1 a NEW MUSIC COMPOSITION TECHNIQUE USING NATURAL SCIENCE DATA D.M.A Document Presented in Partial Fulfillment of the Requiremen

Home , Algorithmic composition, Euclidean rhythm, Herma (Xenakis)

A NEW MUSIC COMPOSITION TECHNIQUE

USING NATURAL SCIENCE DATA

D.M.A Document

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Musical

Arts in the Graduate School of The Ohio State University

Joungmin Lee, B.A., M.M.

Graduate Program in Music

The Ohio State University

2019

D.M.A. Document Committee

Dr. Thomas Wells, Advisor

Dr. Jan Radzynski

Dr. Arved Ashby

Copyrighted by

Joungmin Lee

2019

ABSTRACT

The relationship of music and mathematics are well documented since the time of ancient Greece, and this relationship is evidenced in the mathematical or quasi- mathematical nature of compositional approaches by composers such as Xenakis,

Schoenberg, Charles Dodge, and composers who employ computer-assisted-composition techniques in their work. This study is an attempt to create a composition with data collected over the course 32 years from melting glaciers in seven areas in Greenland, and at the same time produce a work that is expressive and expands my compositional palette.

To begin with, numeric values from data were rounded to four-digits and converted into frequencies in Hz. Moreover, the other data are rounded to two-digit values that determine note durations. Using these transformations, a prototype composition was developed, with data from each of the seven Greenland-glacier areas used to compose individual instrument parts in a septet.

The composition Contrast and Conflict is a pilot study based on 20 data sets.

Serves as a practical example of the methods the author used to develop and transform data. One of the author’s significant findings is that data analysis, albeit sometimes painful and time-consuming, reduced his overall composing time. The variety and richness of data that exists from all academic areas and disciplines conceivably provide a rich reservoir of material from which to fashion compositions. As more composers ii explore this avenue of work, different methodologies will develop, and the value of works produced by this method will be evaluated.

iii

Dedicated to my wife, Hyejin

ACKNOWLEDGMENTS

I am heartily thankful to my advisor, Dr. Thomas Wells, who provided encouragement, guidance, and advice from beginning to end. Completing a DMA degree is a long journey, one which I would not have been able to achieve without his steadfast support and enthusiasm. I also would like to express my gratitude towards Dr. Jan

Radzynski and Dr. Arved Ashby for their outstanding teaching and guidance.

Furthermore I would like to express my gratitude to Dr. Seongsu Jeong who provided the scientific data for this study, and to my former mentors, Dr. David Gompper and Dr. Hyunsook Choi.

I want to thank my wife Hyejin Hong and our family for their support and patience. Most of all I wish to give thanks to God for my musical gift, opportunities, blessings, and everything He has done for me.

- Joungmin Lee

VITA

April 18, 1975 ………………………………Born Kwangju, South Korea

2001…………………………….……………B.A. Church Music,

Seoul Jangsin University, South Korea

2004……………………………….…………M.M. Computer Music,

SangMyung University, South Korea

2010…………………………………….……M.M. Music Technology,

New York University

2004……………………………………….…Lecturer of Computer Music/Recording,

Changwon College, South Korea

2002-2005……………………………………Vice Director of Image Enterprise Inc.

for Film, TV, Online Game, South Korea

2003-2005…………………………………....Lecturer of Composition/Computer Music

Seoul Artist Music School

Commercial Music Academy, South Korea

2010-2012……………………………………Lecturer of Composition/Computer Music

Apple Bridge Academy. Annandale, VA

2011-2013……………………………………Director of Music, Annandale United

Methodist Church, Annandale, VA vi

2013-2014……………………………………Director of Music, The Korean Choir of

Greater Kansas City, KS

2016-Present…………………………………Director of Music, Bethel International

United Methodist Church, Columbus, OH

PUBLICATIONS

2016………………………………... CD by Editor Sconfinarte at XXI Century Archives,

Milan string quartet ‘Vexatious’

2016…………………………………CD by the 31st volume Series of the SCI CD

for string quartet ‘Vexatious’

2016…………………………………CD by ABLAZE Records on Electronic Masters

Vol. 5 for 'Heterogeneous'

2016…………………………………CD by ABLAZE Records on Hong Kong New

Music Ensemble Live from Prague Vol. 1 for string

quartet ‘Vexatious’

2018………………………………….CD by ABLAZE Records Electronic, Masters

Series Vol. 7, ‘3 Sounds’ for fixed media

2019………………………………….Offered CD release by ABLAZE Records on

Pierrot Ensemble Series Vol. 4 for 'Abandoned'

vii

FIELDS OF STUDY

Major Field: Music

Studies in Music Composition and Electroacoustic Dr. Thomas Wells

Dr. Jan Radzynski

Studies in Choral Conducting Dr. Robert J. Ward

viii

TABLE OF CONTENTS

Abstract………………………………………………………………………….………...ii

Dedication…………………………………………………………………………….…..iv

Acknowledgements………………………………………………………………….…….v

Vita………………………………………………………………………………….…….vi

Publications………………………………………………………………………….…...vii

Fields of Study………………………………………………………………….….…... viii

Table of Contents………………………………………………………………………....ix

List of Tables…………………………………………………………….……………...xiv

List of Figures…………………………………………………………………………...xvi

List of Picture…………………………………………………………………………..xvii

List of Scores..………..………………………………………………………………..xviii

Chapter 1. Introduction……………………………………………………………………1

Chapter 2. Music and Mathematics: Their Relatedness…………………………………...3

Part 1. Introduction…………………………………….………………………...... ……3

Part 2. Philosophy in Music History…………………….…...………………………….3

1. Ancient Greece and the Idea of Music…………………………………….…...... 4

2. Medieval Philosophy………….…………………………………...…………...…..4

3. Premodern Music Philosophy………………………...……………...…………….5

3.1 Renaissance………………………………………...……………...…………...5

3.2 The 17th and 18th Centuries……………………………....……………….…..5

3.3 The 19th Century……………………………………….……….……………...6

3.4 Modern Music Philosophy……………………………….……….….……..…..7

Part 3. Overview other Works Related to Mathematics & Music………….….….….…8

1. Schillinger System of Musical Composition………………….……….……………8

1.1 Joseph Schillinger………………………………………………………………8

1.2 Joseph Schillinger and Scientific Principles……………………………………8

1.3 Inference………………………………………………………………………..9

2. Set Theory and Probability in Iannis Xenakis’s Herma…………………………..11

2.1 Iannis Xenakis………………………………………………………………...11

2.2 Herma…………………………………………………...…………………….12

Chapter 3. The Strengths and Weaknesses of the Use Scientific Data As a Compositional

Resource……………..…..……………………………………………………………….14

Part 1. Strengths………………………………………………..……...………………14

1. Productivity………………………………………………………...……………..14

2. Cross-Genre Collaboration………………………………………...….…………..15

3. Composition with Software………………………….…...…………...…………..15

Part 2. Weaknesses……………………………………….………………...………….15

Chapter 4. Discussion of Electroacoustic & Acoustic Works by using Natural Science Data

……………………………………………………………………………………………16

Part 1. Discussion, ‘Threatened By’, an Electroacoustic Work…………...……..….…16

1. Introduction…………………………………………………………...…………..16

2. Purpose…………….…………………………………………...…………………17

3. Discussion the Data in Graphics…….………………………...………………….17

3.1 The Velocity Data………………………...……………………...……..……..17

3.2 Discussion the Process of Music Composition………….…….………………19

3.3 Discussion about Composition Styles for Music inspired by Science Data

………………………………………………….…………………...... ………………….21

3.4 Discussion of Sine Waves……………………………………………………..21

4. Video……………………………….………...…………...………………………22

5. Discussion of Audience Reactions and Responses at the AGU Meeting………….23

6. Result………………………………………………...……………………....……24

Part 2. Analyses, “Contrast & Conflict”, a New Acoustic Work………….…….…….24

1. Introduction……………………………………………………………...….…….24

2. Purpose……………..…………………………………...... ………………………25

3. Method………..………………………………………………...……………...…26

3.1 Overview - Data Analysis………….………………………………………….26

3.2 Data for Helheim Area………………….……………………………………..26

3.3 Humboldt Area…………………………………….…………………....…….31

3.4 Jakobshavn Area………………………………….……………...……………37

3.5 Kangerdlugssuaq Area……………………………...…….…………………..40

3.6 Kongoscar Area………………………………….……….……………...……43

3.7 Petermann Area………………………………….…………………...……….47

3.8 Upernavik Area………………………………….……………………...…….51

4. Discussion………………..…………………..……………………………...……54

4.1 Enharmonicity and the Autonomy of the Octave……….……………………..54

4.2 Atonal (in C)……………….……………………………………...…………..54

4.3 The Length of Notes……….…………………………………………...……..55

Chapter 5. Composition, Contrast & Conflict, a New Acoustic Work……………………56

1. Autonomy of the Octave…….…………………………………………………….56

2. Repetitive High-Pitch Sound……………..………….………………...………….57

3. Enharmonicity as Cure for Monotony……………..………………………………57

4. Use of Complex Rhythms………………..…………………………..…………...57

5. Performance Levels…………..……………………………………...……………57

6. Shortening Composition Time……………..………………………..……………58

7. Liquidity of Data Analysis………………..……………………………...………..58

8. Philosophy…………………..…………………………………………………….58

Chapter 6. Conclusions…………………………………………………………………...61

Bibliography…………………………………………………………………………….. 63

xii

LIST OF TABLES

Table 1. Base frequency and the frequency variation used for each glacier………..……19

Table 2. a: Value before round-off; b: value after post-decimal point round-off for the

Helheim area……………….…………...…………………………………………..……27

Table 3. Frequencies of equal temperament……….……………………………….……27

Table 4. Frequency-to-musical note conversion for the Helheim area…………………..28

Table 5. Length of notes for the Helheim area………………………………..…………30

Table 6. Result of the pitch and length for the Helheim area…………..……..…………..31

Table 7. a: Value before round-off; b: value after post-decimal point round-off for the

Humboldt area…………………………………………………………….………..…….32

Table 8. Frequency to musical note conversion for the Humboldt area…………………33

Table 9. Rounded-off numbers for the Humboldt area……………………….………….34

Table 10. A result of the pitch and length for the Humboldt area…………………………36

Table 11. a: Value before round-off; b: value after post-decimal point round-off for the

Jakobshavn area………………………………………………………………………….37

Table 12. Frequency to musical note conversion in the Jakobshavn area ……………….38

Table 13. Result of the pitch and length for the Jakobshavn area………………….……39

Table 14. a: Value before round-off; b: value after post-decimal point round-off for the

