How Musicians Develop the Ability to Improvise: a Cross-Cultural Comparison of Skill Development in the Egyptian, Hindustani Classical, and Jazz Traditions

How Musicians Develop the Ability to Improvise: A Cross-cultural Comparison of Skill Development in the Egyptian, Hindustani Classical, and Jazz Traditions

Christian Watson

School of the Arts and Media University of New South Wales

This thesis is presented for the degree of Doctor of Philosophy in Music

2012

THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet

Surname or Family name: WATSON

First name: CHRISTIAN Other name/s: PETER

Abbreviation for degree as given in the University calendar: PhD

School: School of the Arts & Media Faculty: Arts

Title: How musicians develop the ability to improvise: a cross- cultural comparison of skill development in the Egyptian, Hindustani classical, and jazz traditions

Abstract 350 words maximum:

This thesis demonstrates what musicians in three distinct traditions do in order to develop the ability to improvise. Rather than retrospectively examining what musicians have improvised, or investigating what occurs cognitively during a performance, this thesis has at its crux a unique cross-cultural comparison that details the specific methodologies that musicians practise when developing improvisational ability. Concurrently, this study identifies the principal modes of transmission and learning environments that facilitate and nurture this development. Detailed findings provided by fieldwork that the author conducted in Egypt are compared to existing research on the jazz and Hindustani classical music traditions.

Research presented in this thesis shows that improvisational ability emerges through a multi-layered process founded within multiple learning environments. By interacting in these enriching milieus, musicians benefit from direct instruction and are inspired to engage in autodidactic modes of learning. The extent to which they participate in either learning model is emphasised differently in each tradition, and more specifically from musician to musician.

In documenting this multipartite developmental process, this thesis establishes that musicians undertake various activities that allow them to internalise the style and content that is embodied within composed and improvised prototypes. These prototypes are drawn from the canon of each tradition and are defined in part by the constraints that uniquely characterise each musical culture. The entire process is effective because musicians are in aural contact with the prototypes, which can be emulated in both unabridged and truncated forms, and are imitated repetitively until internalised. This lengthy and arduous imitative process enables musicians to improvise: that is, to intuitively generate novel musical statements in real time that adhere to genre-specific parameters.

The manner and means through which musicians become expert improvisers, as explicated by this cross-cultural study, is shown to adhere predominantly to a pre-existing framework of expertise acquisition.

This thesis concludes by demonstrating that the imitative process through which musicians acquire improvisational ability has a developmental origin in early childhood, a conclusion that is derived by contextualising the findings of this study within recent research into neonatal psychobiological development.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

.. ..... Signature Witness Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

THIS SHEET IS TO BE GLUED TO THE INSIDE FRONT COVER OF THE THESIS

‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

Signed ......

Date ......

AUTHENTICITY STATEMENT

‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’

Signed ......

Date ......

ORIGINALITY STATEMENT

‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project’s design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed ......

Date ......

ii Abstract

Concurrently, this study identifies the principal modes of transmission and learning environments that facilitate and nurture this development. Detailed findings provided by fieldwork that the author conducted in Egypt are compared to existing research on the jazz and Hindustani classical music traditions.

iii intuitively generate novel musical statements in real time that adhere to genre-specific parameters.

iv Table of Contents

Abstract iii Acknowledgments ix List of Tables xi List of Figures xii List of Audio Examples xiii

Chapter 1: Introduction and Context 1

1.1 Introduction 1 1.2 Thesis structure 2 1.3 The methodology 4 1.4 Research consultants: musicians of Cairo 5 1.5 The musical traditions examined by this dissertation 7 1.5.1 Jazz 9 1.5.2 Hindustani classical music 9 1.5.3 Egyptian music 11 1.6 Definitions of analytical terminology 12 1.6.1 Improvisation 12 1.6.2 A contextual definition of improvisation 18 1.6.3 Professional musician 23 1.6.4 Autodidact 24 1.6.5 Mode of transmission and learning environment 25 1.6.6 Osmotic and osmosis 25 1.6.7 Mimetic and mimesis 26 1.6.8 Oral-aural tradition 26 1.6.9 An instrumental and vocal focus 26

Chapter 2: Contextual Explication 28

2.1 Hindustani classical music 29 2.1.1 Rāga 29 2.1.2 Tāla 33 2.1.3 The ensemble context 34 2.1.4 The performance structure and associated improvisatory practices 35 2.1.5 The improvisational process 38 2.2 Jazz 41 2.2.1 The traditional performance context 42 2.2.2 Improvisation 44 2.2.3 The ensemble context 46

v 2.3 Egyptian music 48 2.3.1 Taqsīm and mawwāl 49 2.3.2 Maqāmāt 53 2.3.3 Modulation 66 2.4 Conclusions 68

Chapter 3: Learning to Improvise in Hindustani Classical Music 70

3.1 Primary learning environment: guru and shisyā parampara 71 3.1.1 Modern adaptations of the guru-shisyā parampara 77 3.1.2 Detailed studies 80 3.2 Primary pedagogical methodologies 82 3.2.1 Imitation and vocalisation 82 3.2.2 Imitation and sārgām 84 3.2.3 Improvisation and the role of prototypes 86 3.2.4 Bandish: why are they taught? 90 3.2.5 Improvisation pedagogy 92 3.2.6 Springboards: the mukhrā and model improvised phrases 93 3.2.7 Tān 94 3.2.8 Systems of variation and recombination 96 3.2.9 Tihāī phrases 100 3.2.10 Concentrating effortlessly 101 3.3 Riaz 105 3.4 The conceptual aims of pedagogy 106 3.5 Conclusions 107

Chapter 4: Learning to Improvise in the Jazz Tradition 109

4.1 Childhood in the jazz tradition 110 4.2 Informal networks and jam sessions 111 4.3 The jazz performance format and approaches of mastery 113 4.4 Learning the jazz harmonic context 115 4.4.1 Memorising the melody 115 4.4.2 Benefits of aural memorisation: aural acuity and improvisational ability 117 4.4.3 Individual processes of memorisation 119 4.5 Memorising and transcribing improvisations 120 4.6 Towards procedural automation 123 4.6.1 Phenomenology and automatic processes 124 4.6.2 Speech and improvisation analogies 125 4.7 Jazz in the university and college curriculum 129 4.7.1 The history of jazz education 129

vi 4.7.2 The nature and content of jazz pedagogy at the tertiary level 132 4.8 Conclusions 140

Chapter 5: Learning to Improvise in the Egyptian Music Tradition 144

5.1 The learning environments and pathways to becoming a musician 144 5.1.1 Childhood in Egypt and the greater Arab world 144 5.1.2 The family environment 145 5.1.3 The rise of Westernised environments 147 5.2 From childhood interest to training 149 5.2.1 Apprenticeship-like relationships 149 5.2.2 Have these apprenticeship-like relationships been replaced by institutionalised training? 152 5.3 Reinvention and adaptation of past learning environments 154 5.3.1 Apprenticeship-like relationships are still essential 154 5.3.2 Arabic musical content in institutionalised tertiary training 155 5.3.3 The tertiary curriculum and the (in)ability to improvise 160 5.3.4 Why is improvisation not taught? 161 5.4 What do musicians do in order to develop the ability to perform taqāsīm and mawwāl? 163 5.4.1 Aural imitation of prototype melodies: vocal songs 164 5.4.2 From vocal song to instrumental composition 168 5.4.3 Instrumental compositions as prototypes 171 5.4.4 Autodidactic preparatory strategies stemming from personalised conceptions of taqsīm and mawwāl performance 175 5.5 Conclusions 195 5.6 A final thought: can improvisation be taught? Or must it be learned? 197

Chapter 6: The Cross-cultural Comparison 204

6.1 Comparison of learning environments 204 6.1.1 Early childhood learning environments 204 6.1.2 Apprenticeships 206 6.1.3 Informal apprenticeship-like relationships with master musicians and in ensembles 210 6.1.4 In situ training: the jam session 211 6.1.5 Institutional training in modern-day contexts 215

vii 6.2 Comparison of practices and related modes of transmission 220 6.2.1 Melodic prototypes 220 6.2.2 The role of discrete musical phrases in the development of technique related to improvisational fluency 229 6.2.3 The role played by the intentional formation of a phrase lexicon and by the premeditation of cadential figures and other structural devices 234 6.2.4 The pedagogical interaction of voice and instrument: vocal imitation by instrumentalists, and vocalisation as pedagogy 236 6.3 Conclusion: how improvisational ability is developed most effectively 242

Chapter 7: Correlation and Conclusions 244

7.1 The acquisition of musical expertise: the pre-existing framework of expertise development 245 7.2 The development of expertise in the jazz tradition, Hindustani classical music, and Egyptian music 253 7.2.1 Starting age and the role of teachers 256 7.2.2 The role of deliberate practice 260 7.2.3 The ‘ten-year’ rule 262 7.2.4 Motivation 265 7.3 The developmental origin of imitative behaviour 268

Bibliography 274 Internet Bibliography 282 Discography 283 Appendix 1 284

viii Acknowledgements

I extend my sincere thanks to my research supervisor Dr John Napier. The depth and breadth of his expert knowledge was reinforced frequently during the writing of this dissertation, and his perceptive analysis and advice has impacted the quality of this thesis markedly.

