, this system is an addition to,

not overthrow of, rhythmic notation as developed since the late 13 th century. Thus, it seamlessly

grafts onto classical notation without any change in original elements or operators or their

pointers, barring a slight modification of time signature definition, which we should recognize

changes nothing about the meaning of classical time signatures. Consequently, teaching this

system, even as part of the rudiments of music, rather than as some exceptional advanced

practice, is reasonable and easy, if not altogether logical and advisable since it more correctly

defines musical operators and elements, gives the young musician a diverse lexicon of rhythm,

and inculcates composers and musical culture generally to a new manner of musical thinking that

can beget prescriptive music both performable and significantly unique at present.

Pan-rationalism allows the composer to cleanly and freely notate performable pan-

rational rhythmic structures through the reallocation of pointers to new discrete temporal values.

As noted before, this has been accomplished to its fullest rational extent via the addition of the

384
fewest new operators (global and local) and elements as possible, all of which accord to the

traditional system of operators and elements and are logically derivative from them.

Pan-rational time signatures (either with single, double, or more prime factors) allow for

the unencumbered mapping of dyadic note shapes onto non-dyadic values (in the case of one

prime factor) as well as the mapping of pan-rational hierarchies (dyadic or non-dyadic) onto

other pan-rational hierarchies (in the case of two or more prime factors). Furthermore, these new

time signatures and the definitions of their constituent parts allow for one to clearly prescribe

with one number all the subdivision and groupings of a hierarchy continuously, something

unavailable under Creston’s and Adès’ prototype of this system.

Additionally, all care has been taken throughout the first chapter, this final chapter, and

all Synthesis chapters, to outline what makes a good system of notation such that this system

itself might accord to those precepts and create a notation optimized for human communication

and performance, though it certainly can extend beyond those limitations if one wish. Thus, pan-

rationalism is designed to help limit the composer to likely performable pan-rational rhythmic

structures through my definitions, suggestions of implementation, dimensional structures, and

field boundaries provided herein, though the system can allow for any imaginable rhythmic

structure up to approximations of irrational rhythmic values.

Lastly and most importantly, this system emancipates the composer from purely dyadic

thinking, freeing them to select any singular rhythmic value (i.e. apart from the tyranny of the

tuplets aggregate) and intermingle them with other singular rhythmic values of various natures to

form individualized and unique rhythmic structures under imperfectly-regular pan-rational time

signatures, no longer necessitating the segregation of differing rhythmic hierarchies by the bar

line as under Creston’s and Adès’ system. Through the freedom afforded by this refashioned and

385
extended tool of temporal prescriptive encoding, any composer is encouraged to imagine, write,

and audiate rhythmic ideas never before prescribed, no longer subconsciously bound to an

artificial tyranny of dyadicism as handed down for centuries.

The possibilities for further exploration under pan-rationalism are plentiful. Given the

variety of music composed under our dyadic system over the past half millennium and given the

relatively limited set of rhythms available therein, the variety now open to discovery is

staggering and exciting, especially when one considers that many of these new potential

rhythmic ideas are quite performable and not merely virtuosically “hyper-rhythmic.” In addition

to simple rhythmic ideas, yet to be explored further are pan-rational rhythmic macro-structures

and musical forms as presaged by Nancarrow’s Studies for Player Piano and works like Adès’

Piano Quintet, wherein pan-rational rhythmic hierarchies of differing and indivisible partitions

are superimposed, creating a complex network of hierarchical rhythmic counterpoints according

to differing FTUs. Given the utility of this system as a “compositional” tool, the composer is not

limited by human performability in what they can imagine and encode with pan-rationalism;

under the capabilities of digital music, the pan-rational organization of discrete prescriptive

rhythm is endless to the point of irrationality. The exploration of music that approximates

irrational mathematical values is worth much deeper study and execution than afforded it during

the 20th and early 21st century since Nancarrow’s provocative irrational studies for player piano.

It is genuinely surprising that more composers have not explored this interesting field of

irrational rhythmic discourse under the advent of digital music.

Irvine Arditti, violinist and one of the foremost interpreters of avant-garde music of the

20th century, notes that “Every era seems to uncover new realms of possibilities for the player,”

whether these realms be harmonic, timbral, contrapuntal, instrumental, or rhythmic. Arditti,

386
concisely summarizing the narrative of music history as a continual progress from that which

was impossible to that which is now not only possible but perhaps typical, declares that “What

was not possible earlier this century [20th], is or will be possible,” believing that “It is the

player’s responsibility to transcend traditional limitations and find new possibilities of

interpretation.”12 Here, at last, I encourage any composer who has tirelessly persisted to the end

of this tome, to take upon themselves at least a modicum of the heavy responsibility Arditti gives

to every performer: transcend traditional limitations given to us and seek, with our gracious and

assiduous performers, new possibilities of human expression in music.

12
Irvine Arditti, Questionnaire response in Complexity in Music?, 9.

387
 APPENDIX A
COMPLETE MUSICOLOGICAL EDITIONS FROM CHAPTER 3

Example 3-1. Johannes Stockem, Ave maris stella from the Materanensis Manuscript.1

1
Stockem, Johannes. (2019). Ave maris stella (Biblioteca Comunale Augusta, Perugia, Italy [I-PEc] Ms.
1013 [M.36] ff. 63v-70r, "Johannes Stochem") (Key's Historical and Performance Editions) (J. A. Key, Ed.).
Gainesville, FL: Jordan Alexander Key.

388
Example 3-01 (continued).

389
Example 3-01 (continued).

390
Example 3-01 (continued).

391
Example 3-2 John Hothby, Ora pro Nobis from the Faenza Codex.2
2
Hothby, John. (2020). Ora Pro Nobis (Biblioteca Comunale, Faenza, Italy [I-FZc] MS 117 (Faenza), ff.
26v - 27r) (Key's Historical and Performance Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.
392
Example 3-02 (continued).
393
Example 3-02 (continued).
394
Example 3-3. Adam, De tous biens playne a2 from the Segovia Manuscript.3

3
Adam. (unpublished). De tous biens playne (Archivo Capitular de la Catedral, Segovia, Spain [E-SE], Ms.
s. s. “Segovia Codex; Cancionero musical de Segovia; Cancionero del Alcázar.” fol. CCIv) (Key's Historical and
Performance Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

395
Example 3-4. Tinctoris, Le Souvenir a2 from the Segovia Manuscript4

4
Tinctoris, Johannes. (unpublished). Le Souvenir (Archivo Capitular de la Catedral, Segovia, Spain [E-SE],
Ms. s. s. “Segovia Codex; Cancionero musical de Segovia; Cancionero del Alcázar.” fol. CCCIIIv) (Key's
Historical and Performance Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

396
Example 3-5. Obrecht, Regina Caeli a2 from the Segovia Manuscript5

5
Obrecht, Jacob. (2019). Regina caeli (Archivo Capitular de la Catedral, Segovia, Spain [E-SE], Ms. s.
s. “Segovia Codex; Cancionero musical de Segovia; Cancionero del Alcázar.” fol. CCv) (Key's Historical and
Performance Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

397
Example 3-05 (continued).

398
Example 3-6. Anonymous, Le serviteur a2 from the Materanensis Manuscript6

6
Anonymous. (unpublished). Le serviteur (Biblioteca Comunale Augusta, Perugia, Italy [I-PEc], MS 1013
[M.36]. ff. 103v-104r) (Key's Historical and Performance Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan
Alexander Key.

399
Example 3-06 (continued).

400
Example 3-06 (continued).

401
Example 3-7. Mensural tricinia from Dygon’s Proportiones practicabiles7
7
Dygon, John. (unpublished). Three Part Proportional Examples from Proportiones practicabiles
secundum Gaffurium, (Cambridge, Trinity College [GB-Ctc], 0.3.38) (Key's Historical and Performance Editions)
(J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.
402
Example 3-8. Mensural tricinia from Morley, A Plaine and Easie Introduction.8
8
Morley, Thomas. (unpublished). Mensural Tricinia, “Christes Crosse” from A Plaine and Easie
Introduction to Practicall Musicke (Key's Historical and Performance Editions) (J. A. Key, Ed.). Gainesville, FL:
Jordan Alexander Key.
403
Example 3-08 (continued).
404
Example 3-08 (continued).
405
Example 3-08 (continued).
406
Example 3-08 (continued).
407
Example 3-08 (continued).
408
Example 3-08 (continued).
409
Example 3-08 (continued).
410
Example 3-08 (continued).
411
Example 3-08 (continued).
412
Example 3-08 (continued).
413
Example 3-9. Nathaniel Giles, Miserere from the Baldwin Commonplace Book.9
9
Giles, Nathaniel. (2018). Miserere (British Library, London, England [GB-Lbl] RM 24 d. 2 “Baldwin
Commonplace Book.” fol. 102v - 103r) (Key's Historical and Performance Editions) (J. A. Key, Ed.). Gainesville,
FL: Jordan Alexander Key.
414
Example 3-09 (continued).
415
Example 3-09 (continued).
416
Example 3-09 (continued).
417
Example 3-09 (continued).
418
Example 3-09 (continued).
419
 APPENDIX B
NOTED ERRORS AND DISCREPANCIES IN MILTON BABBITT’S PHILOMEL AND
JORDAN KEY’S RENOTATION OF BARS 1-23 OF PHILOMEL

Noted Errors and Discrepancies between Babbitt’s Manuscript and Norton’s Edition of
Philomel (mm. 1-18)

The precise rhythmic structure of Philomel is difficult to fully determine from the score since there
are discrepancies between the two “authoritative”1 editions of Section 1: the first, a facsimile-
printing of Milton Babbitt’s own manuscript and, the second, a digital-printing under the
supervision of Norton publishers from their Norton Anthology of Western Music, Volume 3. The
second of these seems to be less erroneous, which suggests either a careful study of the score by
Norton or the submission of revisions to Norton by some authority, perhaps Babbitt.

Error 1: mm. 4-5 of Babbitt2 and Norton Edition3 (m. 7 in Key edition)
The 8:9 16th-notes tuplet in the lowest two staves should be 8:13 16th-notes. This tuplet is
incorrectly notated by Babbitt as well; neither Babbitt nor Norton noticed this problem. To
account for all parts of both measures (m. 4-5), this tuplet must actually be 8:13 16th-notes,
since it comprises the last 3 16th-notes of measure 4 and the first 10 of measure 5.