Kangerdlugssuaq area……………………………………………………………………40 xiii

Table 15. Frequency to musical note c conversion for the Kangerdlugssuaq area………..41

Table 16. A result of the pitch and length for the Kangerdlugssuaq area…………………42

Table 17. a: Value before round-off; b: value after post-decimal point round-off for the

Kongoscar area…………………………………………………………………….……..43

Table 18. Frequency to musical note conversion for the Kongoscar area…………...……44

Table 19. A result of the pitch and length for the Kongoscar area……………………....46

Table 20. a: Value before round-off; b: value after post-decimal point round-off for the

Petermann area…………………………………………………………………………...47

Table 21. Frequency to musical note conversion for the Petermann area……………….48

Table 22. A result of the pitch and length for the Petermann area……………………....50

Table 23. a: Value before round-off; b: value after post-decimal point round-off for the

Upernavik area…………………………………………………………………………...51

Table 24. Frequency to musical note conversion for the Upernavik area……………...…52

Table 25. A result of the pitch and length for the Upernavik area……………….………..53

xiv

LIST OF FIGURES

Figure 1. 4:3 interference pattern………………………………………………………….9

Figure 2. Graphic notation where each horizontal square represents a 16th-note and each vertical square represents a semitone………………………………………………….…10

Figure 3. Pitch occurrence frequencies within specific pitch sets articulated in Herma

Bullets indicate notes which did not occur in the given section….……………………….11

Figure 4. (left): Locations of the outlet glaciers whose time series speed are using for the work. InSAR velocity mosaic (Joughin et al., 2017) with InSAR velocity mosaic as the base map. The surface flow speed is color-coded in log scale as shown in the color bar on the right. (right): Time series plot of the rate of the seven glaciers (right)………………...18

LIST OF PICTURES

Picture 1. Threatened By, electroacoustic composition, track view in Logic Pro

X……………………………………………………………………………...…………..20

xvi

LIST OF SCORES

Score 1. Score excerpt from Herma showing the last four measures of the exposition of set an indicating clangs (circled) and sequences (tied)………………………………………13

Score 2. Joungmin Lee: Contrast & Conflict …………………………………………….60

xvii

Chapter 1. Introduction

It is well-documented in studies and research that music and science are closely related in both method and structure. This bond has inspired increasingly large numbers of composers to use mathematical operations such as set theory in their compositional process. However, these past endeavors fall short in addressing the complexities of contemporary music, with its interest in multidimensional parameters of music such as timbre and pitch density.

This study will present a new compositional procedure based on the use of scientific data from the field of earth studies. It will describe a method that will make use of 32 years of seismic data of melting icebergs in seven areas of Greenland as material for a work for ensemble. These data were first employed in a piece of music in 2017 in an electroacoustic composition, Threatened By, written in collaboration with an earth- studies scientist as part of a presentation to highlight concerns about the impact of global warming and climate change in Greenland. This represented an unusual and insightful use of music to underscore an increasingly serious example of global warming.

The study will show how the abovementioned data used in the composition of an electroacoustic piece is applied to the composition of an instrumental ensemble work.

The homogeneity and heterogeneity of the two compositions will be compared and

1 analyzed, and the wealth of possibilities for employing scientific data in music composition will be addressed.

Scientific data was measured and obtained from seven regions of Greenland, and each region’s data is applied to one of seven instruments: flute, oboe, clarinet, bassoon, violin, cello, and piano.

The speed (Y-axis) showing the melt amount in m/yr, will be a resource for pitch determination.. Moreover, the time (mm/yyyy, X-axis) will be a resource for the note lengths.

Chapter 2. Music and Mathematics: Their Relatedness

Part 1. Introduction

Music and mathematics have evolved together throughout time. Relationships between the two were investigated by the ancient Greeks, with the writings of Plato and contributions of Pythagoras, Euclid, and others being well known. More recently, the musician Vincenzo Galilei1, A. Helden, Galileo Galilei’s father, explored the relationship of string tension to pitch. (Helden 2005)1. Mathematical concepts provide the basis for further examination of musical phenomena, through the work of Mersenne, Bernoulli,

Euler, Fourier, Ohm, Helmholz, to the present day.

Part 2. Philosophy in Music History

At various points in the history of music, attempts were made to fuse mathematics and sciences with music. What brought them together was philosophy. Science is an

1 Vincenzo Galilei (3 April 1520, c. 1520, or late 1520s – buried 1 or 2 July 1591) was an Italian lutenist, composer, and music theorist, and the father of astronomer and physicist Galileo Galilei and the lute virtuoso and composer Michelagnolo Galilei. He was a seminal figure in the musical life of the late Renaissance and contributed significantly to the musical revolution which demarcates the beginning of the Baroque era. 3 endeavor to find and organize objective and rational knowledge in a particular area.

Philosophy seeks a universality that systemically brings together all elements. Let us examine how music’s relations with mathematics, science, and philosophy played out in each period.

1. Ancient Greece and the Idea of Music

The mathematical idea of music, the founding theory for music, first emerged with Pythagoras in 5th and 6th centuries B.C. He believed that the world is a harmonious cosmos that can be explained objectively through universality. His view of music as an influencer of personality formation and as an educational tool was further developed by

Plato and articulated by Aristotle. His view that music has ethical powers, and can affect character and influence emotion was further developed by Plato and Aristotle.

2. Medieval Philosophy

Pythagoras’ universal music idea and Plato’s abstract view of music took root in the religious idea or divinity during the medieval era. The music of the era is characterized by a critical fusion of ancient Greek ideas and Christianity. The Greek concept of harmony and Pythagoras’ mathematical concepts2, by way of Platonian music ideas, were merged to create an abstraction of the relationship between God, the soul and the universe. However, the inaudible divine music of the medieval era began to decline

2 Stanford Encyclopedia of Philosophy, The Philosophy of Music, First published Mon Oct 22, 2007; substantive revision Tue Jul 11, 2017. 4 by the late medieval era. The concept of music became limited to audible musical phenomenon. Inspired by Aristotle, J. Grocheo3 defined the essence of music as a movement4, the one captured by senses and understood audibly. He delinked music from mathematical concepts and instead defined it as harmonious sound tailored to singing. He defined acoustic arts as an academic field.

3. Premodern Music Philosophy

3.1 Renaissance

Renaissance music, a continuation of Pythagoras’ mathematical music idea, made a break with the medieval preoccupation with theological functionality and the

Pythagorean-influenced study of proportions and, and in the spirit of humanistic philosophy, emphasized melodic beauty, expressiveness and unity. Music came to be seen as a branch of the fine arts.

3.2 The 17th and 18th Centuries

Continuing the humanistic view of music of the Renaissance, questions of musical aesthetics began to shift to the experiential—to the perception of music’s beauty. In the

18th century, the rivalry between empiricism and rationalism centered on the issue of

3 John Haines and Patricia DeWitt, Johannes De Grocheio and Aristotellan Natural Philosophy, Cambridge University Press, 2008. 4 Philipp Brüllmann, Music Builds Character Aristotle, Politics VIII 5, 1340a14–b5, De Gruyter, 2013. 5 objectivity and certainty. A.G. Baumgarten5, of 1714-1762, a disciple of Gottfried

Wilhelm Leibniz6, reformulated aesthetics as logic from a rationalist perspective. He defined aesthetics as a deduction of principles of artistic or natural beauty from individual taste. He understood beauty as the completeness of emotional awareness. Completeness referred to a complete harmony between content and form. In this way, he put the emphasis more on emotional intuition or sense than rational insight. In the 17th and 18th centuries, phenomena were perceived conceptually. In understanding music, clarity and judgment have become more critical than abstraction. The rise of instrumental music made an expression of emotions by the late 18th century.

3.3 The 19th Century

The 19th century was a time of extravagant claims about the expressive power of music.

Whereas Kant, working mainly in the previous century, argued that instrumental music was a less effective genre, E. T. A. Hoffman and Schopenhauer argued the opposite: that instrumental music is alone in its ability to describe the metaphysical organization of reality. This concept also touched on the modern idea of music visualization—using

5 German philosopher Alexander Gottlieb Baumgarten was born in Frankfurt (Oder), Brandenburg. He famously introduced the current definition of the philosophical discipline of aesthetics in his Halle master’s thesis when he was only twenty-one years of age. He called this epistêmê aisthetikê, or the science of what is sensed and imagined (Baumgarten, Meditationes, §CXVI, pp. 86–7). https://artlark.org/2018/05/26/a-g- baumgarten-the-man-who-invented-aesthetics/. 6 Gottfried Wilhelm (von) Leibniz German: 1 July 1646 [O.S. 21 June] – 14 November 1716) was a prominent German (of Slavic origin) polymath and philosopher in the history of mathematics and the history of philosophy. His most notable accomplishment was conceiving the ideas of differential and integral calculus, independently of Isaac Newton's contemporaneous developments. 6 music to alter the perception of film or graphic images. E. Hanslick7, of 1825-1904, believed that music could not constitute an representational expression of emotion and, could not express ideas, emotions, images, or a literary narrative plot. Hanslick’s view helped to inspire Hugo Riemann8 to devise a general theory to explain all aspects of music.

3.4 Modern Music Philosophy

In the 20th century, it was B. Croce9, of 1866-1952, that initiated a debate on philosophy and arts. He divided the spirit into the theoretical and the practical. The theoretical division can be split between aesthetic and logic while the practical division included economics, or study of all practical matter, and ethics.

In music, the concept of Beautiful Art began to decline and be replaced by a new concept that underscores the transformation away from the idea that music is music idea as the expression of emotion. The new concept held that music does not have to be

7 Hanslick was born in Prague (then in the Austrian Empire), the son of Joseph Adolph Hanslik, a bibliographer and music teacher from a German-speaking family, and one of his piano pupils, the daughter of a Jewish merchant from Vienna. At the age of eighteen Hanslick went to study music with Václav Tomášek, one of Prague's renowned musicians. https://en.wikipedia.org/wiki/Eduard_Hanslick 8 Karl Wilhelm Julius Hugo Riemann (18 July 1849 – 10 July 1919) was a German music theorist and composer who was born at Grossmehlra, Schwarzburg-Sondershausen. His first musical training came from his father Robert Riemann, a land owner, bailiff and, to judge from locally surviving listings of his songs and choral works, an active music enthusiast. Hugo Riemann was educated by Heinrich Frankenberger, the Sondershausen Choir Master, in Music theory. He was taught the piano by August Barthel and Theodor Ratzenberger (who had once studied under Liszt). 9 Croce, Benedetto (1866 - 1952), italian philosopher, historian and critic. In its original and most influential formulation Croce's aesthetic theory is part of a general philosophy of civilization (largely derived from Vico and Hegel). Croce's view is both ‘idealist’ and ‘historicist’. His idealism is evident when he poses a strong contrast between ‘intuition’ and ‘intellect’, and argues that art is ‘intuition. 7 harmonious or beautiful. “Theoretical” music is not necessarily beautiful but could be stimulating and has its way of achieving aesthetic values.