I also thank Dr Christine Logan, Associate Professor Emery Schubert, Associate

Professor Dorottya Fabian, and Dr John Peterson for creating a robust and stimulating musicological environment at the University of New South Wales, School of the Arts and Media. Their impact on my tertiary music education is recognised in thanks. I acknowledge with gratitude the financial assistance granted by the University during my candidature in the form of an Australian Postgraduate Award.

I thank both Dr Sarah Mares for her guidance and advice concerning infant psychobiological development, and Dr Kristin Argall for her perceptive analytical skills. I thank my colleagues and the students of St Andrew’s Cathedral School for their support during the writing of this thesis.

I am indebted to virtuoso ‘ūd player Mr Joseph Tawadros for facilitating my research in Cairo, Egypt. His expert guidance, local knowledge, translation skills, and friendship were a constant source of encouragement throughout this research project.

His musical genius was one of the original motivations for my interest in Egyptian music, and he continues to inspire me to this day. I also offer sincere thanks to Dr

Alfred Gamil, Mr Esawi Dagher, Mr Abdo Dagher, Mr Ahmad Ali El Haggar, Mr

Mohammed Antar, and Mr Fathy Salama. These musicians gave of their time freely and sincerely, opened their homes and rehearsal venues with characteristic heartfelt hospitality, and shared their wonderful musical tradition with inexhaustible enthusiasm.

ix Without their participation in this project this dissertation would not exist, and the opportunity I had to perform at the Cairo Opera House on a hot Ramadan evening with

Fathy Salama will remain one of my most treasured musical memories.

I give profound thanks to my parents Peter and Lesley for their incredible support over the course of this dissertation. They have provided the ultimate role model of hard work and determination. I thank my brother Adam, his partner Alexandra, and my sister

Sarah for their friendship and encouragement. I give thanks to Warren and Sieglinde and the entire Argall family for their love and support.

Finally I thank my wife Joanna, whose unwavering love and encouragement made possible the completion of this work, and I give thanks for my son Elliot, whose birth provided the final powerful motivation to complete this project, and who I hope, one day, will benefit from its conclusions.

x List of Tables

Table 1: Sārgām syllables 30

Table 2: The use of jam sessions as a learning environment 213

Table 3: The nature of historical apprenticeships 214

Table 4: How institutionalisation has been implemented 219

Table 5: The use of melodic prototypes to develop improvisational ability 233

Table 6: Modes of transmission active in each tradition 241

xi List of Figures

Figure 1: Swing rhythm (simplified) 47

Figure 2: Maqām hijāz as described by D’Erlanger 60

Figure 3: Maqām hijāz: modern-day representation as a scale 60

Figure 4: Score realisation of the worded description of bayyātī al-husaynī 185

Figure 5: Maqām husaynī (bayyātī al-husaynī) as represented in

modern theory 185

xii List of Audio Examples

Audio Example 1: Parallel audio comparison of a taqsīm in maqām hijāz with a

recitation of the adhān in maqām hijāz

Audio Example 2: Demonstration of phrases from the melody of the song Sahran Le

Wahdi (‘Sleepless by Myself’), as sung by Umm Kulthūm

These audio examples are on the two compact discs found at the back of the thesis.

xiii Chapter One Introduction and Context

1.1 Introduction

While fledgling when compared to other areas of musicological investigation, the corpus of research that investigates musical improvisation has burgeoned over the last

50 years. This period of augmentation juxtaposed a paucity of research emphasised by

Nettl in the period between Ernst Ferand’s quintessential study of 1938 and the 1960s

(Nettl 1998, pp. 1–2). Since the 1960s, studies have examined the phenomenon with considerable scope, focusing on stylistic, cultural, sociological, historical, analytical, philosophical, and biographical themes. Recent research grounded in cognitive psychology has attempted to define what the internal mental processes are that shape improvised musical responses, by delineating the cognitive constraints that impact on real-time performance, with particular effort invested in constructing models that simulate improvisation.

One goal of research has been to illuminate the improvisational process, a practice often clouded by enigmatic allusions, but enquiries have only rarely extended to a detailed analysis of how the skills of improvisation are acquired, often dwelling on what musicians play when improvising. Instead, I propose that for musicians who hope to develop improvisational ability, surely an essential skill, if not the most important, is to learn how improvisational ability is acquired in the first place.

Literature that undertakes a detailed analysis of improvisatory skill acquisition is limited, and whilst studies of individual traditions do exist, no analysis has attempted to cross-culturally compare a selection of improvised traditions in a comprehensive manner. I assert that forming a narrative of cross-cultural skill acquisition, thereby

1 developing a sequence that emphasises significant points of commonality, is essential in order to ascertain those practices that may be considered most effective. Best practices thus identified may theoretically be replicated successfully in a broad range of musical contexts, and improvisational ability may be developed effectively.

To this end, throughout this dissertation I will undertake a cross-cultural comparison between three traditions. Fieldwork that I conducted in Cairo, Egypt, provides data that I cross-culturally compare to data that I have derived from literature investigating the jazz and Hindustani classical musical traditions.

The primary and secondary research undertaken by this thesis shows that musicians develop the ability to improvise by engaging in imitative behaviour, specifically by internalising aural prototypes through repetitive practice. Doing so (i) enhances their aural acuity, (ii) facilitates the internalisation of archetypes of style and content, and (iii) develops requisite tradition-specific technique. The symbiosis of these skills enables musicians to (i) play internally generated phrases 'by ear' that are (ii) regulated intuitively by an internalised knowledge of genre-specific criteria, and (iii) realised physically by tradition-appropriate technique. The exposition of this developmental process forms the centrepiece of this thesis' cross-cultural comparison, detailed henceforth.

1.2 Thesis structure

Following this introductory chapter, I present a detailed analysis in Chapter Two of the key theoretical and structural elements of each musical tradition, so as to contextualise my subsequent examination of how musicians learn to improvise. Within Chapter Two, data concerning Hindustani classical music has been drawn chiefly from Wade (2000),

Slawek (2000a), McNeil (1993–1994), Rao et al. (1999), Ruckert (2004), and Ruckert

2 and Widdess (2000). Information regarding the theoretical aspects of the jazz tradition has principally been derived from my own experiences as a jazz musician and educator, supplementing this practical experience with data chiefly derived from Berliner (1994).

Theoretical detail concerning Egyptian music theory is drawn from research by Marcus

(1987, 1989, 1992, 1993a), Touma (1971, 1996), Racy (2000), Ayari and McAdams

(2003), Nettl and Riddle (1973, 1998), and the website Maqam World (2001–2005). My research conducted in Cairo (2008) supplements this literature.

In Chapter Three I have presented an historical and current perspective of

Hindustani music pedagogy, extracting data for cross-cultural analysis from research by

Alter (1989, 1997), Scott (1997), Booth (1982, 1983, 1986, 1987, 1996), Meer (1980),

Neuman (1980), Ruckert (2004), Shankar (1969), Shepherd (1976), Schippers (2007),

Khan and Ruckert (1998), and Kippen (1988).

Similarly, in Chapter Four I have derived information for cross-cultural comparison that details how jazz musicians have historically developed the ability to improvise, whilst emphasising how this context has changed over the last century. This chapter is formulated around Berliner’s expansive study (1994) and I have supplemented his pioneering work with research by Prouty (2004, 2005, 2006), Ake (2002), DeVeaux and

Giddens (2009), McDaniel (1993), Sudnow (1978, 2001), and Williams-Jones (1975).

My original ethnographic fieldwork (the methodology of which is detailed forthwith) is analysed in Chapter Five, in which I also draw upon research principally by Thomas (2006), and to a lesser extent by Racy (2003). By Chapter Five’s close I establish how Egyptian musicians acquire the ability to improvise.

The crux of this thesis is found in Chapter Six, in which I conduct a cross-cultural comparison that emphasises significant points of commonality in how improvisational ability is developed in the three traditions examined by this dissertation. The

3 conclusions of Chapter Six highlight the principal means through which improvisational ability is developed most effectively.

In Chapter Seven I draw correlations and present my conclusion. In a tripartite manner, I firstly compare the findings of Chapter Six with research concerning expertise acquisition, and determine the extent to which my findings conform to the framework developed by Ericsson et al. (1993). Secondly, I contextualise the fundamental conclusions of my thesis within a psychobiological narrative of human development derived from Malloch and Trevarthen (2009), Bannan and Woodward (2009),

Mazokopaki and Kugiumutzakis (2009), Gratier and Apter-Danon (2009), and

Custodero (2009). Finally, I return to the definition of improvisation that I developed in

Chapter One and reconsider it in light of the findings of this thesis.

1.3 The methodology

As the preceding discussion regarding thesis structure has indicated, the cross-cultural comparison that I conduct in Chapter Six of this dissertation was formed by comparing data extracted from literature that examines the jazz and Hindustani classical musical traditions with data derived from fieldwork that I undertook in Cairo, Egypt, in 2008. I supplemented my focussed research in Egypt with studies by Racy (2003) and Thomas

(2006), both of which investigate Arabic music more expansively. Mr Joseph Tawadros, who acted as my translator and guide, facilitated my fieldwork in Cairo. His knowledge of the local music scene and musicians was an invaluable asset, enhanced by the respect he has garnered in Cairo’s music community.