Error 2: (see Chapter 4) mm 15-16 of Babbitt and Norton Edition (mm. 19-22 Key Edition)

1
The “authority” of Norton’s edition is unknown to me.

2
Babbitt, Philomel, 1.

3
Burkholder, "Milton Babbitt,” 1263.

420
Error 3: mm 16-19 of Babbitt4 and Norton Edition5
The lowest two staves in the tuplet spanning the end of m. 17 and the beginning of m. 18
cannot metrically coincide as they appear to do. The upper of the two lowest voices is within
a tuplet that is 12 16th-notes (or 6 8th-notes) in-the-time-of 10 16th-notes. First, this tuplet
should actually be 12 16th-notes (or 6 8th-notes) in-the-time-of 11 16th-notes, similar to the
tuplets in the voices above it. If this tuplet only comprises 10 16 th-notes, then one measure
is not fully accounted for in one of its constituent measures. Second, the bottom voice is
either missing a tuplet that would be similar to the voice above it (in which case it is also
missing an 8th-note within the tuplet in the first encompassed measure), or if no tuplet is
intended in the lower voice then measure 18 is missing a 16th-note in this voice. Furthermore,
if no tuplet is intended, this lowest voice should not have any coincidence with the voice
above it at this moment, as is suggested in Babbitt’s manuscript and Norton’s edition. My
edition amends for the erroneously foreshortened tuplet and does not include the tuplet in
the lowest voice (see m 24); I am ultimately unsure of Babbitt’s intent, though I believe there
should be a tuplet in this lowest voice (though my edition does not reflect this belief).

4
Babbitt, Philomel, 4.

5
Burkholder, "Milton Babbitt,” 1266.

421
Jordan Key’s Renotation of bars 1-23 of Milton Babbitt’s Philomel

422
423
424
 APPENDIX C
ORIGINAL HYPER-RHYTHMIC AND PAN-RATIONAL COMPOSITIONS
BY JORDAN ALEXANDER KEY

Here, I present my works which were referenced in this dissertation as well as those

works which are representative of his composition in hyper-rhythmicity, under both classical

dyadic notation as well as his own pan-rational system. These works represent my slow progress

to his system, attempting to imitate the masters presented in all Chapters herein as well as

develop my own rhythmic voice, both musically and notationally.

We will only provide a few brief remarks on each piece to direct the reader to each

work’s particular hyper-rhythmic aspects and the place each work holds in the development of

my use of hyper-rhythmicity. Consequently, it makes the most logical sense to present these

works in chronological succession rather than alphabetical; in order, we will introduce:

1. Discursus Anachronismus (“Discursivities against Time”) for quintet (flute, clarinet,

piano, violin, and cello
1.1. Passacaglia Mensuras, 'pater meus bac(c)h(us) est'
1.2. Machaut, Mitter, Messiaen, 'A Secret Labyrinth'
1.3. Fantasia & Fugue, 'Sumite Karissimi'
2. Saxophone Quartet No. 1: Polyptych, ‘DOCH, nichts ist genug‘
2.1. Wie doch die Zeir vergeht... (Chanson)
2.2. Das ist doch keine Kunst! (Chaconne)
2.3. Nimm doch etwas Kuchen (Fugue Fantasia)
2.4. Doch, es ist mein Fernweh... (Chorale)
2.5. Haette ich es doch nur geusst! (Canon)
2.6. Versuch’s doch mal! (Fantasia)
3. Octet: Threnody on the Death of Children for flute, E-flat clarinet, vibraphone, piano,
violin, viola, cello, and contrabass
4. To Say Pi: A Black-MIDI Ballet for fixed media
5. Bicinia: Three Pieces on Rhythmic Proportions for clarinet duo
6. Trio, Festive March for the 12th Hour for clarinet, violin, and piano
7. Nachi no Taki (那智滝), On the Inkjet Scrolls of Tomohiro Muda for fixed media
8. Viola Sonata for solo viola
9. Toccata ad Sancta Caecilia for solo organ
10. Rhythmomachia: Icosihexaplex for solo violin

425
Discursus Anachronismus

Discursus Anachronismus is about the interplay of time, both musically and figuratively,

utilizing various historical styles, forms, and references simultaneously, creating a “discourse

against time” couched in a poly-stylistic collage of conflicting musical ideas. I am not only a

composer but also a musicologist and organist; consequently, I find himself regularly pulled

between the worlds of contemporary, Baroque, and early music, since my primary research focus

and performance repertoire are music from the 14th/15th centuries and the early Baroque

respectively. Discursus Anachronismus embraces all three of these worlds, drawing of styles

from my favorite composers and periods.

The first movement, Passacaglia Mensruas, is a passacaglia that references the styles of

Bach, Bartok, and Jonathan Dawe. The second movement, Machaut, Mittner, Messiaen, ‘A

secret Labyrinth’, takes its stylistic inspiration from the eponymous composers, and uses formal

designs from the Middle Ages and Gothic Renaissance: crab canon, prolation canon, mirror

canon, and isorhythm all in one movement to create an undulating polyrhythmic stratification of

a singular musical idea temporally manipulated and played against itself and supporting a hyper-

rhythmic solo piano. The final movement, Fantasia and Fugue, ‘Sumite Karissimi’, opens with a

series of double-leading-tone-lower-third cadences, fashionable during the late Middle Ages and

Renaissance. After this introduction, the piece is a stylistic collage of all the above-mentioned

composers, with Pachelbel’s fugal style functioning as the tonal backdrop to the piece’s second

half. Ultimately, all these styles work against each other is an excited interweaving of temporally

(historically and rhythmically) distinct ideas.

A complete recording of this work can be found on YouTube.

426
Discursus Anachronismus: Movement 1, “Passacaglia Mensuras”

427
428
429
430
431
432
433
Discursus Anachronismus: Movement 2, “Machaut, Mittner, Messiaen: The Secret Labyrinth”

434
435
436
437
438
439
440
441
442
Discursus Anachronismus: Movement 3/4, “Fantasia & Fugue: Sumite, Karissimi”

443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
Saxophone Quartet

Polyptych for saxophone quartet is a set of six style variations on the old German chorale

tune “Es ist genug” (“It is enough”); however, as the subtitle suggests, “doch, nights ist genug,”

it turns out that “nothing is enough.” The suggestion that “nothing is enough” parallels my own

compositional philosophy, which attempts to synthesize from many styles and periods of music,

oftentimes quoting and parodying unapologetically. Consequently, this set of pieces whimsically

re-contextualizes the tune “es ist genug” in each movement in a different style, parodying a

composer or genre I admire.

The first variation, “Wie doch die Zeit vergeht” (“How the time does fly”), is parodying

the style of a Renaissance chanson, particularly the serpentine counterpoint of Alexander

Agricola. The second variation is a chaconne, the harmonic progression of which is based on the

chorale tune’s idiomatic opening phrase. The third variation is a homage to one of my favorite

baroque composers who is the sad victim of the “one-hit-wonder” phenomenon: Johann

Pachelbel. If you are only familiar with this composer’s “Canon in D,” I encourage you to listen

further to this giant of the Southern German Baroque. This variation is inspired by his “choral

fantasia” style, albeit with some touches of caprice.

The fourth variation is one that we should be particularly concerned with since it presents

the most hyper-rhythmic work in this piece. This variation is composed in an atonally and

rhythmically undulating style reminiscent of the New Complexity School. This work is a hyper-

abstraction of a chorale, written in homage of Paul Koonce, one of my prior composition

mentors, who would often say that someone ought to try to distill what a chorale really is and

reimagine something new yet “chorale-esque” from that distillation. Thus, this is my attempt: a

variation primarily based on choral harmonic progression with the chorale tune subtly embedded

464
in the counterpoint; however, herein we are awash in abstract non-chord tones, swelling and

snaking between our choral scaffolding. Herein, most, if not all, of the rhythms are given in such

layering that no pan-rational notation would notate this music better than the intricate, indivisible

tuplet network in which it operates.

The fifth variation returns us starkly and unabashedly to the Classical tonal idiom with a

variation in the style of the late 18th century. Pay close attention to the accompanimental material

and you will notice that it is a canon derivative of the chorale tune itself. As this festive jig spins

out of control, we arrive on a brash half-cadence that leads directly into the final variation.

The sixth variation, which was briefly featured in this dissertation, is the movement

wherein I present a stratified, indivisible mensuration executed using layered dyadic-time

signatures. The movement begins comically, failing to resolve the half cadence from the

previous variation and rhythmically stumbling over itself to find some harmonic re-coordination

before exploding into a whimsical cacophony. Listen for the baritone saxophone blasting out the

tones of the chorale amidst this whirlwind of notes. Seemingly from nowhere the whirlwind

settles and, for the first time in the course of the work, we are presented with the chorale setting

of the opening phrase of “es ist genug,” emerging like the suddenly discovered and clarifying

answer to a long befuddling riddle. However, this tranquility does not last long before a

tumultuous rush to the end with the music rapidly imploding to the final cadence.

A complete recording of this work can be found on SoundCloud.

465
Saxophone Quartet No.1: Movement 1, “Wie doch die Zeit vergeht”

466
467
468
469
470
Saxophone Quartet No.1: Movement 2, “Das ist doch keine Kunst”

471
472
473
474
475
476
477
Saxophone Quartet No.1: Movement 3, “Nimm doch etwas Kuchen”

478
479
480
481
482
483
484
485
Saxophone Quartet No.1: Movement 4, “Doch, es ost ,eom Fernweh”

486
487
488
489
490
491
Saxophone Quartet No.1: Movement 5, “Haette ich es doch nur gewusst ”

492
493
494
495
Saxophone Quartet No.1: Movement 6, “Versuch’s doch mal”

496
497
498
499
500
501
Octet: Threnody on the Death of Children

Threnody on the Death of Children was written shortly after the Marjory Stoneman

Douglas high school shooting in Florida. Consequently, this work is inspired by recent and

ongoing tragedies such as school shootings in the United States of America and the use of child

soldiers in conflicts around the world as a by-product of global violence. As a consequence of

this attempt at musical protest in a time of tragic loss, this music intertwines the emotion of

mournfulness with those of fear, anger, horror, and dread, creating a work that bears the ugliness

of our nation's current gun-financed state.

Stratified and indivisible mensurations feature heavily in this work, as noted in Chapter 6

of this dissertation. Consequently, this work is heavily influenced by Conlon Nancarrow, Ben

Johnston, and Akira Nishimura. As with prior works, the materials are a collage of past and

present styles, reaching as far back as the Renaissance, with the tune “Mein Junges Leben hat ein

End,” and Middle Ages with organum textures.

Additionally, featured in this work is my first use of non-dyadic time signatures, inspired

at this time by my study of music by Thomas Ades and Ben Johnston. My use of pan-rational

time signatures in this work, as well as his further exploration of indivisible rhythmic

stratification would ultimately lead to my further research into and development of rhythmic

notation to the ends of deeper and more comprehensive pan-rationalism.

A complete recording can be found on YouTube.

502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
To Say Pi: A Black-MIDI Ballet for fixed media

To Say Pi brings together multidisciplinary efforts in material science, applied mechanics,

art, and music. The goal of the outreach portion of this NSF grant project proposes “novel means

to reach untapped local communities at all ages… enabled through a dance-mechanics education

and outreach program,” highlighting “the innate creativity and correlations involved in [the

sciences and fine arts], and aims to inspire the next generation of STEAM (science, technology,

engineering, arts and mathematics) enthusiasts.” As such, this piece attempts to embody concepts

of numerical irrationality, particularly the number π (approximately 3.1415…). An irrational

number is simply any number that cannot be represented precisely by a ratio of two whole numbers

(a/b). With its nonterminating decimal expansion, pi cannot be expressed by a finite fraction.