Part 3. Overview other Works Related to Mathematics & Music

The study will discuss the author’s use of scientific data for music composition.

To lay the groundwork for this presentation and contextualize it, two other examples of incorporation of scientific aspects into music composition—the Schillinger system, and

Xenakis’s piano work Herma--will be discussed.

1. Schillinger System of Musical Composition

1.1 Joseph Schillinger

Joseph Schillinger (1895-1943) was a composer, music theorist, and pedagogue who, in developing a system of composition purportedly based on mathematical and scientific bases. His system, devised in the 1920s and published in 1946 lays out his method in two volumes comprising 1690 pages.

1.2 Joseph Schillinger and Scientific Principles

The system of musical composition developed by Joseph Schillinger in the 1920s-

1930s is widely held as one of the first attempts at musical composition based on scientific principles. His system has 12 books, ranging from composition to rhythmic

8 structures. His structure has inspired many composers such as George Gershwin, Glenn

Miller, Benny Goodman, Oscar Levant, and many others, solidifying its status in the modern history of music composition.

Figure 1. 4:3 interference pattern10

1.3 Inference

Almost every aspect of Joseph Schillinger's 'System' is derived in some fashion from resultants of interference of simple periodic motions11. He injected these resultants into rhythms and structures proportionally into scales and chords, counterpoint, harmonic progression, orchestration. Moreover, the process retains and even amplifies emotional and semantic aspects of musical composition. This was possible because he could turn interferences into regular patterns11.

10 Schillinger, Joseph (1946). Schillinger System of Musical Composition, Volume I, New York, C. Fischer, Inc., 1946, p. 8, Figure 15 11 Bruno Degazio, The Schillinger System of Musical Composition and Contemporary Computer Music, Sheridan College (Oakville), 1986 9

To address the limited availability of rhythmic patterns from relatively small data sets, he devised ways to extend and vary them. Among them, fractioning is standout because it is used to extend the scope of these rhythmic inferences. Also, Schillinger continued to expand his notion of inference-driven pattern into developing the contrapuntal composition11.

His contribution to musical composition can be highlighted by the following:

First, he established the first-ever principles that present all possible generators between two and nine. He built an analytical paradigm that group these possibilities by family. In many ways, his system of composition a prototype of computer music as he developed a form of algorithmic composition which included such concepts as graphic notation, projection of musical parameters, rule-bound methods, self-similarity, and imitative natural dynamics11.

Figure 2. Graphic notation where each horizontal square represents a 16th-note and each vertical square represents a semitone12

12 Schillinger, Joseph (1946). Schillinger System of Musical Composition, Volume I, New York, C. Fischer, Inc., 1946, p. XIX 10

2. Set Theory and Probability in Iannis Xenakis’s Herma

2.1 Iannis Xenakis

Iannis Xenakis, the Greek-French composer, often used highly sophisticated scientific and mathematical theories to arrive at the music of primitive power. Messiaen connects the subjects in 1700AD-1800AD; musical forms such as the fugue and sonata appeared while mathematicians invented complex numbers and group structures as well as defined continuity (Messiaen 1994). Both of these developments involve groupings, structures, and inventions that are almost identical, yet in different subjects. Continuity was essential to Xenakis, so much so, that he graphically represented continuity with the metaphor of the musical glissando13.

Figure 3. Pitch occurrence frequencies within specific pitch sets articulated in Herma Bullets indicate notes which did not occur in the given section14

13 Robert A. Wannamaker, Structure and Perception in Herma by Iannis Xenakis, School of Fine Art and Music, University of Guelph, 2000

2.2 Herma

Herma (1961), or “bond” in English, for piano, is Iannis Xenakis’s first composition for a solo instrument, commissioned in 1961 by Japanese pianist Yuji

Takahashi. Herma is characterized by complex rhythmic figures and huge leaps that still require perfect evenness of articulation. In a good performance, the effort is repaid by the creation of a sense of seething, amorphous energy and a powerful forward momentum15.

In his book Formalized Music, Xenakis laid out theoretical discussions in connection with Herma’s structure. He developed a model based on specific mathematical relationships between certain pitch sets. The composer adopts a universal pitch set. In Herma, the pitch set is forged in-time as a sequence of pitches derived from the set. However, each pitch is drawn randomly from the given sets without register preference. The use of the stochastic procedure is aimed at avoiding melodic or harmonic patterns that would be used to distract listeners from the perception of the pitch sets.

Score1 shows a typical excerpt from the score: the final four measures of the exposition of set A, where “linear” and “cloud” versions of the set are simultaneously articulated. Circles indicate clangs as I typically hear them, with sequences of clangs connected by ties13.

15 Robert A. Wannamaker, Structure and Perception in Herma by Iannis Xenakis, The Online Journal of the Society for Music Theory, Volume 7, Number 3, May 2001. 12

Score 1. Score excerpt from Herma showing the last four measures of the exposition of set an indicating clangs (circled) and sequences (tied) 13

Chapter 3. The Strengths and Weaknesses of the Use Scientific Data

As a Compositional Resource

Part 1. Strengths

1. Productivity

One strength of the data-driven composition is its ability to save time and labor for composers. Data can help streamline years of composition and arrangement into days.

One time-consuming aspect of the data-driven process is choosing which data set should be employed. There is some similarity between Arnold Schoenberg’s 12-tone system16 of composition and a usual data-driven method. What sets these two processes apart is the data-driven process’s complexity that requires more delicacy.

16 Deborah H. How, Arnold Schoenberg’s prelude from the suite for piano, op. 25: from composition with twelve tones to the twelve-tone method, A Dissertation Presented to the faculty of the graduate school University of Southern California in partial fulfillment of the requirements for the degree doctor of philosophy (historical musicology), 2009

2. Cross-Genre Collaboration

Data-driven composition has opened a path for interdisciplinary- and cross-genre collaboration. Data-driven composition can facilitate the integration of multimedia works. For instance, an ocean wave-based composition could be integrated with a video work based on ocean waves.

3. Composition with Software

Analysis of data can influence harmony, rhythm and other musical elements, all which can be translated and be reinterpreted into music scores under the control of the composer with- or without assistance from software.

Part 2. Weaknesses

A certain familiarity with mathematical and scientific processes is necessary to engage in this type of composition. In some cases, the composer may need to work with a collaborator knowledgeable in data science. It is not always easy to access data suitable for data-driven composition. Some data may require consent and release to use.

Chapter 4. The Composition of Electroacoustic and Acoustic Works

Using Natural-Science Data:

Part 1. Discussion, ‘Threatened By’, an Electroacoustic Work

1. Introduction

In the eyes of many, the words “glaciers” and “contemporary classical music” bear little association. The commonality between the two is probably “losing something irreplaceable”: One is diminishing in physical mass; The other in popularity and mass appeal. The author was the beneficiary of a project by The American Geophysical Union17 to attempt a collaboration between art and environmentalism at its Fall 2017 meeting18, where he was selected as a composer to make a poster presentation.

17 AGU galvanizes a community of Earth and space scientists that collaboratively advances and communicates science and its power to ensure a sustainable future. https://sites.agu.org/leadership/strategic-plan/. 18 AGU meeting in the world was in New Orleans, Louisiana, Dec 11 - 15, 2017.

2. Purpose

This paper aims to help to inspire future music work in data-driven composition by providing an analysis of the composition process and the application of science data used in the composition of two of my original works: one electroacoustic; the other acoustic. The article will provide data in graphic form and analyze the process of music composition and video production. Further, it will discuss audience reactions and responses at the AGU meeting.

3. Discussion the Data in Graphics

3.1 The Velocity Data

The surface velocities of the flow are measured by a tracking technique embedded with MIMC2 software19. Figure 1. shows the revised version of the multiple-image multiple-chip algorithm by Ahn and Howat (2011), further edited by Dr. Seongsu Jeong20 et al. (2017) by incorporating quadra matching, dynamic linear constraint, and pseudo smoothing21.

19 Seongsu Jeong, Ian M. Howat, Yushin Ahn, 2017, Improved Multiple Matching Method for Observing Glacier Motion With Repeat Image Feature Tracking, IEEE Transactions on Geoscience and Remote Sensing, 55(4), pp. 2431-2441 20 Dr. Seongsu Jeong: Currently Assistant Project Scientist at University of California, Irvine 21 Seongsu Jeong, Ian M. Howat, and Yushin Ahn, Improved Multiple Matching Method for Observing Glacier Motion With Repeat Image Feature Tracking, IEEE Transactions on Geoscience and Remote Sensing, Vol. 55, No. 4, April 2017 17

Figure 4. (left): Locations of the outlet glaciers whose time series speed are using for the work. InSAR velocity mosaic (Joughin et al., 2017) with InSAR velocity mosaic as the base map22. The surface flow speed is color-coded in log scale as shown in the color bar on the right. (right): Time series plot of the rate of the seven glaciers (right).

The velocity measurement in this project comes from optical satellite images, including Landsat 4, 5, 7, 8, and ASTER. Because there is no sunlight in polar winter, no visual images are obtained during this season. As a result, the lack of data has caused gaps in the winters in the time series. In the case of Humboldt, however, year-long temporal baselines have been used to measure annual velocities, whose timestamp occurs in winter. The readers have to keep in mind that winter velocity results in annual speed.

The speed data are interpolated using inverse distance weight to define velocity at a particular time (v(t)) by inserting the velocity from the time series data. The speeds of

each site are then converted to frequency (fv), where v is speed, vm and vM are minimum

22 Heming Liao, Franz J. Meyer, Bernd Scheuchl, Jeremie Mouginot, Ian Joughin, Eric Rignot, ‘Ionospheric correction of InSAR data for accurate ice velocity measurement at polar regions’, Remote Sensing of Environment Volume 209, May 2018, Pages 166-180 18 and maximum of the speed in the timeline, fb and fV are base frequencies and its variation as presented in Table 1. The converted frequency is then utilized to generate sine waveforms y(t), defined as:

Site Base frequency Frequency variation (Hz) (Hz) Helheim 110 55 Humboldt 220 110 Jakobshavn 440 220 Kanderdlugssuaq 880 440 Kongoscar 1760 880 Petermann 3520 1760 Upernavik 7040 7040

Table 1. Base frequency and the frequency variation used for each glacier.