Specifically, I conducted and recorded interviews with a variety of Egyptian musicians, meeting with them in their homes, at rehearsals, and before and after concerts. I also attended numerous concerts given by my research contacts and other

4 local musicians. These interviews and observations form the foundation of my original research, conducted for the purposes of the cross-cultural comparison that I undertake in

Chapter Six of this dissertation. Significantly, I was also given the opportunity to rehearse and eventually perform at the Cairo Opera House with the ensemble Sharkiat, led by Fathy Salama. This ensemble performs a fusion of jazz and Egyptian music, and whilst the experience does not feature in my analysis, it certainly provided an important insight into one aspect of the Egyptian music industry.

1.4 Research consultants: musicians of Cairo

During my time in Cairo I met many musicians and conducted formal recorded interviews with several. I will now provide brief biographical introductions for those musicians whose experiences form the centrepiece of data derived from this fieldwork.

Esawi Dagher (Dagir) (b. 1941) has worked principally as a professional violinist in Cairo since 1960. He has also supplemented his income by repairing and restoring violins during this time. The younger brother of the more famous Abdo Dagher, Esawi has extensive experience within the Egyptian music community that began during his upbringing in rural Upper Egypt and has continued throughout his urban life in Cairo.

Abdo Dagher (Abdu Dagir) (b. 1936), the elder brother of Esawi, is recognised throughout Egypt as the malik at-taqāsīm (king of improvisation). A master of Arabic violin, he has performed throughout the Arab World and Europe as a soloist and with various ensembles that he has led. Throughout his career in Egypt, he has performed and recorded with the two most famous Egyptian vocalists of the twentieth century:

Umm Kulthūm (pronounced Umme Kolsum) and Mohamed Abdul Wehab. He has also regularly given concerts leading his own orchestra, performing what has been labelled as Fan el-musiqä el-misriyya el-hadithah el-’arabiyyah es-sufiyyah (The New Egyptian-

5 Arabic Sufic Art Music). Religious hymns and Quranic reading have inspired these instrumental compositions, and I was able to observe rehearsals and performances of this ensemble during my fieldwork. Dagher’s compositions draw upon his extensive experience gained whilst living and performing with Sufi religious orders after leaving his father’s home at age 13 (Dagher 2000).

Ahmad Ali El Haggar (b. 1989) was the youngest musician that I interviewed in

Cairo. The son of famous Egyptian vocalist Ali El Haggar, Ahmad inherited a musical tradition preserved by his grandfather, uncle and father. As a child, he performed in the

Children’s Choir at the Cairo Opera House, and he graduated as a vocalist from the

Higher Institute for Arab Music in 2010. Today he pursues a professional musical career as a composer, ensemble leader, and vocalist.

Mohammed Antar (b. 1978) is a nāy (end blown bamboo flute) player. He has performed throughout the Middle East and Europe as a soloist and conducts the

Ensemble Munajah and the Oriental Secrets Ensemble. Antar pursues a conservative musical style derived from his research into Middle Eastern musical traditions and has studied three schools of nāy playing: Arabic, Turkish, and Persian.

Dr Alfred Gamil (b. 1957) provided an extensive range of data for this dissertation due to his multifaceted professional career as a violinist, ‘ūd 1 player, composer, ensemble leader, and tertiary educator. Gamil studied formally at The Higher Institute for Arabic Music, and after completing post-graduate qualifications, took up an assistant professorship in 1996. During his professional career he has led Ensemble Qithara, a traditional instrumental chamber ensemble with vocalists, and has performed throughout the Middle East, Europe, and the USA. As pedagogue and performer, Gamil proved to be an essential research consultant.

1 The ‘ūd is a Middle Eastern fretless lute.

6 Joseph Tawadros (b. 1983) is an Egyptian-born Australian musician. Immigrating to Australia as an infant, Tawadros began a professional career as an ‘ūd player and composer shortly after leaving high school, establishing a reputation whilst still studying music at the University of New South Wales. He has performed throughout the world as a virtuoso soloist and ensemble leader. Highlights of his extensive professional experiences thus far include performances and recordings as ‘ūd soloist with the

Australian Chamber Orchestra, and a recent recording with jazz luminaries John

Abercrombie, John Patitucci, and Jack DeJonette (Tawadros 2010).

The findings presented in Chapter Five, drawn from my fieldwork in Cairo, also feature references to other musicians whom I have not introduced individually in this chapter. Their input was less substantial than the information gleaned from those consultants introduced above, and as such, I introduce them briefly within the context of

Chapter Five.

I will now clarify, where necessary, the identity of the traditions examined in this dissertation. Definitions will not be detailed, so as to not pre-empt a more comprehensive explication in Chapter Two. Rather, they will establish definitional parameters within which my forthcoming analysis will be conducted.

1.5 The musical traditions examined by this dissertation

I submit that as a diverse conglomerate, the traditions chosen for comparison in this thesis represent an important cross-section of improvisatory practices the world over.

Whilst they are diverse, a principal reason for comparing these three specific traditions is that vocalists and instrumentalists perform linear improvisations in each. This parallel provides fertile ground for comparison, and allows for the means through which musicians learn to improvise to be identified and emphasised where similarities in

7 methodology exist. Furthermore, these traditions have been chosen for study because each is in the process of transitioning from a completely oral-aural mode of transmission to a partially oral-aural and a partially institutionalised context. The latter is characterised by the standardisation of learning and teaching, and a movement towards written materials over oral transmission. Analysis of this commonality enables this thesis to determine if these transformations have had any impact over the means through which improvisational ability is developed, and in doing so, reinforces the principal means through which this takes place.

From this fundamentally similar position, these traditions have also been chosen for comparison due to their inherent differences. Expressed in genre-specific stylistic idiosyncrasies, these differences are identified by the way in which musicians conceptualise musical components and relationships. Each tradition has distinctive principles of theoretical organisation, ranging from how melodic, harmonic, and rhythmic aspects interrelate, to the interface through which ensemble members participate. Furthermore, the nexus between fixed (precomposed) and flexible

(improvised) aspects of performance, and related notions of musical originality, are theorised and practised uniquely in each tradition.

The sum of these distinctive attributes provides a two-fold advantage for comparative research. Firstly, these individual differences create correspondingly unique methods through which musicians develop improvisational ability. Through comparison these unique methods are uncovered and best practices can be considered.

Subsequently, methodological adaptations can be contemplated where appropriate.

Secondly, and perhaps more significantly, where it is discovered that musicians employ similar methods to overcome the challenges of improvisation in distinctive musical contexts, the existence of these similar methods lend emphasis to their global import.

8 1.5.1 Jazz

For many readers, the jazz tradition is a familiar genre. Recognised as the bastion of modern Western improvisatory practices, it has been propagated from its American birthplace in the early twentieth century throughout the world. Continuing the fusion that characterised its evolution from African-American roots, it has exploded into a myriad of localised forms. Within this diversity of local interpretations, a fundamental stylistic character has been maintained in varying degrees, defined by a strophic performance structure, the rhythmic conceptualisation known as swing, small to medium ensemble sizes, a typical instrumentation, and an integrated focus on both soloists and group interaction.

Concurrently, jazz has been sustained by the synergy of stylistic extremes present within its diversity and all creative spaces between; those musical activities aligned closer to historical models preserve, whilst those associated with innovation invigorate.

This interpretive dialogue simultaneously reinforces the music’s foundations and inspires new audiences to engage. For the purposes of my analysis of how improvisational ability is developed, my investigation will be inclusive, viewing the jazz tradition as the sum of its diversity. At times I present more traditional examples, such as Chapter Two’s analysis of structural and theoretical aspects of jazz performance, but in my analysis of skill development I have not felt the need to distinguish between differing stylistic approaches. The manner in which improvisatory ability is developed remains relatively constant across this broad spectrum.

1.5.2 Hindustani classical music

Hindustani classical music may also be well known to readers, popularised throughout the second half of the twentieth century through its association with Western popular

9 music. Even so, I will clarify the identity of Hindustani classical music to avoid ambiguity. Two defining principles determine what is considered art music in India, of which Hindustani classical music is a genre. As Qureshi explains:

In South Asia the concept of art music is based on an amalgam of scriptural foundation and oral transmission. Authoritative theoretical doctrine and a disciplined oral tradition of performance extending back over several generations are the two complementary criteria that serve to legitimize Indian art music and its theoretical and historical foundation (Qureshi 2012).

Representative of this criterion, Hindustani classical music is the legitimate art music of northern India. The term ‘classical’ distinguishes Hindustani music from folk or popular music (Qureshi 2012), and throughout this dissertation I refer to the tradition as

Hindustani classical music and Hindustani music interchangeably. In Chapters Two and

Three I will expand upon the twofold authoritative amalgam given by Qureshi, particularly concerning generational oral transmission as it operates in Hindustani classical music. The geographic descriptor ‘northern’ distinguishes Hindustani music from its southern art music equivalent, Karnatak music, and it is also representative of musical differences that are defined by degree, resulting from localised stylistic development and differing emphases on elements derived from their common musical origin (Qureshi 2012). I have restricted my comparison to the north of India for reasons of economy,2 a personal affinity for Hindustani style, and the existence of a substantial collection of literature from which a suitable foundation could be formed for my cross- cultural comparison.