To Say Pi attempts to bring together not only mathematics and music in a non-trivial way

but also bridge numerous musical genres, looking both to music’s past and future in the 21 st

century. Just as composer Conlon Nancarrow realized the limitations of human performability and

consequently sought new modes of expression in the mechanical performers of his day – player

pianos – To Say Pi seeks modes of expression in the digital capabilities of the computer. Thus, To

Say Pi exploits the digital orchestra allowed through MIDI not merely to synthetically replicate

what human orchestras can do or create sounds wholly unique to the digital world, but rather to

create a music of familiar sounds organized in a truly superhuman fashion, pointing to a new

“third-stream” music in the space between sub-genres of digital pop music, like Black-MIDI, and

20th century classical music, like Sound Mass, New Complexity, and Post-Minimalism.

To sonify pi, I have superimposed various approximations of and recursive calculating

processes for pi onto a few musical dimensions, particularly rhythm, time, and harmony. For

example, a relatively good approximation of pi is 355/113. Throughout the piece, we undulate

554
between oversimplified approximations and more precise approximations, giving the music a

sense of acceleration and deceleration, which attempts to simulate elliptical motion, such as that

of the planets around the sun. During the piece, you might notice four large sections, each with its

own instrumental colors: first, metallic sounds forming a digital gamelan; second, plucked sounds

forming a digital orchestra; third, more abstract tones and waves making a digital organ; fourth,

all these ensembles combined. The development of the materials in each section (lasting

approximately 2.5 minutes each) follows a figurative circular trajectory: building from simple

structures, adding more rhythmic layers, morphing harmonies from equal temperament to “pi-

temperament,” and then returning to simplicity with the diffusing of these materials. Within each

of these large formal circles are smaller formal circles which follow similar trajectories as the

larger ones. The temporal length of both these macro and micro formal circles follows the ratios

given above: the largest formal circles designed to encompass a polyrhythmic structure of 355/113

or 333/106 and the smaller ones designed to encompass 22/7. The choice is somewhat arbitrary as

to how far one will take the endless process of irrational computation. The conclusion of this piece

attempts to point to this possibly infinite process and the precision which it demands.

To Say Pi was commissioned under the National Science Foundation’s grant-1751989,

awarded to Dr. Leslie Lamberson and sponsored by Drexel University. Special thanks to Lucy

Bowen McCauley, choreographer of the dance component of this mini ballet, entitled Lissajous

and premiered at the Kennedy Center for the Performing Arts in May of 2019.

A recording, along with a video of the dance performed by Lucy Bowen-McCauley

Dance Studio at the Kennedy Center for the Performing Arts can be found on YouTube.

555
Bicinia: Three Pieces on Rhythmic Proportions for clarinet duo

Bicinia (plural of Bicinium) is an invented word, reminiscent of Latin, from the early 16th

century. Historically, bicinia were two-part compositions from the late Renaissance and Baroque

designed for teaching both singing and counterpoint. Throughout the 16th century, numerous

volumes of bicinia were published throughout Germany, the Low Countries, Poland, and Italy. In

modern usage, a bicinium can refer to any composition from the Renaissance for two voices;

consequently, stand-alone duets (not part of larger works like Masses) from the 15th century are

also now frequently referred to as bicinia. Not until the term “invention” arose in the mid-

Baroque with the first compositions of this type by Francesco Antonio Bonporti and came into

full usage with the popularization of the 15 Inventions by Johann Sebastian Bach did the term

“bicinia” fall out of use for two-voiced compositions. However, by the late 18th century, both

“bicinium” and “invention” were mostly abandoned as genres, and most people returned to the

common appellation “duet” or “duo” for compositions for two monophonic instruments.

Given the historical connotations of this word, Bicinia is referring directly to duets from

the 15th and 16th century. However, this work, being “three pieces on rhythmic proportions,”

does not suggest a pedagogical piece for counterpoint. Rather, this piece pays homage to the

little-known sub-genre of mensural bicinia (see Chapter 3 of this dissertation), which attempted

to teach the nuances of advanced mensural techniques to composers and performers. There are

very few extant examples of these pieces and fewer still have made it into the “canon” of

Renaissance music despite their mathematical complexity and virtuosic understanding of

counterpoint and mensuration. They are unrivaled in rhythmic technique until the music of

Henry Cowell and Conlon Nancarrow in the early and mid-20th century. The largest collection

of such pieces come from the Baldwin Commonplace Book (see Chapter 3 of dissertation) from

556
England in the late 16th century, wherein there is a whole section dedicated to complex mensural

bicinia, avant-garde for their time. The most virtuosic pieces from this collection include those

by John Baldwin and his colleague in the Royal Chapel, Nathaniel Giles. Composers from the

continent also wrote a few such pieces in the late 15th and early 16th century: Alexander

Agricola, Jacob Obrecht, Johannes Tinctoris, and Johannes Stockem to name a few. Given this

historical connection, Bicinia can be understood as a convolution - in the mathematical and

aesthetic sense - of Renaissance idioms and modern aesthetics, demonstrating the rhythmic

capabilities, quirks, and limitations of our present system of musical notation. Bicinia filters a

music from half a millennium ago through the sieve of centuries of musical development,

hopefully offering us something worthy of the 21st century.

Herein, we present three congruent versions of this work, two complete and one only the

first movement. As discussed in Chapter 6 of this dissertation, this works sits in a hyper-

rhythmic space between classical hyper-rhythmicity and pan-rationality and consequently can be

notated under a variety of forms: proportional tuplets, proportional tempos without tuplets, and

pan-rational time signatures (see Chapter 6). When Key was composing this work, he was unsure

which version would be most appealing to the performers, and so tasked himself with notating it

in these three ways. Through a side-by-side comparison one can see the limitations, benefits, and

inefficacies of each system equally.

A complete performance recording of this work can be found on YouTube.

557
Bicinia: Version 1, Tuplet Realization

558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
Bicinia: Version 2, Tempo Realization

576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
Bicinia: Version 3, Pan-Rational Realization (Mvt. 1 only)

592
593
594
595
596
597
Trio, Festive March for the 12th Hour

A trio for clarinet, violin, and piano written for and premiered by The Bold City

Contemporary Ensemble of Jacksonville, Florida, this work attempts to make a mockery of and

musically problematize the suggested violence and toxic nationalism inherent in the tradition of

"marches." As we all revel in the glory and celebration of a march, we are often celebrating and

endorsing the likely, if not inevitable, death of those marching. Furthermore, even if the march is

not militarily oriented, it still rests in a violent tradition of propagandized warmongering.

This work begins with a "tongue-in-cheek," cheery march that slowly dissolves,

reemerging as a slightly more sinister, militaristic march. This march also works itself out,

decaying again, to reemerge as something only vaguely reminiscent to a march in tempo and

strong, regulated beat structure. This third, chaotic march gradually transforms into abstract

chaos, reminiscent of battle, the implicit ultimate result of any "march" activity. This progression

represents the easy and often subtle path from innocent celebration and festivity to violence, war,

and mass death. Following this battle, a wasteland emerges, with only fragments of the marches

persisting. Here, the implicit consequence of the tradition of march is symbolized in the final

death of the music, the end to the pulsing that is both the dissolution of the march and the

heartbeat of our metaphorical "hero," who marched to their death for our celebratory delight.

Written concurrently with Bicinia, this work also features rhythmic complexity, although

not nearly as intricate as in Bicinia. In this work, I use prepared and unprepared mensural

modulations to continuously accelerate the music’s discourse towards a chaotic and furious

climax. The slow development of rhythmically dissonant polyphonic networks between the

violin, clarinet, and piano builds to the freely imitative final combat of the piece before the

598
complete desolation of the musical structure. In the final section, the haziness is infused with

subtle rhythmic nuance over an indivisible and offset polyrhythmic stratification.

A complete performance recording of this work can be found on YouTube.

599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
Nachi no Taki ( 那智滝 ), On the Inkjet Scrolls of Tomohiro Muda for fixed media

Nachi no Taki attempts to sonify the conversation between tradition and modernity

present in the work of Muda using sounds that are acoustically as well as digitally generated,

juxtaposing the "real" and "unreal," the "human" and "super-human" to comment on our present

position in artistic creation, a position that grapples regularly between "tradition" and the future.

Muda’s scroll is modern, perhaps subversive, art disguised in the traditional: a hanging

silk scroll. Once one recognizes this work is a photo, does it loose value or artistic merit as

compared to when one thought it painted? Furthermore, once one realizes that the photo is

simply “printed” on normal paper rather than meticulously developed from film onto specialized

photograph paper does it loose artistic merit even further? Perhaps this seemingly unceremonious

presentation masked by ostensible tradition now appears almost heretical. Maybe it is. But why

should it be? Why might it be? Is it still beautiful? If not, why? If so, then what does this tell us

about our obsession with traditional methodologies?

Thus, "Nachi no Taki " embodies the intersection of tradition and innovation - of our

past, present, and possible future modes of expression - challenging critiques that posit

“inhuman” tools lack human expression. In many incalculable ways, tools like the inkjet printer

and the wave audio file have revolutionized our world. If used thoughtfully, they can also

revolutionize our art, giving us creative facilities never before possible. An example close to any

composer is the use of a tool developed in the 1980s and 1990s: MIDI or Musical Instrument

Digital Interface. “MIDI Sounds” are instrumental sounds constructed from the combination of

simple digital waves on a computer and then audiated using speakers. If you listen to a video

game soundtrack from the 1990s, you will hear the stereotypical sound associated with MIDI.

MIDI can be useful for composers because one can write music and hear a synthesized

636
approximation of the music before ever handing the piece to a human performer. So, rather

than having to clunk out your work at a piano or waste a performer’s time by asking them to

read your drafts, you may, in the privacy of your own studio, sonify your music before anyone

else hears it.

The “problem” with MIDI, however, is that it often sounds “canned” or “hokey,” only

the palest imitation of the actual instrument and human performer. A MIDI Mozart Symphony

will unlikely sound as nice to you as the “real” thing. Thus, MIDI is often seen simply as a tool

for the composer, and its sound is often mocked because of its inability to “be human.” The

actual “problem” with MIDI, however, is not the “canned” nature of it sounds, but the ends to

which it is only ever put: approximating acoustically possible and acoustically intended music.

Of course, this problem lies not in MIDI itself, but in those who intend to utilize it.