3.2 Discussion the Process of Music Composition

The work, titled Threatened By, is an electroacoustic music composition. It is integrated with data of the volume of time that each glacier melts in Greenland’s seven regions over 32 years between 1985 and 2017. The data were recorded in Matlab23 and transformed into sine waves, which was saved in wave format. The file was imported in

Logic Pro X24 and integrated into electroacoustic music25. Downloaded from www.soundsnap.com, the sound of melting glaciers and icebergs was reprocessed and

23 MATLAB is a multi-paradigm numerical computing environment and proprietary programming language developed by MathWorks. 24 Logic Pro is a digital audio workstation and MIDI sequencer software application for the macOS platform. It was originally created in the early 1990s as Notator Logic, or Logic, by German software developer C-Lab, later Emagic. 25 Electroacoustic music is a style of Western art music which originated around the middle of the 20th century, following the incorporation of electric sound production into compositional practice. 19 integrated. The sound of a steaming rice cooker was also applied. With a jazz drum brush, scratching and beating were generated musically. The sound from beans in a plate, pills in a drug container, steel and wood were recording by a Zoom H6 Handy Recorder26.

Sound Forge software was used to edit the sound samples which were integrated as a musical work by Logic Pro X. Along the way, EQ, reverberation, distortion, delay, and reversal were applied to modify and manipulate acoustic images in correspondence with the ups and downs of the data shown in the graphs. Chromax Special Filter and

Bufconvolve of Max/MSP27 were used to manipulate acoustic images.

Picture 1. Threatened By, electroacoustic composition, track view in Logic Pro X

26 24-bit/96kHz, 6-in/2-out Modular Field Recording System and USB Audio Interface with 4 x XLR/TRS Combination Inputs, 4 x Microphone Preamps, and Included XY and Mid-Side Microphone Attachments. 27 Max, also known as Max/MSP/Jitter, is a visual programming language for music and multimedia developed and maintained by San Francisco-based software company Cycling '74. 20

3.3 Discussion Composition Styles for Music inspired by Science Data

To emphasize one of the essential composition methods, good sound sources alone cannot be turned into good music without the composer’s attention and style. All music work is subjective, driven by its composer, which makes it harder to assess objectively. However, analysis of the method of the background of a music piece will create some degree of evaluation possible. The same is true of MUDIV28 electroacoustic music, whose assessment is based on the analysis of creativity and the quality and diversity of sound sources. Threatened By was made with these metrics in mind. To unleash creativity, the arrangement and motions of sound sources were organized. To offer better sound quality, high-quality recording, 48Kh, 92Kh, or above, was needed.

Finally, the sense and perception of music vary individually. However, the format and development of music work can create an impact on vibration. Electronic music often brings into question the need for such structures and formality.

3.4 Discussion of Sine Waves

Sine wave29, a concept developed by French mathematician Joseph Fourier to describe smooth repetitive oscillations, is in extensive use in some fields including music.

Many composers have been employed the idea to write electronic music pieces.

However, the approach is now being used less because it has been used to the point of

28 MuDiv is a tool for verifying concurrent systems. 29 A sine wave or sinusoid is a mathematical curve that describes a smooth periodic oscillation. A sine wave is a continuous wave. It is named after the function sine, of which it is the graph. It occurs often in pure and applied mathematics, as well as physics, engineering, signal processing and many other fields. 21 becoming a banality. What is more, musicians can now work with more sophisticated forms of software to meet their needs for new ideas and experiments. Indeed, fewer and fewer musicians depend on the sine wave. I based Threatened By on the absolute value determined by scientific data, which prompted to generate sine waves from data recorded in Matlab. The waves are integral to the main sound. The sound sampling was sine waves. Diminishing sine waves in use, sine waves are still capable of precisely addressing data. Alternatives fall short of adequately converting data into the art form.

The sound from Matlab can become good, sampling after volume and wave adjustment.

Matlab’s graphs framed numerical data in music scores. The sampling further unleashed musical inspirations to me.

In some discussions, Dr. Seongsu Jeong made some suggestions. Glaciers, formed during the ice age, began to melt because of global warming and such simple pollutants as industrial and urban noises and even hairspray. Melting glaciers are a sensitive issue subject to more research, according to Dr. Jeong, who advised me cautiously to apply data and sound to my work. I want to emphasize the fact that Threatened By is a collaborative project.

4. Video

About 32-years’ worth of satellite photographs of Greenland was used in the production of video portions of this work. The images were edited in animation fashion to show changes in glaciers in real time. This attempt ran into technical difficulties. At least seven areas were to be integrating into the video. However, I lacked the skills.

About 200 gigabytes of satellite images were too large to be processed in tiff format by

Adobe Premier CC 2017. An expanded version of the software could turn them into

Photoshop, only at the considerable cost of time. After consultation with Dr. Jeong, I decided to depict one area, Jakobshavn. However, the issue has not been thoroughly addressed. Multi-images can be rendered, but expansions remain unsolved because the city has more than 1,000 images.

These difficulties led me to use PicGIF Lite30 to create animated images. The app allows only 300 copies to edit into animation, but this solution enabled me to further explore other technical possibilities. The app was adequate at the speed of animated images. All these resulted to create a poster presentation at the AGU Fall 2017 meeting.

5. Discuss Audience Reactions and Responses at the AGU Meeting

The AGU was uncharted territory for me. I was the only artist at that get-together who was to give a poster presentation. After setting up for performance, I was bombarded with questions by the scientists and artists. They seemed to have little background with music, but after my brief explanations, they listened to the music and responded to it well—because they understood the context. They asked me a variety of questions: “How did you apply data to the music?” “Is the work part of my dissertation,” etc. One biologist instantly proposed a joint project. However, performed on a laptop computer, the video and listening were not fully appreciated. Also, the brightness of the interior made it

30 PicGIF Lite is a free software only available for Mac, belonging to the category Design & Photography. 23 difficult to understand the video images fully. The poster could have been expanded to include a flow chart to show the entire production process.

Overall, it was a meaningful endeavor because I could confirm and even pique interest in music among those who otherwise would not be interested in it. I would like to know more about how scientific data can be musically employed and used, which has encouraged me to study further ways to integrate data into music proactively.

6. Results

The entire process was not natural, but I stayed positive throughout. Also, it was a pleasant surprise to find shared experiences between music and science—which further unlocked the potential of music for me. I believe similar methods can be applyed to efforts to link music with construction engineering, philosophy, futurology, and psychology.

Part 2 – Analyses, “Contrast & Conflict”, a New Acoustic Work

1. Introduction

This section introduces an acoustic piece based on scientific data. The same data will be applied to a contemporary music composition—a task requiring greater detail because each data subset is assigned to a specific pitch. It takes more analysis of data to determine whether it fits in with acoustic music. Apart from the melody, rhythm, and

24 harmony, notes will be assigned by the composer. This way, the creator’s emotion and ideas can inform the work and be inspired by the data.

Using science data measured in the seven regions of Greenland for 32 years, pitch and length of the note are easily assigned. Additionally, the seven areas are assigned to each of the instruments within the ensemble: flute, oboe, clarinet, bassoon, violin, cello, and piano. There are only two measurements of the amount of glacier melting, time and melting speed. The rate (Y-axis) shows how many meters per year melts in m / yr, which will be the melody of the three elements of music. And the time (mm / yyyy, X-axis) will be reflected by rhythm. The length of the notes are calculated using two measurement distances. Finally, the melody and rhythm of each region (musical instrument) are established, and harmony will automatically appear when you insert it into the music program.

All of this is unexpected, and some sounds or results may appear. In some ways, it is similar to John Cage's Chance Music, but it differs by using objective data. For these, a small ensemble is composed by using scientific data then analyzed.

2. Purpose

This paper is an attempt to shed light on a new data-driven method of music composition to enhance the scope of musical creativity and productivity.

3. Method

3.1 Overview - Data Analysis

Central to data analysis is the length of notes. The longer the note, the shorter the numbers become. A quarter note is longer than an eighth-note although four is less than eight. All these complicate the analysis process. Therefore, it is needed to assess in a retroactive way, showing that the numbers in data are in line with the length of notes.

However, this method can only be used when the length of notes and the numbers in data do not correspond with each other. Namely, the number 16 is identical to a 16th note. The number 32 is similar to a 32nd note.

3.2 Data for Helheim area

Let us put these into a spreadsheet:

Column A refers to time

Column B refers to velocity (m/yr)

Column C refers to daily changes in sattelite images

The numbers in these columns are too detailed to be relevant (Table 2: A). The numbers have to be rounded out after the decimal point (Table 2: B). Meanwhile, Column

A shows numbers that register little change in length. It is difficult to convert into music, unless otherwise into a quiet, tranquil composition.

Let us look at Column B (Table 2: B). Compared with Column A, Column B fluctuates in numbers within the audible range of frequency (20h-20kh). These numbers

26 are appropriate to convert into melody. However, some numbers are in the higher range of audibility. The highest frequency of the piano is 4186.00, compared with 4858 in the first row of Column B. However, the harmonics of the violin can accommodate this number. The following table shows the approximate values measured in piano keys.