2 Examining both Hindustani and Karnatak music would have exceeded this dissertation’s scope.

10 1.5.3 Egyptian music

Organising Egyptian music into specific categories to distinguish between art, folk, and popularly derived musical forms is not an indigenous Egyptian practice. As El-Shawan

Castelo-Branco elaborates with reference to the Egyptian population:

Musicians and audiences make broad distinctions between Western and Arab musics, rural and urban styles, and religious and secular vocal expressions; but the Western paradigms of art, popular and folk music are not applicable to Egyptian musical production, and there are no local terms analogous to ‘art music’ or ‘popular music’. In Cairo indigenous musics can be conceptualized as a central sphere of overlapping and interrelated musical styles characterized by the fluidity of their musical and conceptual boundaries and by constant changes in musical content, behaviour, discourse and meaning (El-Shawan Castelo- Branco 2012).

Alternatively, the phrase al-mūsīqā al-’arabiyya (Arab music) has been used since 1930 to refer to music in Egypt that is performed by Arabs and that conforms to the customs of Arabic music, according to the perceptions of audiences and musicians (El-Shawan

Castelo-Branco 2012). By extension, any musical genre would be labelled al-mūsīqā al-

’arabiyya if it adheres to traditional Arabic musical conventions, which in itself is a subjective assessment. The details of these Arabic musical norms will be discussed at length in Chapter Two, and music conforming to these customs is the focus of research presented in Chapter Five.

Confirming El-Shawan Castelo-Branco’s assertion (2012), during fieldwork that I conducted in Cairo during 2008, I observed that rather than drawing distinctions between genres, of which the boundaries are ambiguous in any case, musicians demonstrated a greater propensity to distinguish their music based on the type of processes they were engaged in. Accordingly, musicians principally differentiated between performing a composition and improvising. The term taqsīm (plural: taqāsīm) is used to label an improvisation, and it is a musical activity that occurs throughout the broad spectrum of Egyptian music. Throughout this thesis I refer to improvisation in

11 Egyptian music in its singular and plural form. Given that taqsīm is practised in a relatively uniform theoretical manner throughout Egyptian music, conforming to the customs of Arabic music, I do not draw a distinction between improvisation that occurs in music that is sacred or secular, rural or urban, and vocal or instrumental. Unlike my analysis of jazz and Hindustani classical music, where I focus on the individual genre paradigms in which improvisation occurs, I refer to taqāsīm as a practice that occurs in multiple musical contexts, unified by common theoretical precepts.

1.6 Definitions of analytical terminology

In the ensuing paragraphs I will clarify various terms that are used as central descriptors throughout my dissertation, in order to remove any uncertainty from my subsequent analysis.

1.6.1 Improvisation

Whilst often maligned historically as a meagre alternative to composition, improvisation has experienced gradual refinement as a musicological concept over the last century. As a result of the detailed research undertaken by ethnomusicologists advocating a departure from Western-centric comparisons, the term is progressively receiving a more coherent appraisal in musicological literature. Whilst some advances have filtered into current discourses, a recently published collection of studies reinforces that ‘we haven’t found ways to study improvised music as efficiently as we do music composed and recorded in writing or memory’ (Nettl, in Solis and Nettl 2009, p. x). This deficiency, identifiable in the language of stylistic and cultural bias, was caused by inadequate methodologies inherited by musicologists who first attempted a study of improvisation

(Nettl, in Nettl and Russell 1998, pp. 8–9). In applying an approach that was designed

12 to evaluate the pre-composed music of the Western music tradition onto musical cultures whose identities eluded such analytical methods, early studies failed to accurately comprehend the organisation of these musical traditions. Consequently, musicological definitions of improvisation have reflected these inadequacies.

Before examining these issues, let us briefly examine the broader context: underlying problems encountered when using the term improvisation to describe actual musical processes. Of course, in the most general, extra-musical sense, the term

‘improvised’ is used in mainstream discourse to refer to the manner in which solutions are spontaneously developed in response to problems, normally unforeseen. As the

Oxford English Dictionary states, improvisation is ‘The production or execution of anything off-hand; any work or structure produced on the spur of the moment’ (Oxford

English Dictionary Online 2012).

This definition summarises elements common to numerous mainstream descriptions of improvisation that often characterise the process as an unplanned, hastily conceived response. Traditional concepts of the term have stemmed from the Latin improvisus (unforeseen), which consequently has resulted in the term being associated with the makeshift: the challenge is unforeseen and so the solutions and materials that are employed must correspondingly be unplanned.

Etymologically, this focus is unsurprising, but it underscores the dilemma encountered by using the term improvisation to describe an actual musical process. If this style of definition is applied to the term in a musicological context, our understanding is obscured, as actual performance practice rarely if ever conforms to these notions. In reality, genres are given their identity by musical parameters. These constraints not only define genres but also structure the challenges associated with improvised performance within these genres. As such, musical challenges are rarely if

13 ever unforseen, and when improvising, musicians need to spontaneously fashion solutions according to these idiosyncratic templates thereby rendering musical challenges successfully. Merker labels this practice as ‘Modes of supplementing the prototype’ (Merker in Deliege and Wiggins 2006, p. 27). Crafting alternate realisations of these traditional templates in real time requires immense skill and preparation. It is therefore ironic that mainstream definitions of improvisation accentuate unplanned responses when in fact the exact opposite is true: a highly cultivated, skilled response is required of musicians.

Bailey reinforces this fact, asserting that musical improvisation requires ‘skill and devotion, preparation, training and commitment’ (Bailey 1992, p. xii). Given the demonstrated incongruity between the mainstream understanding of the term and actual performance practice, Bailey goes as far as to suggest that some improvising musicians

‘express a positive dislike for it … And so they reject the word, and show a reluctance to be identified by what in some quarters has become almost a term of abuse’ (Bailey

1992, p. xii). As such, at the outset of this discussion distance must be placed between the use of the term in mainstream language, and its use specifically relating to actual performance practice. Indeed, given this lexical prejudice, there is certainly a case worth arguing for a new term to be developed which describes more adequately the practice currently referred to as ‘improvisation’ in music. However given that the term improvisation is commonly used in musicological study, the discussion which follows will remain focussed on clarifying exactly how the practice of improvisation in music should be defined, misunderstood though it may be in common parlance.

Returning to early musicological definitions of the performance practice, Nettl gives an account of this state of affairs by offering two entries in Grove’s Dictionary of

Music and Musicians. In 1886 improvisation was included under the topic of

14 ‘extemporization’: ‘The power of playing extempore evinces a very high degree of musical cultivation’ (Nettl, in Solis and Nettl 2009, p. x). By 1942, still included under

‘extemporization’, it was defined as ‘The art of composing and performing music simultaneously; it is therefore the primitive art of music making’ (Nettl, in Solis and

Nettl 2009, p. x). In the 56 years that passed from 1886 to 1942 the definition offered by

Grove’s Dictionary of Music and Musicians became laced with a negative bias. Nettl does not speculate as to why, though I hypothesise that the ‘primitive art of music making’ was an unfortunate comment on the popularisation of jazz that occurred in the intervening period, exacerbated by the earlier decline in extempore playing in Europe, to the extent that the importance of the extempore had ceased to even be a valid memory by 1942.

Out of this historical context, an academic legacy of musicological research was born that has been concerned with debunking misconceptions and detaching improvisation from value judgements derived from Western-centred models. This has allowed a definition of improvisation to be formed that is a more adequate representation of the concept. Nettl nominates the musicologist Ernst Ferand (1887–

1972) as the ‘quintessential specialist’ and pioneer of musicological writing on this subject (Nettl in Nettl and Russell 1998, p. 1). Ferand produced the first significant study of improvisation in 1938: Die Improvisation in der Musik (Ferand 1938). More specifically, Nettl highlights an important contribution made by Ferand in 1957 where he proposed an ‘outline distinguishing medium (vocal or instrumental), personnel (solo or ensemble), technique (monophonic or multivoiced), degree (total or partial, absolute or relative improvisation), and form’ (Nettl, in Nettl and Russell 1998, p. 6).

Whilst Nettl identifies Ferand as the pioneer of musicological research into improvisation, Nettl and his colleagues must also be recognised for their significant

15 contribution to this field. Taking on Ferand’s academic mantle, they have expanded upon his work, moving into the study of non-Western music. Developing Ferand’s groundwork, Nettl suggested in 1974 ‘that the juxtaposition of composition and improvisation as fundamentally different processes is false’, leading him to the opposing contention: ‘that the two are instead part of the same idea’ (Nettl 1974, p. 6).

This theory was derived from research that Nettl previously conducted into the music of the Blackfoot Indian musical culture (Nettl 1967). Reinforcing his findings through comparison with other research on Native American musical cultures, Nettl discovered aspects inherent to their music-making processes that challenged previously espoused theories emphasising the dissimilar nature of composition and improvisation (Nettl

1974, pp. 4–5).

In reference to the musical culture of the Plains Indians, Nettl recognised that the spontaneous nature of a song’s creation, received ‘during periods of ecstasy’ (1974, p.

5), may have lead interpretations of the musical culture to focus on improvisation.

However, Nettl countered that the song’s spontaneous creation, or the absence of notation, did not preclude the music from having compositional characteristics such as enshrined authorship and relative homogeneity over multiple performances, once the song had been received during ecstasy (Nettl 1974, p. 6). As such, Nettl pondered a concluding query: ‘Is this composition or improvisation?’ (1974, p. 5).