While MIDI might be a poor substitute for music ultimately written for acoustic

instruments and human performers, MIDI has all the capabilities of a computer to realize

whatever you might imagine, things that lie well beyond the limitations of acoustic instruments

and human hands, fingers, lungs, brains, et cetera. So, I argue that we should measure the

“virtue” of MIDI-music based not on its approximations of acoustic music, but on music written

specifically with MIDI in mind. Similarly, we should judge acoustic music only after its acoustic

realization, not only merely on its MIDI imitation.

Interestingly, genres of MIDI-music have emerged, despite much criticism of the genre.

One popular genre of MIDI music in the popular music industry is known as “Black-MIDI,” which

realizes the capability of MIDI to play hundreds, if not thousands, of notes per second, resulting

in entire pop songs of 3.5 minutes containing within themselves millions, if not billions, of

637
notes. In some of my latest electronic works, I have attempted to bring MIDI-music, inspired by

some explorations of “Black-MIDI,” into the “classical” arena, writing “orchestral” music that

lies between the “real” and “unreal,” the “human” and “superhuman.” There are things within

this and other works of mine that sound clearly “real” and possibly “human,” but these are

mixed with things likely well beyond the capabilities of either people or their acoustic

instruments, but well within the capabilities of computer-assisted sound generation.

Not intrinsic to the program of this work generally, but part of my compositional process

in generating some of the sounds that you will hear in this work is my use of large-integer

rational approximation of the √2 to generate many of the pitched white noise/waterfall

sounds. While some of the initial waterfall sounds in the work are actual recording of the Nachi

no Taki waterfall in Japan, as the work progresses, I generate my own waterfall through the

superhuman hyper-rhythmicization of the flute choir in this work. Similar to the conclusion of

To Say Pi, I use “Black-MIDI” sound structures, incorporating thousands of musical data points

per second, to create various ‘noise’ textures that are “artificial” or “printed” waterfalls.

A complete recording of this work can be found on YouTube.

638
Viola Sonata for solo viola

"Ceol Mor" is the first part to a larger work for solo viola. This work (both this part and

the whole), takes its inspiration from traditional forms, sounds, and rhythms of Scottish bagpipe

music, particularly from two primary genres: "Ceol mor" (a Gaelic word meaning "big music")

and "Ceol baeg" (meaning "little music"). "Ceol Mor" (also known as "Piobaireachd,"

pronounced "pee-oh-brach") is a traditional form of Scottish bagpipe music from the Middle

Ages, formed around very basic themes, often only a few notes, and continuous variations of that

theme using a system of elaborate embellishments. While this movement uses no specific

piobaireachd for its material, the repetitive structure with slight variations in motivic and

rhythmic execution with a slow build in tempo do pay homage to the style. Being a bagpiper

himself, I often search for new ways to use the bagpipe in music as well as new ways to envision

or contextualize traditional genres of Scottish music in "classical" concert venues. My new

works for bagpipe have received international recognition, publication, and performance; most

recently my work, "Microtonal Piobaireachd on Prime Numbers" was the winner of the

Piobaireachd Society's "Modern Pibroch Library” 2018 competition.

As discussed in Chapter 6 of this dissertation, this work continues to explore non-dyadic

time signatures, after my first use of them in Threnody. It was in this work I encountered the

desire to create pan-rational subdivisions of non-dyadic note values. Thus, concurrently, I began

composing Toccata (see later in this Appendix) to explore the possibility of intermingling pan-

rational rhythmic values within the same measures and establishing differing subdivision trees

among those values. This work would not realize that goal, which would come to fruition later in

Rhythmomachia: Icosihexaplex.

A synthesized recording of this work can be found on SoundCloud.

639
640
641
642
643
644
645
646
647
648
Toccata ad Sancta Caecilia for solo organ

This work uses what I call “non-dyadic time signatures” (known elsewhere and erroneously

as “irrational time signatures”). I do not call them “irrational” since they are in no way “irrational”

(either psychologically or mathematically). They are “rational” proportions, but do not use powers

of 2 for the bottom number, hence “non-dyadic” (not written over 2, 4, 8, 16, 32, 64, etc.). What

we call a triplet is actually just some multiple or divisions of a “third note,” and similarly the

quintuplet some “fifth note.” Systems of subdividing fundamental note values (like the breve or

whole-note) into groups other than “dyadic-rationals” (third, fifth, seventh as compared to half,

fourth, eighth, sixteenth, etc.) have existed at prior points in music history (particularly the late

Middle Ages and Renaissance), using uncommon and arcane systems of mensuration (see Chapter

2 and 3 of this dissertation). Such systems have long fallen out of favor; however, if reincorporated

into our system of musical language, they can allow for the fluid notation of some rather complex

rhythmic ideas without the use of perhaps excessive tempo changes or tuplet (and embedded

tuplet) markings.

For example, imagine if one wanted to write a quarter note triplet in a 4/4 time signature,

but they only wanted two-thirds of that triplet (namely, two quarter notes of the three within the

quarter-note triplet), and then they wanted to return directly to duple time with no hint of the prior

triplet. What would such a music look like? The composer would have to write the two notes of

the triplet, but then, in returning to the simple duple time from the 4/4, they would have to endlessly

tie over values from within triplets. However, if the composer just recognize that they can create

“sixth notes,” then they can specify that they want two “sixth-notes” without the use of triplets and

many ties subsequently. This can be done using a “non-dyadic” time signature, namely 2/6. See

Figure A-C1.

649
Figure A-C1. Traditional vs. Pan-Rational notation of triplets intermixed with quarter notes.

Why use “2/6” and “sixth-notes” rather than “third-notes”? Well, if we consider the

“whole-note” as the fundamental value of rhythm, then to get the quarter-note triplets, we must

recognize that the whole-note must be divided equally into six parts. Then, to get only two of these

six equal divisions of the whole note, we simply specify “2,” just as one might specify “3” in 4

equal divisions of the whole to get 3/4 rather than 4/4.

Similarly, if we want seven units of five equal divisions of the whole note (1 and 2/5’s of

an eighth-note quintuplet), we recognize that we must divide the whole into 10 equal parts and

play seven of those parts, thus requiring a 7/10 time signature and “tenth-notes.” See Figure A-C2.

Figure A-C2. Traditional vs. Pan-Rational notation of septuplets intermixed with quarter notes.

Such a system allows for the rapid alternation between different tempos without the use of

tempo markings on each measure. Furthermore, it utilizes and recognizes those rhythmic

fluctuations that are proportional to each other. The notational possibilities of non-dyadic time

650
signatures as compared to tempo changes are demonstrated, in a rudimentary form, in Figure A-

C3 below.

Figure A-C3. Demonstration of metrical intermingling of pan-rational rhythmic values according

to proportional tempos.

Note that this work is a prototype of the system of pan-rational rhythmic notation given in

this dissertation. All diamond-shaped note heads in this work correspond to “seventh notes” and

all binary multiples/divisions of them (1/14, 1/7, 2/7, 4/7 etc.) by conventional usage of stems,

flags, and/or coloration; the time signatures within the piece will suggest as such. The tempo of

the seventh-note is given at the beginning of the score, but one may discover the “feel” of the

seventh-note simply by finding the tempo of the half-note, dividing it into a septuplet, and taking

the septuplet as the 1/14th note (namely, half of the seventh note).

651
652
653
654
655
Rhythmomachia: Icosihexaplex for solo violin

Rhythmomachia: Icosihexaplex represents the first systematic use of Key’s pan-rational

system of rhythmic notation. This work is technically incomplete at present, the movement

Icosihexaplex only one etude in a set intended to cover all the low prime pq-time signature

combinations. Key is currently composing more to complete the set of Rhythmomachia.

Nothing further needs to be said of this piece since it is extensively covered in Chapter 6 of this

dissertation.

656
657
658
659
660
661
662
663
664
665
 LIST OF REFERENCES

Musical Scores:
Adam. (unpublished). De tous biens playne (Archivo Capitular de la
Catedral, Segovia, Spain [E-SE], Ms. s. s. “Segovia Codex; Cancionero musical de
Segovia; Cancionero del Alcázar.” fol. CCIv) (Key's Historical and Performance
Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

Adès, Thomas. Piano quintet. London: Faber Music, 2017.

_____. Tevot: for large orchestra. London: Faber Music, 2007.

_____. Totentanz: For Mezzo-Soprano, Baritone and Orchestra. London: Faber, 2013.

_____. Traced Overhead for Solo Piano, Op. 15. (1995-96). London: Faber Music, 1997.

_____. Violin Concerto: Concentric Paths : Op. 23 (2005). London: Faber Music, 2010.

Anonymous. (unpublished). French Double Motet upon an ostinato, ‘Cil s'entremet / Nus hom /
Victimae paschali laudes’ (Source: Bibliothèque Interuniversitaire, Section Médecine,
Montpellier, France [F-MO H 196], fol. 223v-223r) (Key's Historical and Performance
Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

______. (unpublished). Le serviteur (Biblioteca Comunale Augusta, Perugia, Italy [I-PEc], MS

1013 [M.36]. ff. 103v-104r) (Key's Historical and Performance Editions) (J. A. Key,
Ed.). Gainesville, FL: Jordan Alexander Key.

______. (unpublished). Mass of Tournai, Ite Missa Est (Source: Chapitre de la Cathedrale,
Tournai,

Babbitt, Milton, and John Hollander. Philomel: for soprano, recorded soprano, and synthesized
sound. 1964.

______, Mimmi Fulmer, Bethany Beardslee, Godfrey Winham, David Dies, and Judith
Kellock. Milton Babbitt's Vision and Prayer: An Introduction, Performer's Guide and
Study Score. Verona, New Jersey: Subito Music, 2016.

Belgium [B-Tc], 476 “Mass of Tournai.” fol. 33v.) (Key's Historical and Performance Editions)
(J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

_______. (unpublished). Mass of Tournai, Kyrie (Source: Chapitre de la Cathedrale, Tournai,

Belgium [B-Tc], 476 “Mass of Tournai.” fol. 28r.) (Key's Historical and Performance
Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

Biblioteca Comunale Augusta [I-PEc], MS 3065 “Mancini Codex/Lucca Codex,” no. 9 fol.
LXXXIII, “Ciconia”

666
Burkholder, Peter J., Claude V. Palisca, Donald J. Grout, and Barbara R. Hanning. "Milton
Babbitt (b. 1916), Philomel: Section 1, Monodrama for Soprano, Recorded Soprano, and
Synthesized Sound (1964)." In Norton Anthology of Western Music, 1263-279. Vol. 3.
New York: W.W. Norton, 2014.

Carter, Elliott. A mirror on which to dwell: (six poems of Elizabeth Bishop) : for soprano and
chamber orchestra. New York: Associated Music Publishers, 2001.

______. Double Concerto for Harpsichord and Piano with Two Chamber Orchestras (1961).
New York: Associated Music Publishers, 1994

______. Quintet for piano and winds. Boosey & Hawkes, 1991.

______. String quartet no. 1. New York, NY: Associated Music Publishers, 1994.

______. Sonata for flute, oboe, cello, and harpsichord (1952). New York: Associated Music
Publishers, 1997.

______. Triple duo: (1982). [New York]: Hendon Music, 1990.

Chávez, Carlos. Tercera sonata para piano. México, D.F.: Ediciones Mexicanas de Música,
1989.