A B C A B C 1 725089.562 4858.19877 31.9999306 1 725090 4858 32 2 725106.558 5834.75636 15.9999421 2 725107 5835 16 3 725129.562 4757.69794 47.9998495 3 725130 4758 48 4 725129.563 4815.48023 47.9998495 4 725130 4815 48 5 725138.558 4498.91418 47.9998843 5 725139 4499 48 6 725161.563 5008.25345 15.9999884 6 725162 5008 16 7 725166.06 5442.73808 24.9956829 7 725166 5443 25 8 725166.06 6416.53886 7.00428241 8 725166 6417 7 9 725170.558 6201.2257 15.9999769 9 725171 6201 16 10 725174.06 6442.4183 8.99569444 10 725174 6442 9 11 725176.567 4984.5126 31.999919 11 725177 4985 32 12 725177.562 5139.39887 47.9998727 12 725178 5139 48 13 725177.563 5166.94794 47.9998727 13 725178 5167 48 14 725178.558 5068.98451 31.9999074 14 725179 5069 32 15 725182.06 4676.20272 24.995625 15 725182 4676 25 16 725184.567 5317.83911 47.9998495 16 725185 5318 48 17 725186.558 4077.92379 15.9999306 17 725187 4078 16 18 725190.06 4477.67241 23.0041898 18 725190 4478 23 19 725197.064 5297.60197 8.9956713 19 725197 5298 9 20 725200.566 5588.51861 15.9999306 20 725201 5589 16 a b Table 2. a: Value before round-off; b: value after post-decimal point round-off For the Helheim area

Octave 0 1 2 3 4 5 6 7 8 9 10 C/B# 16.352 32.703 65.406 130.813 261.626 523.251 1046.502 2093.005 4186.009 8372.018 16744.036

C#/Db 17.324 34.648 69.296 138.591 277.183 554.365 1108.731 2217.461 4434.922 8869.844 17739.688 D 18.354 36.708 73.416 146.833 293.665 587.330 1174.659 2349.318 4698.636 9397.273 18794.545 D#/Eb 19.445 38.891 77.782 155.563 311.127 622.254 1244.508 2489.016 4987.032 9956.063 19912.127 E/Fb 20.602 41.203 82.407 164.814 329.628 659.355 1318.501 2637.020 5274.041 10548.082 - F/E# 21.827 43.654 87.307 174.614 349.228 698.456 1396.913 2793.826 5587.652 11175.303 - F#/Gb 23.125 46.249 92.499 184.997 369.994 739.989 1479.978 2959.955 5919.911 11839.822 - G 24.500 48.999 97.999 195.998 391.995 783.991 1567.982 3135.963 6271.927 12543.854 - G#/Ab 25.957 51.913 103.826 207.652 415.305 830.609 1661.219 3322.438 6644.875 13289.750 - A 27.500 55.000 110.000 220.000 440.000 880.000 1760.000 3520.000 7040.000 14080.000 - A#/Bb 29.135 58.270 116.541 233.082 446.164 932.328 1864.655 3729.310 7458.620 14917.240 - B/Cb 30.868 61.735 123.471 246.942 493.883 987.767 1975.533 3951.066 7902.133 15804.266

Table 3. Frequencies of equal temperament 27

The first row of Column B shows 4858, which is approximate to D#/Eb8,

4987.032. The first row of Column B matches D#/Eb8. The value of the second row is

5835, approximate to F#/Gb8, 5919.911.

The first 20 rows of Column B can be showing as the following:

Column B Frequency Note Offset cents 1 4858 4978.032 D#8 -42.25 2 5835 5919.911 F#8 -25 3 4758 4698.636 D 8 21.73 4 4815 4698.636 D 8 42.35 5 4499 4434.922 C#8 24.83 6 5008 4978.032 D#8 10.39 7 5443 5587.652 F 8 -45.40 8 6417 6271.927 G 8 39.59 9 6201 6271.927 G 8 -19.68 10 6442 6271.927 G 8 46.32 11 4985 4978.032 D#8 2.42 12 5139 5274.041 E 8 -44.90 13 5167 5274.041 E 8 -35.49 14 5069 4978.032 D#8 31.35 15 4676 4698.636 D 8 -8.35 16 5318 5274.041 E 8 14.37 17 4078 4186.009 C 8 -45.25 18 4478 4434.922 C#8 16.73 19 5298 5274.041 E 8 7.84 20 5589 5587.652 F 8 0.41

Table 4. Frequency to musical note convert for the Helheim area

Table 4 shows a frequency range of C8-G8 that can produce a melody be expressed by the violin. It should be noted that this is a resource for the melody, not the melody itself. The numbers in Column C are smaller than Columns A and B (Table 2).

These values can be converted into notes and length of the note. The first two rows of

Column C are identical to 32nd and 16th notes. The third row has 48, which cannot be 28 expressed with a musical note. However, 48 can break down to 32 plus 16. Accordingly,

48 can assign to a .

The 7th row of Column C has 25, which is bigger than 16 and smaller than 32.

Also, 25 can break down to 16 plus 9. The number 16 can be converted into a 16th note. In turn, 9 can break down to 8 plus 1. The number 16 can be converted into two 8th notes

( ) plus 1. Since none can assign to 1, the composer arbitrarily assigns a note to it.

Therefore, 1 is assigned to a 32nd note and a half. And hence: 1= a 64th note ( ). To recap,

25= + or .

The eighth row has 7, which can break down to 4 plus 3 or a 4th note ( ) plus 3.

The 3 can be three-fourths or . Accordingly, the number 7 can be expressed in + .

However, it should be written as .

Finally, the tenth row has 9 or 8 plus one, which is + 1. The number 1 can be converted into 1/8 (0.125). The value of 0.125 is too short to convert into a musical note.

For extending the length, 9 can be converted into + . The 18th row shows 23, or 16

( ) + 7. The 16 can halve as two 8s. The remainder, 7 can approximately turn into a 16th note ( ) and 23( ).

Data ‘C’ 7 9 16 23 25 32 48 Notes +

Table 5. Length of notes for the Helheim area

The spreadsheet of data of Helheim shows a total of 863 values. In Column C, the numbers can be enumerated as the following: 7, 8, 9, 10, 11, 14, 15, 16, 17, 18, 22, 23,

24, 25, 26, 27, 31, 32, and 48. Out of the 863, 19 values are overlapping, which is so small enough to convert into musical notes.

As a result of the pitch and length can be organized as the following:

B C Note Lengths 1 4858 32 D#8

2 5835 16 F#8

3 4758 48 D 8

4 4815 48 D 8

5 4499 48 C#8

6 5008 16 D#8

7 5443 25 F 8

8 6417 7 G 8

9 6201 16 G 8

10 6442 9 G 8 + 11 4985 32 D#8

12 5139 48 E 8

13 5167 48 E 8

14 5069 32 D#8

15 4676 25 D 8

16 5318 48 E 8

17 4078 16 C 8

18 4478 23 C#8

19 5298 9 E 8 + 20 5589 16 F 8

Table 6. Result of the pitch and length for the Helheim area

3.3 Humboldt Area

Next, I analyzed the Humboldt area. The following two tables show the values and the rounded off values. (Table 7) The Humboldt area posts less value than the 31

Helheim area. However, it provides a frequency range that can accommodate all primary instruments such as flute, oboe, clarinet, bassoon, violin, cello, and piano. Among them, a piece will be composed for the clarinet.

A B C A B C 1 725118.71 632.592623 15.999919 1 725119 633 16 2 725134.71 275.007481 15.9999421 2 725135 275 16 3 725166.71 415.640159 47.9999306 3 725167 416 48 4 725206.71 753.189267 31.999838 4 725207 753 32 5 725230.709 528.751823 15.9998843 5 725231 529 16 6 725302.708 452.181774 383.996319 6 725303 452 384 7 725306.21 453.669602 376.99206 7 725306 454 377 8 725310.708 453.327642 399.99603 8 725311 453 400 9 725310.708 441.282652 367.9964 9 725311 441 368 10 725314.21 454.756892 407.000174 10 725314 455 407 11 725314.21 454.084887 392.991771 11 725314 454 393 12 725314.21 455.123857 360.992153 12 725314 455 361 13 725317.712 455.714102 399.995914 13 725318 456 400 14 725318.708 438.151604 383.996111 14 725319 438 384 15 725322.21 452.302372 376.991863 15 725322 452 377 16 725325.712 447.510003 415.99559 16 725326 448 416 17 725325.712 452.736593 383.996007 17 725326 453 384 18 725326.708 440.60392 335.996447 18 725327 441 336 19 725330.21 450.234907 328.992164 19 725330 450 329 20 725333.712 460.49631 431.995231 20 725334 460 432

a b Table 7. a: Value before round-off; b: value after post-decimal point round-off For the Humboldt area

The following table shows pitches converted from Table 3. Frequencies of the equal temperament.

B Frequency Note Offset cents 1 633 622.254 D#5 29.64 2 275 277.183 C#4 -13.68 3 416 415.305 G#4 2.89 4 753 739.989 F#5 30.17 5 529 523.251 C 5 18.91 6 452 440.000 A 4 46.58 7 454 466.164 A#4 -45.77 8 453 466.164 A#4 -49.58 9 441 440.000 A 4 3.93 10 455 466.164 A#4 -41.96 11 454 466.164 A#4 -45.77 12 455 466.164 A#4 -41.96 13 456 466.164 A#4 -38.16 14 438 440.000 A 4 -7.88 15 452 440.000 A 4 46.58 16 448 440.000 A 4 31.19 17 453 466.164 A#4 -49.58 18 441 440.000 A 4 3.93 19 450 440.000 A 4 38.90 20 460 466.164 A#4 -23.04

Table 8. Frequency to musical note convert for the Humboldt area

Let us analyze Column C (Table 9), in which values will be converted into the lengths of musical notes. A close look at the data turned up 16, 32, and 48 in rows 1-5— which is similar to the data of the Helheim area. However, after these rows, values change substantially, to the point of requiring modification.

A B C 1 725119 633 16 2 725135 275 16 3 725167 416 48 4 725207 753 32 5 725231 529 16 6 725303 452 384 7 725306 454 377 8 725311 453 400 9 725311 441 368 10 725314 455 407 11 725314 454 393 12 725314 455 361 13 725318 456 400 14 725319 438 384 15 725322 452 377 16 725326 448 416 17 725326 453 384 18 725327 441 336 19 725330 450 329 20 725334 460 432

Table 9. Rounded-off numbers for the Humboldt area

The sixth row of Column has 384, the value that has to reduce. The lowest number in Column C is 16, which will be used to divide values into all columns. The results after rounding out after the decimal point are 0, 0, 3, 2, 0, 24, 24, 25, 23, 25, 25,

23, 25, 24, 24, 26, 24, 21, 21, 27.

The above numbers represent the lengths of musical notes. With 0s converted into full notes, the following are the conversions.

The number 3 is assigned to three-four time. The number 2 is assigning to two- four time. The conversions are the following:

, , , , , 24, 24, 25, 23, 25, 25, 23, 25, 24, 24, 26, 24, 21, 21, 27

The remaining two numbers, 23 and 25 are converted into and similar to the data for the Helheim area. To recap, , , , , , 24, 24, , , , ,

, , 24, 24, 26, 24, 21, 21, 27.

Let us convert 21, 24, 26 and 27. The number 21 equals 16 plus 5. The number 16 is converted into a whole note. The remainder of 5 is assigned to a quintuplet: +

The number 24 equals 16 plus 8. The number 8 is half 16, which can be converted into a thirty-second note: + =

The number 26 can break down to 16 plus 10, and 10 can break down to 5 plus 5.

A quintuplet and a thirty-second note can be assigned: + + .

The last number is 27 or 16 plus 11, which in turn breaks down to 16+8+3. The first two numbers can be converted into , and 3 can be converted into a triplet. The number 27 can be + .

All in all: , , , , , , , , , , , , , ,

, + , , + , + , + .