In order to accommodate instances in which improvisation and composition interact collaboratively within the one musical culture to the point of being indistinguishable, and also to temper absolutist differentiations of composition from improvisation, Nettl proposed a continuum based definition where composition and improvisation are thought of as being at the ‘opposite ends of a continuum’ (1974, p. 6). All musical processes exist somewhere on the continuum relative to each other and to the

16 continuum’s distinct extremities. The continuum concept was later revised and restated in 1998 by Nettl (Nettl in Nettl and Russell 1998, p. 6), at this time more specifically taking into account the many manifestations of the improvisation concept. This inclusive continuum underpins Nettl’s most recent definition published in Grove Music

Online. He writes that improvisation is:

The creation of a musical work, or the final form of a musical work, as it is being performed. It may involve the work’s immediate composition by its performers, or the elaboration or adjustment of an existing framework, or anything in between. To some extent every performance involves elements of improvisation, although its degree varies according to period and place, and to some extent every improvisation rests on a series of conventions or implicit rules. The term ‘extemporization’ is used more or less interchangeably with ‘improvisation’. By its very nature – in that improvisation is essentially evanescent – it is one of the subjects least amenable to historical research (Nettl 2012).

This definition, emphasising a ‘wide spectrum of improvisation’ (Nettl, in Nettl and

Russell 1998 p. 6), is predicated on the rationale: ‘If we take an absolute stance, then, perhaps every performing musician improvises to some degree’ (Sutton in Nettl and

Russell 1998, p. 73), a concept proposed by Nettl when he stated ‘we must abandon the idea of improvisation as a process separate from composition and adopt the view that all performers improvise to some extent’ (Nettl 1974, p. 19). Nettl’s conclusion at this time was: what musicians do when performing in both composed and improvised traditions differs only in ‘degree, not in nature’ (1974, p. 19).

Similarly, Benson proposes a quantitative definition of improvisation taking an equivalent stance: ‘no performance is possible without some form and degree of improvisation’ (Benson 2003, p. 26). Thus Benson expands the definition of improvisation in a manner similar to Nettl’s continuum, to include many acts carried out in performance, from the most minimal filling-in of details not recorded in the score, to the creation in real time of a new structural element (Benson 2003, pp. 26–30). Whilst much of Benson’s analysis conforms to Nettl’s ‘degree of improvisation’ continuum,

17 Benson’s list also includes acts of musical creation undertaken before performance, such as arranging or transcribing a pre-existing score (2003, p. 27) to be performed in a way that differs from the original score.

In this area, Benson’s theory rests on his assertion that improvisation works ‘with the given in order to “create” something new’ (2003, p. 32). However, what is missing from Benson’s analysis is an appropriate consideration of the temporal aspect of performance as it relates to the improvisation concept. Changes to a score made by, ‘a conductor or an editor’ (Benson 2003, p. 27) occur before a performance takes place, and whilst the pre-existing has been transformed into something new, they can not be considered improvisatory in a musical context given that these changes have been made before a performance takes place. Expanding the definition to include this type of musical activity fails to acknowledge the vital temporal notion of real-time creation emphasised by Nettl, and concisely, albeit broadly, summarised by Dean:

‘Improvisation in music is the simultaneous conception and production of sound in performance’ (Dean 1989, p. ix).

1.6.2 A contextual definition of improvisation

I will now undertake an analysis of Nettl’s theory in order to refine his definition of improvisation for use in the context of this thesis. Given that this dissertation examines the practice of improvisation in three traditions – Hindustani classical music, jazz, and

Egyptian music – I propose that any definition of improvisation must incorporate a component that recognises the importance of tradition-specific parameters of performance, as these fixed aspects and structural devices facilitate the creation of non- fixed, improvised features. Whilst any definition need not specify the details of the improvised process in those traditions that are the subject of this thesis – an area that I

18 will clarify in Chapter Two – it should mitigate overly inclusive approaches that risk expansive amorphousness.

Considering more closely the second sentence of Nettl’s definition excerpted above, ‘To some extent every performance involves elements of improvisation’ (Nettl

2012): it is true that an element of spontaneity does direct the final outcome of most, if not all musical performances undertaken by live musicians. However, by examining the concepts of performer intention and audience expectation, further clarification regarding the identity of improvisation can be gained.

Take an example that would be placed at one end of Nettl’s continuum: interpreting and performing a pre-composed notated work in the Western art music tradition. This performance model involves the realisation of various score markings such as pitch, rhythm, tempo, articulation, timbre, dynamics, and other expressive techniques. Whilst pitch and rhythm markings are commonly considered to be fixed non-negotiable entities that must be realised accurately to produce a performance of the work in question, other aspects – such as tempo, rubato, ornamentation, articulation, timbre, dynamic nuances, and other expressive techniques – are less fixed as Solis emphasises (Solis, in Solis and

Nettl 2009, p. 9). Whether this is due to the fact that it is difficult, if not impossible for a composer to notate these variables in a way that allows them to be translated according to exact measurements, or whether it is the intention of a composer to leave these aspects open to the interpretation of performers, the fact remains that they are variable entities traditionally left to a performer’s discretion. As such, even these pre-composed pieces do take their final form during performance.

This being the case, a performer may intend these variables to be realised in a way that conforms to a premeditated design, or alternatively may intentionally leave

19 decisions regarding how they will perform these variable aspects to the inspiration of the moment.

With reference to the former, in one hypothetical performance, whilst intending the exact replication of a rehearsed interpretation, the performer may fail to reproduce these variables as intended. The shortfall in their performance, the spontaneous moments, mean that whilst the final form of the musical work is created as it is being performed, it occurs in a manner unintended by the performer. Consider for a moment the intention of the performer. In the course of rendering the composer’s score, the performer is constantly aiming to adhere to the score markings and is also continually shaping their performance to their premeditated interpretation. Their inability to do either, whilst spontaneous relative to their plan, is contrary to their intentions, and it is only spontaneous as a consequence of their lacking ability. Should their performance be considered improvised? Spontaneous perhaps, but not intentionally so.

Contrastingly, in a second hypothetical situation a performer intends to realise the final nature of these variables in the moment of performance, and once again the final shape of the musical work is determined during its performance. However, in this case the musician’s intention – that is, to perform the variable aspects according to the inspiration of the moment – aligns their performance more closely with an act of improvisation.

Even so, whilst the final rendering of a composer’s score occurs in the moment of performance, especially with specific regard to the more variable aspects of performance, it makes more sense, as argued by Sutton, ‘to reserve the word improvisation for more substantial choices being made at the moment of performance – choices of pitch and rhythmic structure whose difference from previous instances … are

20 intended by the performer to be apprehended by listeners’ (Sutton, in Nettl and Russell

1998, p. 73). Indeed, as Nettl himself asserts,

It is axiomatic that improvised music requires a greater creative effort on the part of the performer than does [the performance of] composed music, and, indeed, one way we may perhaps define improvisation is by measuring the degree to which the performer is creatively involved (Nettl 1974, p. 17).

The requisite degree of creative involvement should include, among other entities of structural importance and content, the aspects specifically identified by Sutton: ‘choices of pitch and rhythmic structure’ (Sutton, in Nettl and Russell 1998, p. 73).

Further clarification of this point can be gained through an examination of audience expectation. Informed audiences attending performances such as that specified above, a pre-composed notated work in the Western art music tradition, expect to hear a performer rendering a composer’s score. They expect to witness the realisation of an existing musical work. As such, the variable natures of entities – such as tempo, rubato, articulation, timbre, dynamic nuances, and other expressive techniques – whilst differing from one performance to the next, do not combine to create a new work, as these variables are expected when a performer renders a composer’s score. Their variability does not alter the work being performed in the ears of informed audiences, and as such, even if a performer intends to leave the interpretation of these variables to the moment of performance, an audience expects the differences within an expected degree of variability. Often variations might be appreciated by audiences, and at other times derided, but they are still accepted as an appropriate performance practice, whereas variations of pitch and rhythms would generally be frowned upon. As such, the final form of the work is expected to adhere to a particular range of interpretation within a small degree a variation, and so, whilst some variables are realised in the moment of performing the work, this process will not be viewed as improvisation in the context of this thesis.

21 Additionally, improvisation must also be understood as taking place within the defining parameters (constraints) of the genre in which performers place themselves.

These parameters place limitations on performance, but in constraining they also give structure and context to otherwise amorphous music making, and equally amorphous understandings of what is taking place. Freedom is relative, and the constraints that govern performance in each tradition essentially facilitate the creative originality that audiences expect of musicians by providing vehicles of departure and a means through which audiences can engage and identify novelty.

The final provision is that the novelty experienced during performances occurs in a spontaneous manner. Considerable simulated practice of the flexible features that accommodate originality may take place before a performance; indeed, arduous preparation is necessary. However, their final form is realised in real time, in the course of performance.

As such, developing Nettl’s definition excerpted above for the specific purposes of my musicological study, it is reworked to become in the context of this thesis:

Improvisation: The intentional creation of a musical work, or the final form of a musical work as it is being performed. Audiences witnessing the performance expect to hear substantial originality, such that the work, or a substantial part of it, is identified as new. The creation of novelty is facilitated by constraints that are embodied in the stylistic parameters of individual musical traditions. The existence of these constraints require musicians to adhere to a template of performance in which they manipulate, in real time, the flexible features of traditional performance models alongside recognisably fixed features that define their performance.