Ciconia, Johannes. (2018). Le ray au soleyl (Mancini Codex/Lucca Codex, I-PEc MS 3065 no. 9
fol. LXXXIIIr) (Key's Historical and Performance Editions) (J. A. Key, Ed.). Gainesville,
FL: Jordan Alexander Key.

Cowell, Henry. Fabric: for piano. New York: Breitkopf, 1922.

______. Quartet Romantic and Quartet Euphometric. New York: Peters, 1974.

______. Rhythmicana for piano. New York: Associated Music Publishers, Inc, 1975.

______. Six ings: for piano. New York: Associated Music Publishers, 1950.

Dawe, Jonathan. Liber de arte contrapuncti 1477/2001: for string quartet. 2001.

Dygon, John. (unpublished). Three Part Proportional s from Proportiones practicabiles

secundum Gaffurium, (Cambridge, Trinity College [GB-Ctc], 0.3.38) (Key's Historical
and Performance Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

Ferneyhough, Brian. In nomine a 3: for piccolo, oboe and clarinet, [2001]. London: Peters,
2009.

_____. Second string quartet. London: Peters, 1981.

667
Finnissy, Michael. English country-tunes: [1977, for piano solo]. London: United Music
Publishers, 1986.

Giles, Nathaniel. (2018). Miserere (British Library, London, England [GB-Lbl] RM 24 d.

2 “Baldwin Commonplace Book.” fol. 102v - 103r) (Key's Historical and Performance
Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

Gordon, Michael. Four kings fight five: for nine instruments (1988). New York, NY: Red Poppy
Music, 2020.

Guido. Or voit tout en aventure. Chantilly, Bibliotheque du Chateau, MS. 564, f. 25v.

Hothby, John. (2020). Ora Pro Nobis (Biblioteca Comunale, Faenza, Italy [I-FZc] MS
117 (Faenza), ff. 26v - 27r) (Key's Historical and Performance Editions) (J. A. Key, Ed.).
Gainesville, FL: Jordan Alexander Key.

Ives, Charles. In re con moto et al; for piano quintet. New York: Peer International Corp, 1968.

______. Largo Risoluto, Piano Quintet No. 2. New York: Peer International, 1961.

______. Scherzo: Over the pavements: for chamber orchestra, performable also as chamber
music. New York: Peer International, 1954.

______. Scherzo, for 2 violins, viola, and 'cello. New York: Peer International Corp, 1958.

______. String quartet no. 2, for two violins, viola and cello. New York: Peer International Corp,
1980.

______. The unanswered question: for chamber orchestra or chamber ensemble. New York:
Southern Music Pub. Co, 1953.

______. Three places in New England, III. The Housatonic at Stockbridge. New York: Mercury
Music Corporation, 1935.

______. Tone roads no. 3. New York: Peer international, 1952.

Johnston, Ben. String quartet number 4: for the Fine Arts Quartet. Baltimore, MD: Smith
Publications, 1990.

_____. String quartet no. 5. Baltimore, MD: Smith Publications, 1992.

_____. String quartet no. 10: 1995. Baltimore, MD: Smith, 1999.

_____. Sextet for flute and strings. Baltimore, MD: Smith Publications, 1991.

668
Jones, Daniel. Sonata: for three unaccompanied kettledrums (one player). Jan Williams
Collection of Annotated Scores, 1950-1999. London: Hinrichsen Edition Ltd, 1953.

Khachaturian, Karen Surenovich. Trio, fuer Violine, Viola und Violoncello. Wien: Universal
Edition, 1990.

Médecine, Montpellier, France [F-MO H 196], fol. 9v-12r) (Key's Historical and Performance
Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

Morley, Thomas. (unpublished). Mensural Tricinia, “Christes Crosse” from A Plaine and Easie
Introduction to Practicall Musicke (Key's Historical and Performance Editions) (J. A.
Key, Ed.). Gainesville, FL: Jordan Alexander Key.

Nancarrow, Conlon. Collected Studies for Player Piano Vol. 2: Study No. 41 for Player Piano.
Santa Fe, CA: Soundings Press, 1981.

_____. Collected Studies for Player Piano Vol. 5: Studies No. 2, 6, 7, 14, 20 ,21, 24, 26, and 33.
Santa Fe, CA: Soundings Press, 1984.

_____. Rhythm study no. 1: for player piano. Bryn Mawr, Pa: T. Presser, 2000.

Nishimura, Akira. String quartet no. 4: "Nrsimha" = Gengaku shijūsoukyoku dai-4-ban :

Nurushinha. Tokyo: Zen-On, 2009.

Obrecht, Jacob. (2019). Regina caeli (Archivo Capitular de la Catedral, Segovia, Spain [E-SE],
Ms. s. s. “Segovia Codex; Cancionero musical de Segovia; Cancionero del Alcázar.” fol.
CCv) (Key's Historical and Performance Editions) (J. A. Key, Ed.). Gainesville, FL:
Jordan Alexander Key.

Perotin. (unpublished). Alleluia, Nativitas Gloriosae (Source: Bibliothèque Interuniversitaire,

Section

Redgate, Roger. Genoi hoios essi: pour piano. Paris: H. Lemoine, 1988.

Rodericus. Angelorum psalat tripudium. Chantilly, Bibliotheque du Chateau, MS. 564, f. 48v.

______. (unpublished). Angelorum psalat tripudium (Chantilly, Bibliotheque du Chateau, MS.

564, f. 48v) (Key's Historical and Performance Editions) (J. A. Key, Ed.). Gainesville,
FL: Jordan Alexander Key.

Stockem, Johannes. (2019). Ave maris stella (Biblioteca Comunale Augusta, Perugia, Italy [I-
PEc] Ms. 1013 [M.36] ff. 63v-70r, "Johannes Stochem") (Key's Historical and
Performance Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

Thoegersen, Peter Alexander. Hypercube: a polytempic polymicrotonal work for string quartet.
2012.

669
Tinctoris, Johannes. (unpublished). Le Souvenir (Archivo Capitular de la
Catedral, Segovia, Spain [E-SE], Ms. s. s. “Segovia Codex; Cancionero musical de
Segovia; Cancionero del Alcázar.” fol. CCCIIIv) (Key's Historical and Performance
Editions) (J. A. Key, Ed.). Gainesville, FL: Jordan Alexander Key.

Wellesz, Egon. Die Bakchantinnen. Berlin: Bote & Bock, 1931.

Primary Sources:
Agricola, Martin. Musica figuralis deudsch (1532): Musica instrumentalis deudsch (1529),
Musica choralis deudsch (1533), Rudimenta musices (1539). Hildesheim: George Olms,
2011.

_____. Musica instrumentalis deudsch. Wittemberg, 1528.

Archivo Capitular de la Catedral, Segovia, Spain (E-SE), Ms. s. s. “Segovia Codex; Cancionero
musical de Segovia; Cancionero del Alcázar”

Arditti, Irvine. Questionnaire response in Complexity in Music? An Inquiry into its Nature,
Motivation and Performability. Edited by Joel Bons. Netherlands: Job Press, 1990, 9.

Artusi, Giovanni Maria. L'arte del contraponto. Sala Bolognese: A. Forni, 1980.

Babbitt, Milton. “Twelve-Tone Rhythmic Structure and the Electric and the Electronic Medium,”
Perspectives in New Music I, no. 2 (Fall 1962).

Baldwin, John. Commonplace Book. London, British Library (GB-Lbl), R.M. 24.d.2, “Anthology
of Miscellaneous Music Including Textless Madrigals and Motets, Masses, Songs,
Canons and Excercises by English and Italian Composers. 1591.”

Bartolozzi, Bruno. "Proposals for Changes in Musical Notation.” Journal of Music Theory. 5, no.
2: 297-301, 1961.

Bathe, William, and Cecil Hill. A Brief Introduction to the True Art of Music. Colorado Springs:
Colorado College Music Press, 1979.

Berlioz, Hector. A Travers Chants. Paris: Michel Lévy frères, 1862.

Beswick, Laura H. Musical Notation. U.S. Patent No. 733,351, U.S. Patent and Trademark
Office, 1903.

Biblioteca Comunale Augusta, Perugia, Italy (I-PEc), MS 1013 (M.36)

Boethius, Anicius Manlius Severinus, Claude V. Palisca, and Calvin M. Bower. Fundamentals of
Music. Music theory translation series. New Haven: Yale university press, 1989.

670
Bononcini, Giovanni Maria. Musico prattico che brevemente dimostra il modo di giungere alia
perfetta cognizione di tutte quelle cose, che concorrono alia composizione di i canti, e di
cid ch’all’ arte del contrapunto si ricerca. Bologna: per Giacomo Monti, 1673.

Borin [formerly attributed to Didier Saurin]. La musique theorique, et pratique, dans son
ordre naturel: nouveaux principes par mr. XXXXX. Paris: T. B. C. Ballard, 1722.

Burmeister, Joachim, Pascal Dubreuil, and Agathe Sueur. Musica poetica (1606) augmentée des
plus excellentes remarques tirées de "Hypomnematum musica poetica" (1599) et de
"Musica autoschediastike" (1601). Wavre: Mardaga, 2007.

Cambridge, Trinity College (GB-Ctc), 0.3.38

Carter, Elliott. 1963. "Letters to the Editor". Journal of Music Theory. 7, no. 2: 270-273.

_____. n.d. "Music and the Time Screen". Current Thoughts in Musicology.

_____. The Writings of Elliott Carter: an American Composer Looks at Modern Music. Edited
by Else Stone and Kurt Stone. Bloomington, Indiana: Indiana University Press, 1977.

_____, and Sue Knussen. 1996. "Elliott Carter in Interview". Tempo. New Series. 2-7.

Caza, Francesco, and Johannes Wolf. Tractato vulgare de canto figurato, Mailand 1492. Berlin:
Breslauer, 1922.

Coclico, Adrianus Petit, and Manfred F. Bukofzer. Compendium musices. Kassel: Bärenreiter,
1954.

Cole, Hugo. Sounds and Signs: Aspects of Musical Notation. London: Oxford University Press,
1974.
Cologne, Franco de, Gilbert Reaney, and Franco. Franconis de Colonia Ars cantus mensurabilis.
[Rome]: American Institute of Musicology, 1974.

Cowell, Henry. New Musical Resources. New York: A.A. Knopf, 1930.

Creston, Paul. Rational Metric Notation: The Mathematical Basis of Meters, Symbols, and Note-
Values. Hicksville, N.Y.: Exposition Press, 1979.

David, Francois. Methode nouvelle ou principes generaux pour apprendre facilement la

musique, et l’art de chanter. Paris: Vve. Boivin, 1737.