As a result, pitch and length can be organized as follows:

B C Note Lengths 1 633 16 (0) D#5 2 275 16 (0) C#4 3 416 48 (3) G#4

4 753 32 (2) F#5

5 529 16 (0) C 5 6 452 384 (24) A 4

7 454 377 (24) A#4

8 453 400 (25) A#4

9 441 368 (23) A 4

10 455 407 (25) A#4

11 454 393 (25) A#4

12 455 361 (23) A#4

13 456 400 (25) A#4

14 438 384 (24) A 4

15 452 377 (24) A 4

16 448 416 (26) A 4 + + 17 453 384 (24) A#4

18 441 336 (21) A 4 + 19 450 329 (21) A 4 + 20 460 432 (27) A#4 +

Table 10. A result of the pitch and length for the Humboldt area

3.4 Jakobshavn Area

Now, let us observe the data of the Jakobshavn area. Column B of Table 12b has

2000 Hz - 3000 Hz, which can be accommodated by the flute, which ranges from 250 to

2500 in frequency with the fundamental frequency of 1375. One octave will be lowered to better accommodate the flute—in the same manner, that lowered an octave for the flute.

A B C A B C 1 725103.605 3568.39099 15.9999421 1 725104 3568 16 2 725111.605 3406.63417 31.9998611 2 725112 3407 32 3 725119.605 3353.21991 15.999919 3 725120 3353 16 4 725123.107 3053.47261 23.0041667 4 725123 3053 23 5 725131.107 2369.18967 7.00424769 5 725131 2369 7 6 725175.605 3712.45064 17.9913657 6 725176 3712 18 7 725176.6 3284.6347 15.9999537 7 725177 3285 16 8 725188.102 3524.70425 7.00425926 8 725188 3525 7 9 725192.6 3674.70412 15.9999306 9 725193 3675 16 10 725200.6 3581.41523 31.9998495 10 725201 3581 32 11 725208.6 3632.33078 15.999919 11 725209 3632 16 12 725232.6 3388.00735 31.9997685 12 725233 3388 32 13 725245.098 2272.39614 7.00423611 13 725245 2272 7 14 725249.596 3304.82982 15.9998611 14 725250 3305 16 15 725253.098 3948.46221 8.995625 15 725253 3948 9 16 725261.098 3617.02821 24.9954861 16 725261 3617 25 17 725263.604 2696.78511 30.0083333 17 725264 2697 30 18 725265.596 3166.11125 15.9998611 18 725266 3166 16 19 725471.601 3062.85829 15.9997569 19 725472 3063 16 20 725477.095 3834.09488 8.99555556 20 725477 3834 9

a b Table 11. a: Value before round-off, b: value after post-decimal point round-off for Jakobshavn area

The following Table 12 shows pitches converted from Table 3. Frequencies of the equal temperament.

B Frequency Note Offset cents 1 3568 3520.000 A 7 23.45 2 3407 3322.438 G#7 43.51 3 3353 3322.438 G#7 15.85 4 3053 3135.963 G 7 -46.41 5 2369 2349.318 D 7 14.44 6 3712 3729.310 A#7 -8.05 7 3285 3322.438 G#7 -19.61 8 3525 3520.000 A 7 2.45 9 3675 3729.310 A#7 -25.39 10 3581 3520.000 A 7 29.74 11 3632 3729.310 A#7 -45.77 12 3388 3322.438 G#7 33.83 13 2272 2217.461 C#7 42.06 14 3305 3322.438 G#7 -9.10 15 3948 3951.066 B7 -1.34 16 3617 3520.000 A 7 47.06 17 2697 2637.020 E 7 38.93 18 3166 3135.963 G 7 16.50 19 3063 3135.963 G 7 -40.75 20 3834 3729.310 A#7 47.93

Table 12. Frequency to musical note conversion in the Jakobshavn area

Column C of Table 11b shows values that vary in length, which can be translated into the following: 16= , 32 = , 23= , 7= , 9= , 25= .

This leaves us with 18 and 30. Let us break down the remaining numbers. First,

18=16+2 can be converted into + 2. As noted earlier, composers can assign value to approximates. As such, 18 can be converted into + after 2 assigned with a half note.

Moreover, finally, let us analyze 30 which can turn into 16 +14, which can break down to +14. The value “14” can translate into 8+4. To recap, 30= + +4, with 4 converted into . And hence, 30= + + . –as shown in the following table.

B C Note Lengths 1 3568 16 A 7

2 3407 32 G#7

3 3353 16 G#7

4 3053 23 G 7

5 2369 7 D 7

6 3712 18 A#7 + 7 3285 16 G#7

8 3525 7 A 7

9 3675 16 A#7

10 3581 32 A 7

11 3632 16 A#7

12 3388 32 G#7

13 2272 7 C#7

14 3305 16 G#7

15 3948 9 B7

16 3617 25 A 7

17 2697 30 E 7

18 3166 16 G 7

19 3063 16 G 7

20 3834 9 A#7

Table 13. Result of the pitch and length for the Jakobshavn area 39

3.5 Kangerdlugssuaq Area

Data for the Kangerdlugssuaq area in Column B of Table 14. There is a frequency level of 2200 Hz ~ 4200 Hz, which can be adjusted downward octave-wise to accommodate the piano, the instrument with 4186.009 in frequency.

A B C A B C 1 725090.558 4217.55167 15.9999653 1 725091 4218 16 2 725106.558 3003.99259 15.9999421 2 725107 3004 16 3 725108.549 3395.18316 15.9999421 3 725109 3395 16 4 725138.558 3839.22815 47.9998843 4 725139 3839 48 5 725140.549 3621.91473 47.9998843 5 725141 3622 48 6 725156.549 2271.71146 15.9999884 6 725157 2272 16 7 725161.562 3007.29895 15.9999884 7 725162 3007 16 8 725161.562 3202.55256 15.9999884 8 725162 3203 16 9 725170.558 3602.21488 15.9999769 9 725171 3602 16 10 725172.549 3841.62538 15.9999769 10 725173 3842 16 11 725186.558 3753.00446 47.999838 11 725187 3753 48 12 725186.558 3843.77567 15.9999306 12 725187 3844 16 13 725193.562 3641.15571 15.999919 13 725194 3641 16 14 725193.562 3638.06548 15.999919 14 725194 3638 16 15 725202.558 3725.82844 15.9999306 15 725203 3726 16 16 725218.558 3290.74197 15.9998958 16 725219 3291 16 17 725220.549 3791.5748 15.9999074 17 725221 3792 16 18 725234.557 3670.8166 47.9996412 18 725235 3671 48 19 725234.557 4055.54757 15.9998843 19 725235 4056 16 20 725236.549 3744.25998 15.9998727 20 725237 3744 16

a b Table 14. a: Value before round-off; b: value after post-decimal point round-off for the Kangerdlugssuaq area

The following Table 15 shows pitches converted from Table 3. Frequencies of the equal temperament.

B Frequency Note Offset cents 1 4218 4186.009 C 8 13.18 2 3004 2959.955 F#7 25.57 3 3395 3322.438 G#7 37.40 4 3839 3951.066 B7 -49.81 5 3622 3520.000 A 7 49.45 6 2272 2217.461 C#7 42.06 7 3007 2959.955 F#7 27.30 8 3203 3135.963 G 7 36.62 9 3602 3520.000 A 7 39.86 10 3842 3951.066 B7 -48.45 11 3753 3729.310 A#7 10.96 12 3844 3951.066 B7 -47.55 13 3641 3729.310 A#7 -41.48 14 3638 3729.310 A#7 -42.91 15 3727 3729.310 A#7 -1.07 16 3291 3322.438 G#7 -16.45 17 3792 3729.310 A#7 28.86 18 3671 3729.310 A#7 -27.28 19 4056 3951.066 B7 45.38 20 3744 3729.310 A#7 6.80

Table 15. Frequency to musical note convert for the Kangerdlugssuaq area

Column C of Table14b has 16 and 48. As broken down earlier, 16 equals and 48 -

-as shown in the following table.

B C Note Lengths 1 4218 16 C 8

2 3004 16 F#7

3 3395 16 G#7

4 3839 48 B7

5 3622 48 A 7

6 2272 16 C#7

7 3007 16 F#7

8 3203 16 G 7

9 3602 16 A 7

10 3842 16 B7

11 3753 48 A#7

12 3844 16 B7

13 3641 16 A#7

14 3638 16 A#7

15 3727 16 A#7

16 3291 16 G#7

17 3792 16 A#7

18 3671 48 A#7

19 4056 16 B7

20 3744 16 A#7

Table 16. A result of the pitch and length for the Kangerdlugssuaq area

3.6 Kongoscar Area

Let us now study the data of Kongoscar area. Column B of Table 17 shows

2000Hz-3000Hz, higher for the cello which necessitates the active being adjusted.

A B C A B C 1 730598.706 2680.31888 14.0371759 1 730599 2680 14 2 730616.683 2672.83531 49.9911227 2 730617 2673 50 3 730654.176 2593.22599 24.9955671 3 730654 2593 25 4 730662.176 2715.68416 40.9954167 4 730662 2716 41 5 730668.185 2812.07633 53.0124074 5 730668 2812 53 6 730669.18 2746.50584 55.0038194 6 730669 2747 55 7 730670.176 2541.85311 56.9952199 7 730670 2542 57 8 730674.674 2860.4093 15.9998495 8 730675 2860 16 9 730680.682 2979.32689 28.0168403 9 730681 2979 28 10 730681.678 2968.31212 30.0082523 10 730682 2968 30 11 730682.674 2447.6768 31.9996528 11 730683 2448 32 12 730686.176 2909.76102 39.0038773 12 730686 2910 39 13 730689.678 3031.76573 14.0084028 13 730690 3032 14 14 730690.674 2131.17444 15.9998032 14 730691 2131 16 15 730694.176 2899.562 23.0040278 15 730694 2900 23 16 730989.192 2698.22116 23.0298264 16 730989 2698 23 17 730993.69 2667.15646 14.0342361 17 730994 2667 14 18 730998.175 2712.20222 40.9953819 18 730998 2712 41 19 731001.677 2564.14949 47.9995718 19 731002 2564 48 20 731002.673 3027.00636 31.9997917 20 731003 3027 32

a b Table 17. a: Value before round-off; b: value after post-decimal point round-off For the Kongoscar area

The following Table 18 shows pitches converted from Table 3. Frequencies of the equal temperament.