The parameters that define traditional performance models, which in turn determine the creation of novelty in the musical traditions under examination, will be the focus of

Chapter Two. The cross-cultural comparison of Chapter Six demonstrates that musicians in all three traditions learn what the customary rules and conventions of their

22 tradition are, and how they may balance tradition and originality when improvising, by internalising prototypes within a mimetic cultural practice.

Whilst now defined, I recognise that ‘improvisation’ continues to be an elusive and precarious term, and one of the most problematic for this thesis. The definition that I have provided is for the purpose of exposing the skill sets as practised in three specific traditions, and therefore the definition provided here may not necessarily satisfy universality. However, it conforms to the expectations of performance in the three traditions to be scrutinised. I will now proceed to other, more straightforward definitions and delimitations.

1.6.3 Professional musician

Within my analysis of how musicians learn to improvise, I use the term professional musician. This term carries with it the notion that musicians engage in this activity as their profession, earning their living through performing. For many musicians discussed throughout this dissertation this is the case. However, in my analysis the term professional musician is used in a manner that refers to someone possessing musical expertise, and whilst they may not earn their living as a full-time musician, they possess an ability to do so. Consequently I include these musicians when using this terminology.

Concurrently, whilst much of my research is derived from a retrospective analysis of educational experiences encountered by professional musicians, my analysis should not be viewed as being limited to information derived solely from professional musicians, nor should it be viewed as being only applicable to professional development. Though professionals are a focus of my dissertation, the developmental issues thus exposed often apply equally to amateurs and professionals given their

23 common origin as novice musicians, as methods of education run seamlessly into each other and do not discriminate predictively. This fact is particularly evident in the

Hindustani tradition, where educational experiences are often universal and musicians may have the thorough training of a professional without ever becoming one. Where I have observed specific developmental elements that impact on the attainment of expertise in all traditions, these are identified, particularly in Chapter Seven.

Furthermore, it is important to emphasise at this stage that in some cultural contexts, working as a musician is not necessarily the basis from which respect is earned. For example, in the Egyptian context, earning a living as a musician may involve accompanying belly dancers, and as such, drawing a wage in this occupation is considered disreputable by association. Moreover, it is not possible for many of the

Egyptian musicians that I interviewed to earn a living completely from performing music. In fact, it is rarely possible. Economic circumstances dictate a necessity to earn money from other sources, either related to music, such as instrument repair, or unrelated, such as being a taxi driver or chef. I emphasise that in reference to Egyptian music, as for the jazz tradition and Hindustani music, I use the term professional to mean possessing high-level expertise.

1.6.4 Autodidact

When discussing educational experiences, I use the term autodidact in the conventional manner, to refer to a person who is self taught, specifically in the context of improvisational ability. In doing so, I emphasise a distinction between time spent practising individually, where a novice is often revising and reinforcing pedagogical materials taught to them by a teacher, and autodidactic behaviour.

24 Moreover, for those musicians who are autodidacts, drawing a distinction between periods of time spent teaching themselves, and time spent reinforcing newly self- discovered information or techniques, is not so easily demarcated, though individual practice is obviously an important aspect of autodidactic development. Furthermore, I acknowledge that whilst a musician may learn from a teacher, they may also engage in autodidactic behaviour. Whether they would therefore be labelled as an autodidact is not an essential determination for the purposes of my dissertation.

1.6.5 Mode of transmission and learning environment

Borrowing the term from a medical origin, I use mode of transmission to refer to the manner or method through which information is transmitted or conveyed from one person to another. I label specific instances of this transference as a mode of transmission. Related to the transference of information is the situation in which this transference occurs. I label an educational relationship, institution, and any other enriching circumstance in which information is transmitted to novices as a learning environment.

1.6.6 Osmotic and osmosis

Associated with modes of transmission, I have similarly adapted osmotic from a scientific source to figuratively refer to a form of learning where information is passively internalised. That is, information is transmitted osmotically, without explicit teaching, and accordingly it is received via osmosis. I often use a metaphorical compound to describe a passive aural reception process, that is, ‘by aural-osmosis’, while also referring to passive processes as osmotic.

25 1.6.7 Mimetic and mimesis

Similarly derived from a biological origin, these terms are used to describe learning by imitation – that is, ‘internalised through mimetic behaviour’, ‘internalised mimetically’, or ‘by mimesis’.

1.6.8 Oral-aural tradition

Given that this thesis is examining musical traditions that were oral traditions historically, and to varying extents still are, I have employed the term oral tradition to refer to a tradition in which its conventions are transmitted without recourse to writing, and often verbally. Within this paradigm, osmosis and mimesis are active modes of transmission, as is explicit verbal teaching. However, in perceiving that this phrase only accounts for transmission, and whilst the phrase does imply that information is received by ear, I assert that it is important to recognise the biological attributes that are responsible for the perception and assimilation of information by ear more explicitly.

Additionally, I stress that this necessity is reinforced by the fact that information is often transmitted by musical instrument in these traditions, where a novice again receives this information aurally. Thus, throughout this dissertation in order to explicitly recognise the means of reception in the transmission process, I have employed the phrase oral-aural tradition in order to account for processes of transmission and reception.

Other terminology that is specifically related to theoretical and structural aspects of each musical tradition is discussed and defined in Chapter Two.

1.6.9 An instrumental and vocal focus

My analysis of improvisation in each tradition accounts for modes of transmission and related activities that occur in both vocal and instrumental music. The emphasis that I

26 place on either musical practice in each tradition is determined by the respective prominence of each. Thus, in reference to the jazz tradition, my emphasis is on instrumental music although the analysis does extend to consider vocal practices. In my analysis of the Hindustani classical tradition, given the primacy of vocal music in North

Indian classical music (Stevens 1975, Wade 2000, Slawek 2000a, and the prominence of vocalisation in instrumental pedagogy (Booth 1983), I consider vocal pedagogy as the archetype. Where relevant, I have included an examination of instrumental pedagogy simultaneously. In my examination of Egyptian music I have treated instrumental music with more emphasis, although educational practices in vocal music have also been documented to an extent.

When considering improvisation in instrumental music, my analysis is predominantly restricted to melodic instruments. While drawing upon some aspects of how percussionists learn to improvise, I have not undertaken a more significant examination primarily because I considered this was beyond the scope of this dissertation. I have included a discussion of some aspects of improvisation pedagogy in

North Indian tablā performance, but only so as to reinforce universal aspects of improvisation pedagogy in Hindustani music.

27 Chapter Two Contextual Explication

In this chapter I will contextualise my examination of what musicians do in order to develop the ability to improvise by explaining the distinctive frameworks within which improvisation takes place in the three traditions under examination. I will outline theoretical concepts that are pertinent to a fluent understanding of the investigations undertaken in Chapters Three, Four, and Five which explore how musicians develop the ability to improvise in Hindustani classical music, Jazz, and Egyptian music.

Specifically, I will outline the performance context and the tradition-specific musical parameters, particularly regarding how various components impact upon the creation of originality3 in these traditions.

From this basis I will also establish how the act of improvisation is conceptualised in each tradition. For the jazz and Hindustani traditions, I have included a concise amount of technical description, limited to information that I have determined necessary for the clarification of my argument. The information that explicates the Egyptian tradition is more comprehensive for two reasons. Firstly, a historical perspective is advantageous in order to establish that a contradiction exists between how Arabic music theory is conceptualised in the literature, and how performers verbalise various theoretical components and practices in Egypt today. Secondly, this historical context provides a developmental explanation for why performers who claim to be unrestricted during performances actually conform to an underlying stylistic template, albeit unknowingly.

3 By originality, I mean identifiable newness as apprehended and expected by informed audiences. This is mediated by tradition-specific criteria.

28 2.1 Hindustani classical music

A variety of vocal and instrumental genres exist in North Indian music. Whilst each has distinct performance characteristics, for the purposes of this thesis the overarching performance concepts that characterise the Hindustani classical tradition will be the basis upon which my analysis takes place. Given that vocal music has always dominated Indian music to the extent that the development of instrumental music cannot be separated from vocal origins (Wade 2000, p. 162, Slawek 2000a, p. 188), my analysis will view the performance context as practised in the principle genres of vocal music in North India as the archetype. In doing so, I will highlight those practices shared by instrumental music, whilst also indicating instrument-specific practices.

2.1.1 Rāga

Rāga: the central concept concerning how pitch interacts with other musical elements in

Hindustani classical music is renowned in Western literature for the elusiveness of its definition. Caution is taken in the literature to avoid presentation of the concept using

Western terminology that would see rāga converted to an abstract theoretical entity such as a Western scale or mode. Attempts to analyse the concept in an extracted fashion are viewed as being predominantly inadequate. Whilst these Western-derived perspectives do ‘convey something of the basics’, considered on their own they fail to present a complete view of the true nature of a rāga (McNeil 1993–1994, p. 70). As such, at the outset it is helpful to emphasise that a rāga can be viewed as only fully coming into existence in the totality of performance, and attempts to reduce it to component parts, whilst helpful, will always be incomplete until a performance occurs.

Meer emphasises that ‘Rāga has an ideational and a technical side’ (1980, p. 3), balancing the abstract imagery associated with a rāga’s identity, previously often

29 portrayed in painting or poetry, with an acknowledgement of theoretical musical components. McNeil differentiates this dual identity in terms of ‘tonal elements which are associated with the melodic formula of the rāg’ and ‘extramusical elements which generate the desired aesthetic effect or results’ (McNeil 1993–1994, p. 71).