Dupont, Pierre. Principes de musique, par demande et par reponse, par lequel toutes personnes
pourront aprendre d’eux meme a connoitre toutte la musique. Paris: Chez 1’hauteur,
1718.

671
Dygon, John, Theodor Dumitrescu, and Franchinus Gaffurius. John Dygon's Proportiones
practicabiles secundum Gaffurium = Practical proportions according to Gaffurius.
Urbana: University of Illinois Press, 2006.

Ferneyhough, Brian, James Boros, and Richard Toop. Brian Ferneyhough: Collected Writings.
Australia: Harwood Academic, 1996.

Finnissy, Michael. 2002. "Biting the Hand That Feeds You". Contemporary Music Review. 21,
no. 1: 71-79.

Foundation Gaudeamus (Amsterdam, Netherlands). Complexity in Music?: An Inquiry into Its

Nature, Motivation and Performability. Rotterdam: Gaudeamus Foundation, Nieuw
Ensemble and the Rotterdam Arts Council, 1990.

Fuhrmann, Martin Heinrich. Musikalischer Trichterl dadurch ein geschickter Informator seinen
Informandis die Edie Singe-Kunst nach heutiger manier bald und leicht einbringen kan/
darinn Vitiosa ausgemustert: Obscura Erlautert: Deficienta aber Erstattert/ mit einer
Vorrede von der heutigen Music Vollkommenheit/ Kraffte/ Nutz und Nothwendigkeit!
herausgegeben durch ein mitglied der Singenden und Klmgenden Gesellschafft.
Franckfurt an der Spree, 1706.

Gaffurius, Franchinus, and Clement Albin Miller. Practica Musicae. [S. l.]: American Institute
of Musicology, 1968.

Ganassi, Silvestro. Opera intitulata fontegara: la quale isegna a sonare di flauto cho tutta l'arte
opportuna a esso istrumento, massime il diminuire il quale sara utile ad ogni istrumento
di fiato et chorde: et achora a, chi si dileta di canto. Milano: Bolletino Bibliografico
Musicale, 1934.

Glarean, Heinrich, and Clement A. Miller. Dodecachordon. [Dallas, Tex.]: American Institute of
Musicology, 1965.

Grassineau, James. A musical dictionary being a collection of terms and characters as well
ancient as modern: including the historical, theoretical and practical parts of music.
Carefully abstracted from the best authors in the Greek, Latin, Italian, French and
English Languages. London: Wilcox, 1740.

Gross, J., & Babbitt, M. (2000). “Milton Babbitt talks about ‘Philomel’.” Perfect Sound Forever,
April. Online journal, retrieved 20 November 2018, from
[Link] [Link]/

Handlo, R. D., Hanboys, J., & Lefferts, P. M. (1991). Regule and Summa: A new critical text and
translation on facing pages, with an introduction, annotations, and indices verborum and
nominum et rerum. Lincoln, NE: University of Nebraska Press.

672
Horsley, Imogene. 1951. "Improvised Embellishment in the Performance of Renaissance
Polyphonic Music". Journal of the American Musicological Society. 4, no. 1: 3-19.

Hothby, John, Gilbert Reaney, and Thomas Walsingham. Opera omnia de musica mensurabili.
[Rome]: American Institute of Musicology, 1983.

Ives, Charles, and John Kirkpatrick. Memos. New York: W.W. Norton, 1972.

Jaques-Dalcroze, Emile. Eurhythmies Art and Education. London: Chatto & Windus, 1930.

_____. Rhythm, Music & Education. New York, 1921.

Johnston, Ben. "Maximum" Clarity and Other Writings on Music. Urbana, Illinois: University of
Illinois Press, 2007.

Jones, Daniel Jenkyn. “Some Metrical Experiments.” The Score 3 (June 1950): 32–48.

Karkoschka, Erhard, and Ruth Koenig. Notation in New Music: A Critical Guide to
Interpretation and Realisation. London: Universal Edition, 1975.

London, British Library (GB-Lbl), Additional 10336

London, Lambeth Palace (GB-Llp), 466

Longy-Miquelle, Renée. Principles of Musical Theory. Boston: E.C. Schirmer, 1949.

Loulie, Etienne, and Albert Cohen. Elements or Principles of Music. New York: Institute of
Mediaeval music, 1965.
Malcolm, Alexander. A treatise of musick, speculative, practical and historical, containing. . .
the nature and office of the scale of musick, the whole art of writing notes, and the
general rules of composition, with a particular account of the ancient musick and a
comparison thereof with the modern. Edinburgh, 1721.

Marchetto of Padua, Pomerium, in Marcheti de Padua Pomerium, ed. Giuseppe Vecchi, Corpus
Scriptorum de Musica 6. Rome: American Institute of Musicology. 1961.

Manfredini, Vincenzo. Regole armoniche o sieno precetti ragionati per apprendere i principi
della musica, il portamento della mano, et I’accompagnamento del basso, sopra gli
strumenti da tasto, come I’organo, il cembalo, ec.. Venice: Zerletti, 1775.

Monachus, Guilielmus and Albert Seay. De preceptis artis musicae. [Place of publication not
identified]: American Institute of Musicology, 1965.

673
Montéclair, Michel Pignolet de. Nouvelle methode pour apprendre la musique: par des
demonstrations faciles, suivies d'un grand nombre de leçons à une et à deux voix, avec
des tables qui facilitent l'habitude des transpositions et la conoissance des differentes
mesures. Pre-1801 Books About Music and Related Topics in the Library of Congress
Collections As Listed in RISM, Volume B-VI [Microform]. Paris: Chez l'Auteur, 1709.

Muris, Johannes de, and Christian Berktold. Ars practica mensurabilis cantus secundum
Iohannem de Muris: die Recensio maior des sogenannten "Libellus practice cantus
mensurabilis. München: Verlag der Bayerischen Akadamie der Wissenschaften, 1999.

______, Petrus, and Ulrich Michels. Johannis de Muris Notitia artis musicae et Compendium
musicae practicae. Tractatus de musica. [S.l.]: American Institute of Musicology, 1972.

Morley, Thomas. A Plaine and Easie Introduction to Practicall Musicke. Farnborough: Gregg,
1971.

Nancarrow, Conlon, and Alistair M. Riddell. 1988. "Tempus perfectum: a conversation with
Conlon Nancarrow". Meanjin. 47 (2): 266-273.

Penna, Lorenzo. Li primi albori musicali per li principianti della musica figurata distinti in trë
libri: dal primo spuntano li principij del canto figurati; dal secondo spiccano le regole
del contrapunto; dal terzo appariscono li fondamenti per suonare l'organi ò
clavicembalo sopra la parte. Bologna, P.M.: M. Monti, 1696.

Pennington, Jo. The Importance of Being Rhythmic ; A Study of the Principles of Dalcroze
Eurythmics Applied to General Education and to the Arts of Music, Dancing and Acting.
New York: G.P. Putnam's Sons, 1925.

Pisa, Agostino and Piero Gargiulo. Breve dichiaratione della battuta musicale: (1611): con
alcuni estratti da Battuta della musica (1611). Lucca: Libreria musicale italiana, 1996.

Praetorius, Michael. Syntagma Musicum. Kassel: Barenreiter, 1959.

Printz, Wolfgang Caspar. Compendium musicae signatoriae & modulatoriae vocalis: oder
Kurtzer begriff aller derjenigen dinge, so einem der die vocal-music lernen will zu wissen
von nöthen seyn auffgesetzt und nunmehro zum andern mahl vermehret und verbessert.
Dressden und Leipzig: Bey Johann Christoph Miethen, 1714.

Prosdocimus and Alberto Gallo. Expositiones tractatus practice cantus mensurabilis magistri
Johannis de Muris. 1966.

______ and Alberto Gallo. Prosdocimi de Beldemandis opera. Bologna: [Arti grafiche Tamari],
1966.

674
______ and Jan Herlinger. Brevis Summula Proportionum Quantum Ad Musicam Pertinet and ;
Parvustractatulus De Modo Monacordum Dividendi = A Short Summary of Ratios
Insofar As They Pertain to Music and = A Little Treatise on the Method of Dividing the
Monochord. Lincoln: University of Nebraska Press, 1987.

Purcell, Henry, Frances Purcell, and Henry Playford. A Choice Collection of Lessons for the
Harpsichord or Spinnet. 1963.

Ramos de Pareja, Bartolomeo, and Johannes Wolf. Musica practica, Bartolomei Rami de Pareja,
Bononiae Impressa opere et industria ac expensis magistri Baltasaris de Hiriberia
MCCCCLXXXII. [Wiesbaden]: [Breitkopf u. Härtel], 1969.

Raymondi, Joseph. Examen critique des notations musicales proposées depuis deux siècles.
Paris: Librairie encyclopédique de Roret, 1856.

Read, Gardner. Modern Rhythmic Notation. Ann Arbor, Mich: Books on Demand, 1994

_____. 1965. "Some Problems of Rhythmic Notation". Journal of Music Theory. 9, no. 1: 153-
162.

_____. Source Book of Proposed Music Notation Reforms. New York: Greenwood Press,
1987.

Reynolds, Roger. "Conlon Nancarrow: Interviews in Mexico City and San Francisco." American
Music. 2, no. 2 (1984): 1.

Ross, Alex. 1998. "PROFILE Roll Over, Beethoven - The Musical World of Thomas Ades". The
New Yorker. 110.

Rousseau, Jean-Jacques. Dictionnaire de musique 1768. Oeuvres Completes de. J.-J.

Rousseau, vols. XII and XIII. Paris: Dupont, 1824.

_____. Dissertation sur la musique moderne. Oeuvres Completes de J.-J. Rousseau, vol. XI,
pp. 21-140. Paris: Dupont, 1824.

_____. Projet concernant de nouveau signes pour la musique. Oeuvres Completes de J.-J.
Rousseau, vol. XI, pp. 1-17. Paris: Dupont, 1824.

Schumann, Robert. Neue Zeitschrift für Musik. Leipzig: J.A. Barth, 1835.

Sperling, Johann Peter. Principia musicae, das ist Griindliche Anweisung zur Musik! wie ein
Musik-Scholar vom Anfang instruiret und nach der Ordnung zur Kunst oder
Wissenschaft der Figural-Musik soil gefubret und gwiesen werden. Bautzen: bey Andreas
Richter, 1705.

Tigrini, Orazio. Il compendio della musica. New York: Broude, 1966.

675
Tinctoris, Johannes, and Albert Seay. Opera theoretica. [Place of publication not identified]:
American Institute of Musicology, 1975.

______, and Heinrich Bellermann. Terminorum musicae diffinitorium. 1983.

Ugolino, and Albert Seay. Declaratio musicae disciplinae. Rome: American Institute of
Musicology, 1962.

Valentini, Pier Francesco. Trattato della battuta musicale. 1643.