B Frequency Note Offset cents 1 2680 2637.020 E 7 27.99 2 2673 2637.020 E 7 23.46 3 2593 2637.020 E 7 -29.14 4 2716 2793.826 F 7 -48.90 5 2812 2793.826 F 7 11.22 6 2747 2793.826 F 7 -29.26 7 2542 2489.016 D#7 36.46 8 2860 2793.826 F 7 40.52 9 2979 2959.955 F#7 11.10 10 2968 2959.955 F#7 4.70 11 2448 2489.016 D#7 -28.76 12 2910 2959.955 F#7 -29.46 13 3032 2959.955 F#7 41.63 14 2131 2093.005 C 7 31.14 15 2900 2959.955 F#7 -35.42 16 2698 2637.020 E 7 39.58 17 2667 2637.020 E 7 19.57 18 2712 2637.020 E 7 48.54 19 2564 2637.020 E 7 -48.61 20 3027 2959.955 F#7 38.77

Table 18. Frequency to musical note convert for the Kongoscar area

Let us determine the length of the musical notes. Exclusive of such overlapping numbers as 16, 23, 25, 30, 32 and 48, there are 14, 50, 41, 53, 55, 57, 28, 30 and 39.

The number 14 breaks down to 8+4, or +4. Since 4is a , 14 can be converted into + . The number 50 equals 32 plus 18, or +18. As shown earlier, 18 becomes

16 plus 2, or + . The number 50 can break down to + + , or + ,

44 depending on the preference of a composer. The number 41 is 32 plus 9, or + , which translates into .

The number 53 is 32 plus 21 which is + . The 53 can break down to

+ + or, it is + . The number 55 equals 32 plus 23 ( ), and therefore + , or . The number 57 is 32+25 ( ). It is + or + . The 28 equals 16+12 (or, 8 plus 4) And hence: 28 equals + + , or The number 30 is

16+14 ( +8+6) or + + 4+2, or + + + .

Finally, the number 39 equals 32+7 ( ), or + . Alternatively, it is + .

To recap in the following table:

B C Note Lengths 1 2680 14 E 7 + 2 2673 50 E 7 + + or + 3 2593 25 E 7

4 2716 41 F 7

5 2812 53 F 7 + 6 2747 55 F 7 + or 7 2542 57 D#7 + or + 8 2860 16 F 7

9 2979 28 F#7 + + or 10 2968 30 F#7 + + + 11 2448 32 D#7

12 2910 39 F#7 + or + 13 3032 14 F#7 + 14 2131 16 C 7

15 2900 23 F#7

16 2698 23 E 7

17 2667 14 E 7 + 18 2712 41 E 7

19 2564 48 E 7

20 3027 32 F#7

Table 19. A result of the pitch and length for the Kongoscar area 46

3.7 Petermann Area

The Petermann area posted a frequency level of 1000 Hz-1500 Hz that can be accommodated by the oboe.

A B C A B C 1 725221.714 1138.13249 15.9998958 1 725222 1138 16 2 725525.71 1200.43829 15.9996644 2 725526 1200 16 3 726581.713 1023.04835 15.9998032 3 726582 1023 16 4 726629.712 1123.5082 15.9996991 4 726630 1124 16 5 726997.707 1435.01097 15.9999884 5 726998 1435 16 6 728437.706 1131.53304 15.9997917 6 728438 1132 16 7 728453.706 1135.69425 47.999294 7 728454 1136 48 8 728453.706 1302.47587 15.9997801 8 728454 1302 16 9 728477.706 1058.31681 31.9994329 9 728478 1058 32 10 728517.705 1188.85902 47.9991204 10 728518 1189 48 11 730312.751 1208.22244 17.9914583 11 730313 1208 18 12 730618.872 1148.8779 37.9740394 12 730619 1149 38 13 730619.353 1135.16626 37.0121296 13 730619 1135 37 14 730619.387 1068.46694 39.0037847 14 730619 1068 39 15 730619.868 1028.17425 39.9655671 15 730620 1028 40 16 730619.868 1106.10235 38.041875 16 730620 1106 38 17 730620.349 1087.33674 39.0036574 17 730620 1087 39 18 730622.408 1054.51042 32.9612269 18 730622 1055 33 19 730622.889 1006.75672 33.9230093 19 730623 1007 34 20 730631.404 1064.73624 63.0376042 20 730631 1065 63 a b Table 20. a: Value before round-off; b: value after post-decimal point round-off For the Petermann area

The following table shows pitches converted from Table 3. Frequencies of the equal temperament.

B Frequency Note Offset cents 1 1138 1108.731 C#6 45.11 2 1200 1174.659 D 6 36.95 3 1023 1046.502 C 6 -39.32 4 1124 1108.731 C#6 23.68 5 1435 1396.913 F 6 46.57 6 1132 1108.731 C#6 35.96 7 1136 1108.731 C#6 42.06 8 1302 1318.510 E 6 -21.81 9 1058 1046.502 C 6 18.91 10 1189 1174.659 D 6 21.01 11 1208 1174.659 D 6 48.45 12 1149 1174.659 D 6 -38.23 13 1135 1108.731 C#6 40.54 14 1068 1046.502 C 6 35.20 15 1028 1046.502 C 6 -30.88 16 1106 1108.731 C#6 -4.26 17 1087 1108.731 C#6 -34.26 18 1055 1046.502 C 6 14 19 1007 987.767 B5 33.38 20 1065 1046.502 C 6 30.33

Table 21. Frequency to musical note convert for the Petermann area

With the numbers broken down earlier excluded, there are 37, 38, 33, 34, and 63.

Let us break down each value. The number 32 equals 32 ( ) + 5 ( ). The number 38 can break down to 32 plus 6, with the latter turning into 4 plus 2. It, in turn, can be rewritten as + + , or + The number 33 is 32 plus 1, or + . The number 34, or 32 plus 2, can be converting into + . And the number 63 is 32 ( ) plus 48

31, with the latter being 16( ) plus15. The number 15 can break down to 8 ( ) + [ 4 (

) + 3]. The number 3 equals 2 ( ) + 1 ( ). And hence, 63= + + + + + , the most complex conversion so far.

The following table shows the conversions as mentioned above.

B C Note Lengths 1 1138 16 C#6

2 1200 16 D 6

3 1023 16 C 6

4 1124 16 C#6

5 1435 16 F 6

6 1132 16 C#6

7 1136 48 C#6

8 1302 16 E 6

9 1058 32 C 6

10 1189 48 D 6

11 1208 18 D 6 + 12 1149 38 D 6 + + or + 13 1135 37 C#6 + 14 1068 39 C 6 + or + 15 1028 40 C 6 + 16 1106 38 C#6 + + or + 17 1087 39 C#6 + or + 18 1055 33 C 6 + 19 1007 34 B5 + 20 1065 63 C 6 + + + + + or + +

Table 22. A result of the pitch and length for the Petermann area

3.8 Upernavik Area

Finally, analysis of the data of Upernavik area turned up a frequency level of

2700 Hz – 3200 Hz, necessitating being adjusted for an octave to accommodate the

Bassoon.

A B C A B C 1 725097.63 2894.55595 15.9999653 1 725098 2895 16 2 725117.132 2773.61639 23.0041782 2 725117 2774 23 3 725125.132 2711.08681 39.0041319 3 725125 2711 39 4 725132.136 2930.91425 7.00424769 4 725132 2931 7 5 725136.634 3156.74613 15.9999537 5 725137 3157 16 6 725140.136 3094.14801 8.99570602 6 725140 3094 9 7 725144.634 2849.54579 31.9999537 7 725145 2850 32 8 725148.136 2865.7924 24.995706 8 725148 2866 25 9 725148.136 2825.52707 39.0042361 9 725148 2826 39 10 725149.132 2884.96931 40.9956597 10 725149 2885 41 11 725151.638 2824.53891 31.9999884 11 725152 2825 32 12 725152.634 2923.22593 47.9999306 12 725153 2923 48 13 725168.634 3066.13083 15.9999769 13 725169 3066 16 14 725172.136 3247.07147 8.99569444 14 725172 3247 9 15 725173.132 2876.9461 7.00427083 15 725173 2877 7 16 725174.128 3570.10526 8.99569444 16 725174 3570 9 17 725176.634 2988.15687 31.999919 17 725177 2988 32 18 725182.128 3260.09511 24.9956366 18 725182 3260 25 19 725184.634 3048.72196 15.9999421 19 725185 3049 16 20 725186.626 3058.7449 15.9999421 20 725187 3059 16

a b Table 23. a: Value before round-off; b: value after post-decimal point round-off For the Upernavik area

The following table shows pitches converted from Table 3. Frequencies of the equal temperament.

B Frequency Note Offset cents 1 2895 2959.955 F#7 -38.41 2 2774 2793.826 F 7 -12.32 3 2711 2637.020 E 7 47.90 4 2931 2959.955 F#7 -17.01 5 3157 3135.963 G 7 11.57 6 3094 3135.963 G 7 -23.32 7 2850 2793.826 F 7 34.46 8 2866 2793.826 F 7 44.15 9 2826 2793.826 F 7 19.82 10 2885 2959.955 F#7 -44.40 11 2825 2793.826 F 7 19.21 12 2923 2959.955 F#7 -21.74 13 3066 3135.963 G 7 -39.05 14 3247 3322.438 G#7 -39.75 15 2877 2959.955 F#7 -49.21 16 3570 3520.000 A 7 24.42 17 2988 2959.955 F#7 16.32 18 3260 3322.438 G#7 -32.84 19 3049 3135.963 G 7 -48.68 20 3059 3135.963 G 7 -43.01

Table 24. Frequency to musical note convert for the Upernavik area

Column C of Table 23 showed re-arranged values from Column B, which in turn are translated into musical notes.

B C Note Lengths 1 2895 16 F#7

2 2774 23 F 7

3 2711 39 E 7 + or + 4 2931 7 F#7

5 3157 16 G 7

6 3094 9 G 7

7 2850 32 F 7

8 2866 25 F 7

9 2826 39 F 7 + or + 10 2885 41 F#7

11 2825 32 F 7

12 2923 48 F#7

13 3066 16 G 7

14 3247 9 G#7

15 2877 7 F#7

16 3570 9 A 7

17 2988 32 F#7

18 3260 25 G#7

19 3049 16 G 7

20 3059 16 G 7

Table 25. A result of the pitch and length for the Upernavik area

4. Discussion

So far, methods to compose contemporary music pieces with scientific data have been surveyed, which can lay out the following points.

4.1 Enharmonicity and the Autonomy of the Octave

There are three types of glacier data.

- Column A refers to the time

- Column B refers to velocity (m/yr)

- Column C refers to daily changes in satellite imaging.

m/yr. in Column B translated into Hz to better facilitate the conversion of the values into musical notes. The conversion is based on the proximity between frequency levels and musical values, with the composers’ latitude over approximate conversion.

The same metrics can be applyed to octave conversions. Without adjustment for the octave, some high frequencies exceed the frequency range of the instrument in question.