The initial theoretical element in a rāga’s technical definition concerns the organisation of pitch. In Indian music, syllables known as sārgām are used to label individual pitches, assigned in a similar stepwise fashion to the European solfège moveable tonic system. Table 1 demonstrates a simplified description of the basic sārgām syllables. In addition, lowered versions exist for re, ga, dha, and ni, and a raised version of ma exists.

Table 1: Sārgām syllables

Swara: long-form names as Sārgām Pitch Western equivalent: do as tonic Sanskrit transliteration syllable number when swara are unaltered

sadjah sa 1 do

rishabh re 2 re

gandhar ga 3 me

madhyam ma 4 fa

pancham pa 5 so

dhaivat dha 6 la

nishad ni 7 ti

From this basis, a common tendency in the literature is to introduce a discussion of what a rāga is by listing various qualities of its structure, specifically regarding the organisation of the individual pitches as labelled in Table 1. Rao et al. (1999), Ruckert

(2004, pp. 55–56), Ruckert and Widdess (2000), Meer (1980, pp. 3–5), and McNeil

(1993–1994) are principle examples of this type of description, summarised as follows.

30 The structural features of rāga, as described by Rao et al. (1999), Ruckert and Widdess (2000), Ruckert (2004), McNeil (1993–1994) and Meer (1980)

There are a minimum number of pitches used in a specific rāga: at least five notes, one of which must be sa, and also at least ma or pa. From the remaining potential swara, various pitches are not used and are expressly forbidden. Their incorrect inclusion alters the rāga to the point of destroying its identity. Moreover, successive altered and non- altered forms of a pitch, such as re followed by lowered (komal) re, are rarely experienced. The collection of swara that belong to a certain rāga form an ascending and descending arrangement of pitches known respectively as the aroha and avaroha, or arohavaroh when combined: these arrangements may include or exclude particular swara in ascent or descent. These swara form the pitch inventory of a rāga.

From these pitches, a prescribed melodic movement (chalan), or one or more motivic phrases (pakad), are formed that identify the central musical ambience of a rāga. Whilst a rāga is more intricate and dynamic in performance than a summarised melodic movement can exhibit, the chalan is important in as much as it distinguishes a rāga from another rāga with identical or near-identical pitch inventories, emphasising the melodic hierarchy: which notes should and should not be emphasised, and which notes phrases should end on. The stronger pitch centres that are to be emphasised4 are known as the vadi, and theoretically, a perfect fourth or fifth away from the vadi, the samvadi.

Within the structure of the chalan, as formed from the arohavaroh, subtleties of

‘pronunciation’ (uccar) are definitively associated with a rāga. These are created through the use of ornamentation (gamaks), articulation, and the inclusion of microtonal intonation (shruti) for various pitches. Finally, moving beyond the technical, the effect 4 This emphasis is achieved through various means, so subtle, that definitions often do not account for the practice: a flexible mix of frequency, duration, location in phrase, and dynamic stress (the least important).

31 created by the combination of these theoretical elements results in individual rāgas being identified with a particular mood, and often specific rāgas are performed according to an association with various times of the day or season of the year.

In a creative manner, McNeil (1993–1994) summarises these elements by organising them and recommending a method for their combination in a manner reminiscent of a culinary recipe.

Melodic recipe

Ingredients: Primary – basic scale, mode. Secondary – vādī (the note of the rāg requiring greatest emphasis); samvādī (the note requiring second greatest emphasis); chalan (outlines hierarchy of notes); pakad (characteristic phrase[s]). Tertiary – gamaks (ornamentation); shrutis (microtones) McNeil (1993–1994, p. 72).

The method of combination (cooking) recommended by McNeil:

Prepare the primary ingredients according to established conventions. Temper these primary ingredients with the constrain[t]s of the rāg outlined in the secondary list of ingredients. Do not overcompensate the secondary ingredients in relation to the primary ones. Maintain the correct emphasis on the pūrvāng and uttarāng (lower and upper tetrachords, respectively) of the rāg. Within this framework add the tertiary ingredients to the above structure at the appropriate places in order to attain the desired flavour. Perform the melodic preparation against a suitable drone. Present the rag at the appropriate time of the day … (McNeil 1993–1994, p. 73).

Extending from these descriptions, and in acknowledgment of the fundamental fact that a rāga can only be fully represented in the act of performance, Ruckert and Widdess state that

a rāga is a map a musician follows in his or her creation of a musical performance: a catalog of melodic movements that the artist unfolds, details, and expands while following a traditional performance format that has been passed down orally from teacher to student (Ruckert and Widdess 2000, p. 66).

The expansion that they allude to is an important descriptor of the musical activity that a musician engages in during a rāga performance. Ruckert elaborates on this point more

32 precisely, describing the performance of a raga, and its expansion from the pre- composed bandish, composition, in detail.

When a musician performs a rāg in tāl [cyclical time], he or she bases the performance around a fixed composition (bandish), and then adds sections to it [expands the composition]. After singing or playing the complete bandish a few times, it may be shortened into a refrain, perhaps only the first line of the composition, or even only a few beats leading up to the downbeat of the tāl [beginning of the rhythmic cycle]. This foreshortened phrase usually shows a main feature of the rāg and is called the mukhrā (‘face’) … The mukhrā reappears frequently during a performance, often returning again and again (Ruckert 2004, p. 56).

It is this ‘expansion of the composition’ that is the essential ‘improvisational’ practice in North Indian music, and it takes up the great majority of performance time. Ruckert explains that the nature of continuing expansion or growth (badhāt) is determined by the specific traditions of individual performance genres, whilst also being affected by the circumstances of the performance (2004, pp. 56-57). He summarises that the sections added by a performer, who performs as a soloist over an accompanying drone and most frequently counterpointed by rhythmic accompaniment, are composed in the moment of a performance (Ruckert 2004, p. 56), and ‘usually proceed from slow tempos to fast, and are a meditation on the melodic design of the rāg’ (Ruckert 2004, p.

57). The nature of this continuing expansion occurs according to ‘compositional maps’ learnt by musicians from their teachers, individually characterised by the teacher’s style, as determined by their gharāna (school of playing) (Ruckert and Widdess 2000, p. 84).

2.1.2 Tāla

Unlike rāga, whose definition is not easily containable, tāla, the repeated rhythmic cycle in Indian music, does not suffer from elusiveness of definition. Each tāla has a fixed number of counts (matras) and is made up of two or more sections (Rao et al.

1999, p 7). This succinct definition may suggest simplicity, yet Kippen asserts that tāla

33 is ‘a complex theoretical concept’ (2000, p. 110), and Bailey agrees, stating that ‘the subdivision of this cycle can become quite complex’ (1992, p. 4). Complexity within each tāla assured, Bailey remarks on the variety of tāla, but concludes, ‘Although there are probably over one hundred tāla available to the Indian musician there are only about a dozen in general use’ (1992, p. 4). These rhythmic cycles, substantially internalised by both melodic soloists and audiences and outlined by a percussion accompanist, are the temporal foundation over which the metred section of rāga performances takes place, and through which the expansion of the rāga and associated improvisatory practices unfold.

2.1.3 The ensemble context

The accompanying drone over which most vocal or instrumental soloists undertake the expansion of a rāga is provided by the tambūrā (or electronic equivalent), a stringed instrument that resonates with the fundamental notes of the given rāga: the tonic, and either the fifth, the fourth or the natural seventh. In dhrupad,5 a vocal genre recognised as the oldest in Hindustani classical music (Wade 2000, p. 162), the soloist sings with only tambūrā and drum accompaniment on the two-headed pakhāwaj. In the younger khyāl6 tradition, a tambūrā’s drone also accompanies the solo vocalist and rhythmic accompaniment is provided by the tablā. Musicians playing melodic instruments such as the bowed sarangī and violin, as well as the keyed harmonium, which like the violin has been adopted from the West, also accompany the khyāl vocalist. These instrumentalists ‘complement the vocal line and provide continuity between breathing pauses by playing slight variations a split second behind the soloist’s improvisation’

(Wade 2000, p. 171).

5 The term dhrupad is derived from dhruva (fixed) and pada (words). 6 The term khyāl refers to ‘imagination’.

34 The most common instruments played by soloists in Hindustani classical music are the stringed sitār and sarod, and the bansuri (bamboo transverse flute). Demonstrating the aforementioned influence of vocal music, these instrumental soloists are accompanied by a tambūrā’s drone and by a percussion accompanist. At times multiple melody-producing instrumentalists perform together, though one of these will normally assume the position of soloist, whilst the others accompany (Slawek 2000a).

2.1.4 The performance structure and associated improvisatory practices

Ālāp, bandish, nom-tom, bolbānt, laykari, mukhrā, tihāī, and tān

A variety of improvisatory practices are associated with genres of North Indian vocal and instrumental music. Certain techniques are practised across many genres, and others are more restricted in their use. I will now detail a variety of improvisatory practices as they relate to the principle vocal genres of Hindustani music, dhrupad and khyāl. As the prominent archetypal modes of performance in North Indian music, these vocal genres employ a variety of improvisatory practices that are used in other types of vocal music.