Van der Elst, Johannes. Notæ Augustinianæ: sive Musices figuræ sev notæ novæ concinendis
modulis faciliores, tabulaturis organicis exhibendis aptiores ; quas firmamento ecclesiæ
magno patriarchæ divo Aurelio Augustino musices ecclesiasticæ protectori venerabundus
dedicat F. Ioannes vander Elst. Gandavi: Maximiliani Graet, 1657.

Vega, Carlos. La musica popular argentina: canciones y danzas criollas. Buenos Aires:
Imprenta de la Universidad, 1941.

Virdung, Sebastian, and Klaus Wolfgang Niemöller. Musica getutscht. Kassel: Bärenreiter, 1983.

Vitry, Philippe de, Gilbert Reaney, André Gilles, and Jean Maillard. Ars nova. Stuttgart, W.
Germany: American Institute of Musicology, 1986.

Walther, Johann Gottfried. Praecepta der Musikalischen Composition 1708. Herausgegeben von
Peter Benary. Jenaer Beitrage zur Musikforschung, Band 2. Leipzig: Breitkopf und
Hartel, 1955.

Zarlino, Gioseffo, Iain Fenlon, and Paolo Da Col. Le istitutioni harmoniche: Venezia, 1561.
[Sala Bolognese]: Forni, 2008.

Secondary Sources:
Adamowicz, Emily J. 2011. "Subjectivity and Structure in Milton Babbitt’s Philomel". Music
Theory Online. 17 (2).

Barthelmes, Barbara. Raum und Klang: das musikalische und theoretische Schaffen Ivan
Wyschnedgradskys. Hofheim: Wolke Vlg, 1995.

Baxandall, Michael. Painting and Experience in Fifteenth Century Italy. Oxford, UK: Oxford
University Press, 1988.

Beardslee, Bethany, and Minna Proctor. I Sang the Unsingable: My Life in Twentieth-Century
Music. 2017.

Belling, Hue. “Thinking Irrational, Thomas Adès and New Rhythms.” Master’s thesis, Royal
College of Music, London, 2010.

676
Berger, Anna Maria. Mensuration and Proportion Signs: Origins and Evolution. Oxford:
Clarendon Press, 1993.

______. 1990. "Musical Proportions and Arithmetic in the Late Middle Ages and
Renaissance". Musica Disciplina. 44: 89-118.

Bernard, Jonathan W. 1988. "The Evolution of Elliott Carter's Rhythmic Practice". Perspectives
of New Music. 26, no. 2: 164-203.

Blachly, Alexander. 2001. "‘Mensura‘ Versus ‘Tactus‘“. [Quellen Und Studien Zur Musiktheorie
Des Mittelalters, Bd. 3]. 425-467.

Boeke, Kees and Jill Feldman. “Or voit tout en aventure.” Text translations in accompanying
booklet, Codex Chantilly II: Tetraktys. 2008-2011. Olive Music KTC 1905,
2011. Compact disc.

Bray, Roger, and Ian Spink. The Blackwell History of Music in Britain. Vol. 2. Oxford:
Blackwell, 1995.

Callender, Clifton. "Performing the Irrational: Paul Usher’s arrangement of Nancarrow’s Study
No. 33, Canon 2:√2." Conlon Nancarrow, Life and Music: Online Symposium September
27 - October 27, 2012. September 2012. Accessed December 02, 2017.
[Link]

Cantor, Moritz. Vorlesungen über Geschichte der Mathematik. Bd. 2 Bd. 2. Vaduz: Sändig
Reprint, 2005.

Carapetyan, Armen. A Early Fifteenth-Century Italian Source of Keyboard Music: The Codex
Faenza, Biblioteca Comunale, 117. [Place of publication not identified]: American
Institute of Musicology, 1961.

Coussemaker, Charles Edmond Henri de. Scriptorum De Musica Medii Aevi. Hildesheim, W.
Germany: G. Olms, 1987.

Crawford, Richard. An Introduction to America's Music. New York: Norton, 2001.

DeFord, Ruth. Tactus, Mensuration and Rhythm in Renaissance Music. 2018.

Drott, Eric Austin. 2004. "Conlon Nancarrow and the technological sublime". American
Music. 22 (4): 533-563.

Duncan, Stuart Paul. The Concept of New Complexity: Notation, Interpretation and Analysis.
Part I of thesis (D.M.A.), Cornell University, May 2010.

Eggebrecht, Hans Heinrich. Die mittelalterliche Lehre von der Mehrstimmigkeit. 1984.

677
Ellsworth, Oliver B. The Berkeley manuscript: University of California Music Library, MS. 744
(olim Phillipps 4450) : a new critical text and translation on facing pages. Lincoln:
University of Nebraska Press, 1984.

Fischer, Kurt von. Studien zur italienischen Musik des Trecento und frühen Quattrocento: das
Repertoire. Bern: Haupt, 1956.

Fox, Christopher. “A Darmstadt Diary.” Contact 29, (1985): 44-47.

_____. “Plural Darmstadt, the 1986 International Summer Course.” New Music 87, (1987): 102-
105.

_____. “British Music at Darmstadt 1982-90.” Tempo Sew Series, no. 186 (September 1993):
21-25.

_____. "Music as Fiction: A Consideration of the Work of Richard Barrett.” Contemporary

Music Review 13, no. 1 (1995): 147-157.

French secular music. Manuscript Chantilly, Musee Conde 564, ed. Gordon K. Greene, texts
edited by Terence Scully, 2 vols., Les Pemparts – Monaco, L’oiseau-lyre, 1982
(Polyphonic music of the fourteenth century, 18-19).

Frengel, Mike. “Polyrhythmic Modulation Studies.” Master of Arts thesis, Dartmouth College
1999.

Gallagher, Sean. Johannes Regis. Turnhout: Brepols, 2010.

Gallo, F. Alberto, "Die Notationslehre Im 14. Und 15. Jahrhundert." In Die Mittelalterliche
Lehre Von Der Mehrstimmigkeit, by Hans Heinrich Eggebrecht, 334-36. Darmstadt:
Wissenschaftliche Buchgesellschaft, 1984.

Gann, Kyle. "Conlon Nancarrow's Tempo Tornadoes," Village Voice, New York (5 October
1993), pp. 93 and 97

_____. "Downtown beats for the 1990s: Rhys Chatham, Mikel Rouse, Michael Gordon, Larry
Polansky, Ben Neill". Contemporary Music Review. 10, no. 1 (1994): 33-49.

_____. “Outside the Feedback Loop: A Nancarrow Keynote Address,” Music Theory Online 20,
no. 1 (March 2014): accessed November 4, 2017,
[Link]

_____. The music of Conlon Nancarrow. Cambridge University Press, 2006.

Gagne, Cole, and Tracy Caras. Soundpieces: Interviews with American Composers. Metuchen,
N.J: Scarecrow Press, 1982.

678
Gibbens, John Jeffrey. An Analysis of Ben Johnston's Sonata for Microtonal Piano. 1985.

Goldman, Richard Franko. 1957. "The Music of Elliott Carter". Musical Quarterly. 43: 151-170.

Grant, Roger Mathew. Beating Time & Measuring Music in the Early Modern Era. 2018.

_____. 2012. "Leonhard Euler's Unfinished Theory of Rhythm". Journal of Music Theory. 57,
no. 2: 245-286.

Greeson, Jim. Conlon Nancarrow Virtuoso of the Player Piano. Univ of Arkansas Pr, 2012.

Greitzer, Mary Lee T. Tormented Voices. PhD diss., Harvard University, 2007. Ann Arbor, MI:
UMI/ProQuest, 2007.

Grendler, Paul F. 2004. "The Universities of the Renaissance and Reformation". Renaissance
Quarterly. 57, no. 1: 1-42.

Haas, Max. 1982. "Studien zur mittelalterlichen Musiklehre I: eine Übersicht über die
Musiklehre im Kontext der Philosophie des 13. und frühen 14. Jahrhunderts". Forum
Musicologicum, Bd. 3. 323-427.

Hankel, Hermann. Zur geschichte der mathematik in alterthum und mittelalter (classic reprint).
[Place of publication not identified]: FORGOTTEN Books, 2016.

Hicks, Michael. 1993. "Cowell's Clusters". Musical Quarterly. 77, no. 3: 428-458.

_____. Henry Cowell, Bohemian. Urbana: University of Illinois Press, 2002.

Hodges, Nicolas, James Clarke, and Roger Redgate. Roger Redgate, James Clarke: works for
piano solo. Darnstadt: Coviello Classics, 2008.

Hoppin, Richard H. The Cypriot-French Repertory of the Manuscript Torino, Bibliotheca

Nazionale, [Link].9 Vol. 3, Vol. 3. The Cypriot-French Repertory of the Manuscript Torino,
Bibliotheca Nazionale, [Link].9. Rome: American Institute for Musicology, 1963.

Houle, George. Meter in Music, 1600-1800: Performance, Perception and Notation.

Bloomington, Ind: Indiana Univ. Press, 2000.

Hughes, Andrew. "Franco of Cologne." Grove Music Online. 2001; Accessed 27 Aug. 2020.
[Link]
.0001/omo-9781561592630-e-0000010138.

Josephson, Nors S. “Rodericus, ‘Angelorum Psalat’”, Musica disciplina, XXI, 1971, pp. 113-
126.

679
Juškevič, Adolf Pavlovič, Hans Wussing, and Viktor Ziegler. Geschichte der Mathematik im
Mittelalter. Basel: Pfalz-Verlag, 1966.

Key, Jordan Alexander. (2017, May 25). A Medieval Composer's Thoughts on Art & Pop
Culture: "Popular" (?) Music from 14th Century Italy: Part 3 - Francesco Landini (c.
1325/35 - 1397), Madrigal, "Musica Son - Già Furon - Ciascun Vuol."
[Link]
on-Medieval-Pop-Culture-Popular-Music-from-14th-Century-Italy-Part-3---Francesco-
Landini-c-132535---1397-Madrigal-Musica-Son---Già-Furon---Ciascun-Vuol

_____. "Euler, Cowell, Polyhedra And The Music Genome". Blog. The Neglected Composer,
2019. Accessed December 19, 2020. [Link]
post/2018/12/10/euler-cowell-polyhedra-and-the-music-genome-leonard-euler-e2-80-99s-
theory-of-music-and-i.

_____. (2016, September 24). Mathematical Medieval Music: "Popular" (?) Music from 14th
Century Italy: Part 1 - Johannes Ciconia (c. 1370 - 1412), Canon "Le ray au soleyl."
[Link]
from-14th-Century-Italy-Part-1---Johannes-Ciconia-c-1370---1412-Canon-Le-ray-au-
soleyl-Source-Mancini-CodexLucca-Codex-I-PEc-MS-3065-no-9-fol-LXXXIII

Lang, Paul Henry. Music in Western Civilization. New York: W.W. Norton and Co, 1941.