With the adjustment, the composer will have a broader range of frequency and instrumental choices.

4.2 Atonal (in C)

Ata-driven compositions do not produce keys or scores for transposition instruments. The only instrument discussed in this paper is Clarinet in Bb. Usually, for

54 transposition instruments, a score is produced, with M2 higher. However, by incorporating atonality, all instruments can be performed without a part score.

4.3 The Length of Notes

The length can be converted or arbitrarily determined. For instance, 55 can translate into x + y or a combined eighth note.

In this paper, simplified numbers are used to determine the lengths. However, the method should improve to better determine the length or the inverse relations between the length of a musical note and the value of data.

Chapter 5. Composition, Contrast & Conflict, a New Acoustic Work

In Chapter 5, the data as mentioned earlier will be used to compose a contemporary music piece, ‘Contract & Conflict,’ for flute, oboe, clarinet, bassoon, violin, cello, and piano. This atonal piece consists of 20 notes, with a variety of arbitrary articulations. In this way, artistic aspects remain intact while the work is basing on scientific data.

The final goal of this paper, data-driven music composition, has culminated in the following points.

1. Autonomy of the Octave

Each data set has its characteristics and patterns. However, data of melting glaciers are grounded in pitches and have relatively high-frequency levels. Unmodified, these frequencies will limit the kinds of instruments to be played because only high- frequency instruments can play the composition. This raises the need for the downward adjustment of frequencies because high-frequency instruments can exclusively play

56 music. As for the composition in question, the octave was lowered to accommodate low- frequency instruments, with the pitch unchanged.

2. Repetitive High-Pitch Sound

There are incidences where an instrument repeats the same sound, based or not on the composer’s intention. However, such repetitiveness can be effectively mitigated by adjusting the octave downward or upward.

3. Enharmonic equivalent as Cure for Monotony

When data registers little fluctuation, it will lead to interrupted, repetitive sound patterns. Enharmonic equivalent can remedy this monotony.

4. Use of Complex Rhythms

Much of the data in question translate into complex musical notes. By tying them together, the sound can be streamlined. Also, increasing to upbeat from downbeat will make the sound more performable. Assignment of two alternate notes to a data point also streamline the sound (for example, + or + ), not out of proportion of the analysis of data.

5. Performance Levels

Data-driven compositions are complex, even given the fact that they are part of modern music. This can be a severe weakness in terms of a possible performance of 57 them. Finding musicians capable of playing such compositions will be not easy. After a series of practice and rehearsals, a conductor will be needed to facilitate performances by an ensemble or a trio. Alternatively, they can practice in conjunction with MP3 recordings, video clips or metronome.

6. Shortening Composition Time

As data analysis led to spreadsheets of sound lengths and pitches, it has become self-evident that time spent on composition has shortened substantially. The placement of musical notes has already been done, leaving the composer with previously time- consuming jobs of articulating and placing rests. The overall perspective and instrumental harmony have yet to be taken care of, but composition time has shortened.

The primary factor that shortens time lies with the fact that there is an increasing overlapping in data as the analysis was proceeded with. While it varies depending on data sets, the pitches and lengths of sound can turn into tempo, rhythms, rests, and articulations—or a repeat of these. There is a reasonable likelihood of shortened composition time.

7. Liquidity of Data Analysis

Analysis of data can turn up multiple probable values. For instance, 7 and 8 can turn into 4 ( ) + 3 ( ) = and 8= , threefold of 7. Basically, 7 should be at least less than which is less than . To express 7 in lower notes than ones in which 8 is

58 expressed, it would take more complex calculations—the process that will likely yield hundreds of notes. It will be excessively time-consuming. All these calls for the need for data liquidity. This should be precise and consistency because it must contain the concept of writing music using scientific data and personal philosophy.

8. Philosophy

As discussed briefly above, the composer needs his philosophy. He needs to think ahead to determine: commonality and relatedness between data and his composition style; the subject matter to give expression to; his expectations in the composition; ways to inject his philosophy and ideas; and his audiences and appeal.

Score 1. Joungmin Lee: Contrast & Conflict 60

Chapter 6. Conclusions

This paper has examined new compositional methods using scientific data in an attempt to shed light on a new data-driven method of music composition to enhance the scope of musical creativity and productivity. It also created a spreadsheet of the heights and lengths of sound coming from the data of melting glaciers to compose pieces of musical work for seven instruments.

The most important finding was the ability of such a method to shorten composition time. The number of notes for an instrument was limited to 20 (some instruments need more than 20 because rhythms need splitting as dictated by data analysis). Without data analysis, it would have taken three or four days to compose a piece of such complexity. This composition period includes drafting, instrumental composition, conceptualization, and philosophical founding. Even with more musical notes and essential work needed, the data-driven composition will be more time-saving than traditional methods. Of course, data collection and analysis can be time-consuming.

Transformation of data into musical notes can be more timing-consuming than usual preparations for music composition. Once all in spreadsheets, the process will be substantially more timesaving.

The above finding is conspicuous, and what is more important is: an entirely single method used for composition. Electronic music pieces inspired by glacier data does not represent a leap in music or creativity. However, the acoustic piece presented in the paper epitomized in what future-oriented modern music is about: technique, complexity, creativity and musical values. It also offered an opportunity to unlock unlimited potential because it showed the possibility of developing uniqueness by using any data.

However, the complexity of the composition may diminish the opportunity for it to perform. More research is needed to substitute complex pitches or acoustic lengths with alternatives such as rests and articulations.

Bibliography

A.Koblyakova, S.Petoukhov, I.Stepanian, The Genetic Code, the Golden Section and Genetic Music, The Moscow PI. Tchaikovsky Conservatory, 2015.

Bruno Degazio, The Schillinger System of Musical Composition and Contemporary Computer Music, Sheridan College, 1988.

Bruno Degazio, The Schillinger System of Musical Composition and Contemporary Computer Music, Sheridan College (Oakville), 1986

Catherine Schmidt-Jones, Sound, Physics and Music, Online: , Rice University, Houston, Texas, 2005.

Christopher Dobrian, Realtime Stochastic Decision Making for Music Composition and Improvisation, 1966.

David J Hargreaves, Dorothy Miell, and RAYMOND A. R. MACDONALD, What are musical identities, and why are they important, Article, January 2002.

David R. Lapp, The Physics of Music and Musical Instruments, Wright Center for Innovative Science Education, Tufts University, 2003.

Frans Absil, A Guide to Schillinger’s Theory of Rhythm Second Edition, F.G.J. Absil, the Netherlands, 2015

Hong-Kyu Lee Par, Jin-Ho Park, Jung-Keun Dong, A Study on the Form of his Philips- pavilion based on Xenakis' s the Musical Thought, The Korean Institute of Architects (KAERI) Journal of Architectural Planning, Vol. 25, No. 3, 2009

Ian Cross, Music and science: three views, University of Cambridge, 1998.

Iannis Xenakis & Additional Material compiled and edited by Sharon Kanach, Formalized Music – Thought and Mathematics in Composition, Harmonologia Series No. 6 – Pendragon Press Stuyvsant NY, 1992.

Jae-hyung Cho's guest reporter, Science stories hidden in music(Ancient Greek literature is a textbook of mathematics?), The Science Times, 2009.

James Harley, The Creative Compositional Legacy of Iannis Xenakis, School of Fine art and Music, University of Guelph, 2006.

James Q. Feng, Ph.D., Music in Terms of Science, 2012.

Janelle Anderson, Xenakis' Combination of Music and Mathematics, The Journal of Undergraduate Research - South Dakota State University, 2011.

Jean Maruani, Science and music: from the music of the depths to the music of the spheres, Pierre and Marie Curie University - Paris 6, 2003.

John Haines and Patricia DeWitt, Johannes De Grocheio and Aristotellan Natural Philosophy, Cambridge University Press, 2008.

Joseph Schillinger, The Schillinger System of Musical Composition, The Musical Quarterly, Vol. 32, No3, 1946.

Ki Jeong Lee, Understanding Music Philosophy, Article of Korea Ethnic Music Society, 1996.

Laurel Trainor, The neural roots of music, Nature Vol 453, 2008.

Leibniz, Mathematics illuminated Unit 10, Harmonious Math Textbook, 2017.

Matthew Rankin, A Computer Model for the Schillinger System of Musical Composition, A thesis submitted in partial fulfillment of the degree of Bachelor of Science (Honours) at The Department of Computer Science Australian National University, 2012

MIT OpenCourseWare, 2.161 Signal Processing: Continuous and Discrete - For information about citing these materials or our Terms of Use, 2008.

Moreno Andreatta and Carlos Agon, structure and symmetry in Iannis Xenakis’ nomos alpha for cello solo, Form and Symmetry: art and science buenos Aires Congress, 2007.

NASA Distributed Active Archive Center (DAAC) at NSIDC, GIMP Data (Greenland Ice Mapping Project), https://nsidc.org/data/measures/gimp 64

Pallett PM, Link S, Lee K. New "Golden" Ratios for Facial Beauty." Vision Research. 2010;50(2):149. Park Sun-Kyo (Jeonju Arts Middle School Teacher) [email protected], If science is hidden, music is fun. The National Research Foundation of Korea (NRF)

Philipp Brüllmann, Music Builds Character Aristotle, Politics VIII 5, 1340a14–b5, De Gruyter, 2013.

Reginald Bain, The Harmonic Series - A path to understanding musical intervals, scales, tuning and timbre, University of South Carolina, 2003.

Robert A. Wannamaker, Structure and Perception in Herma by Iannis Xenakis, School of Fine Art and Music, University of Guelph, 2000.

Saloni Shah, An Exploration of the Relationship between Mathematics and Music, Manchester Institute for Mathematical Sciences at School of Mathematics at The University of Manchester, 2010.

Stanford Encyclopedia of Philosophy, Enlightenment, First published Fri Aug 20, 2010; substantive revision Tue Aug 29, 2017.

Stanford Encyclopedia of Philosophy, Scientific Method, First published Fri Nov 13, 2015, https://plato.stanford.edu/entries/scientific-method/.

Stephen Malloch and Colwyn Trevarthen, The Human Nature of Music, US National Library of Medicine National Institutes of Health Search database Search term, 2018.

Stephen R Hoon (Manchester Metropolitan University ) & B.K. Tanner (Durham University), The physics of music, Article in Physics Education, Phys. Educ., Vol 16. 1981 Printed In Great Britain, 1981.

Stockhausen Karlheinz, Four Criteria of Electronic Music, 1989

The Department of Computer Science Australian National University, 2012.

Xenakis Iannis, Formalized Music Thought and Mathematics in Composition, Paris in 1963.