Langer, Susanne K. Feeling and Form, a Theory of Art Developed from Philosophy in a New
Key, by Susanne K. Langer. London: Routledge and Kegan Paul, 1953.

Lewin, David. 2006. Studies in Music with Text. Oxford: Oxford University Press.

Lichtenwanger, William. The Music of Henry Cowell: A Descriptive Catalog. Brooklyn: Institute
for Studies in American Music, 1986.

Long, Michael P. Musical tastes in fourteenth-century Italy notational styles, scholarly

traditions, and historical circumstances. 1981.

Macklin, Christopher. 2010. "Musical Irrationality in the Shadow of Pythagoras". Contemporary

Music Review. 29, no. 4: 387-393.

Marie, Jean Etienne. L homme musical. Arthabaska, 1976.

Marsh, Roger. 1994. "Heroic Motives. Roger Marsh Considers the Relation between Sign and
Sound in 'Complex' Music". The Musical Times. 135, no. 1812: 83-86.

Mead, Rita. Henry Cowell's New Music, 1925-1936. 1981.

680
_____. 2004. “‘One Man’s Signal is Another Man’s Noise’: Personal Encounters with Post-
Tonal Music.” In The Pleasure of Modernist Music: Listening, Meaning, Intention,
Ideology, ed. Arved Ashby. Rochester, New York: University of Rochester Press.

Menninger, Karl. Number Words and Number Symbols: A Cultural History of Numbers. Dover
Publications, 2013.

Miller, Clement A. 1970. "Early Gaffuriana: New Answers to Old Questions". The Musical
Quarterly. 56, no. 3: 367-388.

______. 1968. "Gaffurius's "Practica Musicae": Origin and Contents". Musica Disciplina. 22:
105-128.

Nádas, John L., and Agostino Ziino, eds. The Lucca Codex: Codice Mancini: Lucca, Archivio Di
Stato, Ms. 184, Perugia, Biblioteca Comunale "Augusta" Ms. 3065. Lucca: Libreria
Musicale Italiana, 1990.

Newes, Virginia. "Mensural Virtuosity in Non-Fugal Canons." In Canons and Canonic

Techniques, 14th - 16th Centuries: Theory, Practice, and Reception History; Proceedings
of the International Conference, Leuven, 4-6 October 2005, edited by Katelijne Schiltz
and Bonnie J. Blackburn, 19-46. Leuven: Peeters, 2007. 19-46.

Onians, John. Bearers of Meaning: The Classical Orders in Antiquity, the Middle Ages, and the
Renaissance. Princeton, N.J.: Princeton University Press, 1990.

Planchart, Alejandro Enrique. 1988. "Guillaume Du Fay's Benefices and His Relationship to the
Court of Burgundy". Early Music History. 8: 117-171.

Read, Gardner. Modern Rhythmic Notation. Bloomington: Indiana University Press, 2003.

_____. Source Book of Proposed Music Notation Reforms. New York: Greenwood Press, 1987.

Sachs, Curt. Rhythm and Tempo: A Study in Music History. New York, NY: Columbia
University Press, 1988.

Sartori, Claudio. Una redazione inedita del "Tractatus practice cantus mensurabilis ad modum
italicorum". Firenze: L.S. Olschki, 1936.

Schering, Arnold. Geschichte der Musik in Beispielen: dreihundertfüngzig Tonsätze aus neun
Jahrhunderten. Wiesbaden: Breitkopf & Härtel, 1959.

Schiff, David. "Carter, Elliott." Grove Music Online. Oxford Music Online. Oxford
University Press.

_____. The Music of Elliott Carter. London: Eulenburg Books, 1983.

681
Schmitt, Charles B. Aristotle and the Renaissance. Cambridge, Mass: Harvard University Press,
1983.

Schreur, Philip, trans. Tractatus Figurarum = Treatise on Noteshapes: A New Critical Text and
Translation. Edited by Thomas J. Mathiesen. Greek and Latin Music Theory. Lincoln:
Univ. of Nebraska Press, 1989.

Schuster-Craig, John. Schumann Encounters Berlioz: The Symphonie fantastique and

Schumann's String Quartets, Op. 41.” (2010). Faculty Scholarly Dissemination Grants.

Silverman, Julian. Review of Aspects of Complexity in Recent British Music. Tempo, New
Series 197 (July 1996): 33-37.

Smalley, Roger. 1972. "Avant-Garde Piano". The Musical Times. 113, no. 1558.

Smith, David Eugene. History of Mathematics. New York: Dover, 1990.

______. Rara Arithmetica: A Catalogve of the Arithmetics Written Before the Year Mdci, Part 2.
[Place of publication not identified]: Nabu Press, 2012.

Stoessel, Jason. "Con Lagreme Bagnandome El Viso: Mourning and Music in Late Medieval
Padua." Plainsong and Medieval Music 24, no. 01 (2015): 71-89.

______. (1999). Symbolic Innovation: The Notation of Jacob de Senleches. Acta Musicologica,
71(2), 136-164.

Strizich, Robert. 1982. "Notation in Elliott Carter's ‘Double Concerto’". Ex Tempore. 2, no. 1:
46-80.

Strunk, Oliver. Source Readings in Music History. New York: W.W. Norton, 1998.

Swetz, Frank J. Capitalism and Arithmetic: The New Math of the 15th Century ; Including the
Full Text of the Treviso Arithmetic of 1478. La Salle, Ill: Open Court, 1989.

Swift, Richard. 1976. “Some Aspects of Aggregate Composition.” Perspectives of New

Music 14, no. 2: 236–248.

Tanay, Dorit Esther. Noting music, marking culture: the intellectual context of rhythmic notation,
1250-1400. Holzgerlingen: American Institute of musicology, 1999.

Taruskin, Richard. Music in the Late Twentieth Century: The Oxford History of Western Music.
Oxford: Oxford University Press USA - OSO, 2011.

Thermos, Paul. Questionnaire response in Complexity in Music? An Inquiry into its Nature,
Motivation and Performability. Edited by Joel Bons. Netherlands: Job Press, 1990, 36-37.

682
Thoegersen, Peter Alexander. "Polytempic Polymicrotonal Music: 'a Road Less Traveled.'"
DMA Dissertation, University of Illinois at Urbana-Champaign 2012.

Thomas, Margaret Elida. 1997. Conlon Nancarrow's "Temporal dissonance": rhythmic and
textural stratification in the studies for player piano. Ann Arbor (MI): UMI.

_____. 2000. "Nancarrow's "Temporal Dissonance": Issues of Tempo Proportions, Metric

Synchrony, and Rhythmic Strategies". Intégral. 14/15: 137-180.

Toop Richard. 1998. “Four Facets of ‘The New Complexity’”. Contact 32.

_____. 2002. ""New Complexity" and After: A Personal Note". Polyphony & Complexity. 133-
135.

Truax, Barry. 1994. "The Inner and Outer Complexity of Music". Perspectives of New Music. 32,
no. 1: 176-193.

Ulman, Erik. 1994. "Some Thoughts on the New Complexity". Perspectives of New Music. 32,
no. 1: 202-206.

Von Gunden, Heidi. The Music of Ben Johnston. Metuchen, N.J.: Scarecrow, 1986.

Watkins, Glenn. "Chapter 25: The United States of America." In Soundings: Music in the
Twentieth Century, 528-41. New York: Schirmer Books, 1988.

Westergard, Peter. “Some Problems Raised by the Rhythmic Procedures in Milton Babbitt’s
Composition of Twelve Instruments,” Perspectives of New Music IV, no. 1 (Fall – Winter
1965)

Weisgall, Hugo. 1959. "The Music of Henry Cowell". The Musical Quarterly. 45, no. 4: 484-
507.

Woodley, Ronald. John Tucke: a case study in early Tudor music theory. Oxford: Clarendon
Press, 1993.

Yudkin, Jeremy, and Anonymous of St. Emmeram. 1990. De Music Mensurata: The Anonymous
Of St. Emmeram. Bloomington: Indiana University Press.

Young, Crawford. (2008). “Antiphone of the Angels: 'angelorum psalat tripudium'”. Recercare,
20(1-2), 5-23.

______. “A Collector’s Court.” Essay in accompanying booklet, Corps Femenin: L’avant-garde

de Jean Duc de Berry. 2010. Media SRL 551, 2011. Compact disc.

683
 BIOGRAPHICAL SKETCH

Jordan Alexander Key (b. 1990), blind, gay composer of contemporary and

classical music, received his Ph.D. from the University of Florida in 2021. He earned bachelor’s

degrees in music composition, mathematics, and philosophy from The College of Wooster

(2013), as well as a master’s degree in music composition from the University of Arizona

(2015). His significant recent projects include his Kennedy Center premier of his ballet, To Say

Pi; his residency at the Harn Museum of Art and the production of music for its various

exhibitions including his sextet, Verses from the Scroll of Sondering, and his Black-MIDI dance,

Nachi No Taki; his European premier with Vienna’s PHACE Ensemble performing his octet,

Threnody on the Death of Children; a performance by Boston String Quartet of his String

Quartet No. 1; a concert with Bold City Contemporary Ensemble performing both his trio March

for the 12th Hour, his quintet Discursus Anachronismus, and his sextet Be Bold; a collaboration

with the Vancouver Queer Arts Festival and Calliope’s Call for the premiers of his atheist songs,

God Ourselves and Last Night I Touched Him; and the display of his recent audio-visual projects

as part of the Wolfsburg Kunstmuseum’s new exhibit, Never Ending Stories: The Loop in Art,

Film, Architecture, and Music, in Germany. Jordan is also proficient on his two primary

instruments: the pipe organ and bagpipe. Jordan's interests in early music, bagpipe music, and

modern organ repertoire give his music a distinct contrapuntal and harmonic flare that is both

rhythmically diverse and melodically compelling.

Jordan, along with his husband, Jason Edmond Johnson (PhD in mathematics, University

of Florida, 2021) founded in 2020 a private, non-profit high school, The Aegis Institute, in

Gainesville, Florida: The Aegis Institute hopes to reimagine education in the 21st century and

provide the world with a new generation of critically minded, curious human beings.

684
 ProQuest Number: 28412151

INFORMATION TO ALL USERS

The quality and completeness of this reproduction is dependent on the quality
and completeness of the copy made available to ProQuest.

Distributed by ProQuest LLC ( 2022 ).

Copyright of the Dissertation is held by the Author unless otherwise noted.

This work may be used in accordance with the terms of the Creative Commons license
or other rights statement, as indicated in the copyright statement or in the metadata
associated with this work. Unless otherwise specified in the copyright statement
or the metadata, all rights are reserved by the copyright holder.

This work is protected against unauthorized copying under Title 17,

United States Code and other applicable copyright laws.

Microform Edition where available © ProQuest LLC. No reproduction or digitization

of the Microform Edition is authorized without permission of ProQuest LLC.

ProQuest LLC
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, MI 48106 - 1346 USA