Furthermore, due to the influence that these notable forms have had on instrumental music (Wade 2000, Slawek 2000a), they are also emulated, where technically possible, in these non-vocal contexts. After illustrating these practices in a vocal context, I will draw attention to those that are practised in instrumental performance.

In the dhrupad genre, melodic improvisation begins from the outset of a performance in a lengthy unmetered section known as ālāp, in which a singer introduces the scale, melodic motifs and musical attributes of the rāga being presented.

Improvised phrases are cyclically closed by a recurring melodic formula that is known as a mohrā. Eventually the soloist moves to a section called non-tom ālāp, drawing

35 attention to the element of rhythm by enunciating repeated vocables on repeated pitches, developing a clear sense of pulsation without any explicit metrical cycle (Wade 2000).

The unmetered ālāp eventually moves into a metred bandish (composition), and the drummer will now accompany the singer. Notably, bandish are flexible relative to

Western compositional concepts (Scott 1997, pp. 183–184). During the exposition of the bandish, improvisation takes place, primarily focussed on generating rhythmic complexities. Wade emphasises that ‘The singer must always articulate the composition clearly; but expressing the meaning of the text through melodic improvisation is far less important that using the text for rhythmic purposes’, a practice known as bolbānt (2000, p. 169). Improvisation observed in ‘a part of a performance dominated by rhythmic play, particularly complex rhythms’ (Kippen 2000, p. 111) is known as laykari.

In the khyāl genre, singers employ a greater variety of improvisatory practices.

However, an important distinction between khyāl and the older dhrupad form is the absence of a lengthy ālāp introduction. After briefly introducing the rāga’s multifaceted identity in unmetered time, a shorter bandish is presented in metre. This composition provides a cadential figure that is eventually used when improvising, known as a mukhrā, whilst also providing the textual basis of the forthcoming improvisation. Wade identifies six types of improvisation associated with khyāl, some used in dhrupad, others unique: ‘bolālāp, nom-tom, bolbānt, two types of tān (ākār and boltān), and sārgām’ (Wade 2000, p. 173). Wade also asserts that these are used to structure performances, and are generally employed in an order that calls attention firstly to melody, then to rhythm, and then to speed (2000, p. 173). Wade explains that the bolālāp is a slow ālāp section of improvisation used in slower khyāl performances using text from the previously introduced composition to render the rāga’s mood and characteristics. Unlike ālāp in dhrupad, it is metred, as are nom-tom and bolbānt (2000,

36 p. 175). Of specific significance to the khyāl genre is the use of virtuosic melodic passages known as tān, where a singer either uses vowels unrelated to the words of the compositions (ākār tān) or syllables derived from the composition’s text (boltān)

(Wade 2000, p. 175; Ruckert 2004, p. 60). At any time during a rāga performance a singer may also employ the sārgām syllables to vocalise their improvised melodies, known as sārgām tān, when singing the characteristic virtuosic passages of the khyāl genre (Ruckert 2004, p. 60).

Another facet of improvisatory practice, associated with tāla, are tihāī. These are experienced during performances

when a rhythmic or melodic phrase is repeated three times … calculated to end on the downbeat, but sometimes tailored to come to the beginning of the composition (since not all compositions start on sam) [the first beat of the tāla] (Ruckert 2004, pp. 46–47).

This practice is used in vocal traditions such as khyāl (Wade 1984) and is also an important element of instrumental improvisations (Slawek 2000a, p. 200). At this point,

I will make the retrospective note that much of the preceding discussion relating to improvisatory practices in vocal music applies to instrumental contexts. Slawek (2000, pp. 198–202) draws attention to the fact that instrumental improvisations in the North

Indian context derive their form from vocal origins, particularly dhrupad and khyāl.

Consequently, those improvisatory practices such as the slow ālāp exposition feature also in instrumental performances that draw context from the dhrupad tradition. Slawek also notes the influence of the khyāl tradition, not only in the use of tihāī, but also in the tān of instrumental improvisations (2000, p. 200). In the instrumental context, improvised passages that are not classified as tān are referred to, more globally, as upaj: passages of improvised expansion and elaboration of the rāga. Tihāī are used to conclude both upaj and tān passages. Within this framework, lāykarī again describes

37 moments of complex rhythmic play in instrumental improvisation (Slawek 2000a, p.

200).

2.1.5 The improvisational process

Performances of rāga in Hindustani classical music occur through the interaction of pitch and rhythm within the traditional practical dictates of rāga and tāla. This being so, the gradual expansion of the presentation of a rāga, conducted in the course of a performance, transpires as a musician guides the presentation’s growth and development according to traditional performance parameters, whilst engaging in a dynamic process balancing tradition and novelty. Indeed, the crux of this interplay goes towards understanding exactly what the improvisational process is that Hindustani musicians engage in, a process so subtle that some claim it is misleading to align it with other improvisational practices. Meer asserts that:

This variation has been called improvisation sometimes, suggesting wrongly that everything other than the composition is invented on the spot, within the limits of the rāga. The truth is that most aspects have been precomposed (though unwritten), often handed down by the teacher, and that the word improvisation is only used because there is a degree of recombination and a certain freedom in distributing e.g. an alāpā over time (Meer 1980, p. 143).

Indeed, using the descriptor ‘improvisation’ does carry with it some necessity to qualify the degree to which performer freedoms exist, and Meer clearly feels that the process at work in Hindustani music is better understood as one of recombination with temporal freedom, rather than the carte blanche creative freedom that a mainstream, if not misinformed understanding of ‘improvisation’ may encourage. There are certainly

‘compositional maps’ (Ruckert and Widdess 2000, p. 84) that performers follow, including learnt methods of recombination and elaboration. Paraphrasing Neuman

(1980, p. 23), Napier specifically identifies this variety of activity:

38 Existing phrases are stretched or compressed, and the same may happen to motives from the phrases; further motives may be prefixed, infixed, and suffixed. Phrases may be broken up or telescoped with others, and motives or phrases may be sequenced through different registers (Napier 2006, p. 5).

It is this practice that Neuman asserts is at the heart of the improvisational process: ‘The intricacy of the structure of these musical elements and the ingenuity of the architectural assembly establish the degree of creativity and inspiration in a performance’ (Neuman

1980, p. 23).

Writing directly concerning the interplay between novelty and tradition, Napier suggests that the value affixed to a performance may

depend as much, if not more, on the quality of the pre-composed materials, and on the accuracy of their re-presentation, than on the novelty and inventiveness of a performer in re-presenting them (2006, p. 5).

Asserting that simple descriptions of Indian music as predominately improvised are lacking, given that the term improvised ‘focuses on process and difference rather than the re-presentation of fixed material’ (Napier 2006, p. 5), Meer’s earlier contention is in part explained by Napier’s analysis of pre-existing definitional inadequacies. Equally,

Bailey’s excerpted conversation with Viram Jasani speaks directly to this issue:

And, of course, a rāga can be considered a limiting thing. How, after all, do you recognise a rāga? Because you recognise certain characteristic features about it. And if you are going to play that rāga you can’t help but play those characteristic features. So this, perhaps, is not improvisation (Jasani, quoted in Bailey 1992, p. 9).

So, the question remains, should the term improvisation be used to describe the rāga process of Hindustani music at all? Whilst Meer recommends caution to the point of prohibition (1980, p. 143), Jasani uses the term, asserting that

your improvisation comes into play when you are trying to use the information presented to you in terms of musical facts, using your ability, and the experience acquired over the years of practicing that rāga, and listening to other people play that rāga, to put all this together and create some new phrases or put a new idea within that rāga (quoted in Bailey 1992, p. 9).

39 More cautious, yet not as proscriptive as Meer, Napier summarises the process as ‘a dialogue between the demands of reproduction and distinctiveness, intergenerational continuity and contemporary subjectivity’ (2006, p. 12), and more specifically suggests that whilst derogatory uses of ‘improvisation’ have already been invalidated,

What is still current is the need to avoid using the word in such a way as to only validate novelty, distinctiveness, and subjectivity without acknowledging conservatism, reproduction, and continuity (Napier 2006, p. 12).

Napier is reinforcing that the process at play in a rāga performance involves delicate originality that is closely tied to the process of reproducing tradition, or as coined by

Napier, ‘re-presenting materials so acquired’ (2006, p. 11). As such, Napier argues for a

‘culturally appropriate understanding of improvisation’ that acknowledges ‘the concept of subtle variation’ (2006, p. 12).

Whilst Meer staunchly suggests that most commonly, variation from one performance to the next by the same musician ‘is purely recombination and cannot be considered change within the Indian context’ (1980, p. 143), his conservatism in this regard appears to be a reaction to the use of the term to indicate ‘a real novelty during a concert’ (1980, p. 43): something completely new. Whereas if the term ‘improvisation’ is used in accordance with the limitations established by Napier, thus incorporating ‘the concept of subtle variation’ (2006, p. 12), with care, the term can be applied to

Hindustani rāga performances. The ‘recombination’ identified by Meer is one such vehicle of ‘subtle variation’ highlighted by Napier.

Rāga performances are unique ephemeral musical experiences, and as claimed by

Jasani (in Bailey 1992, p. 9), they do include improvised moments, however subtle, achieved by using methods of recombination: addition, reduction, variation, interpolation, and synthesis. Reflecting back to the definition of improvisation that I proposed in Chapter One, during a rāga performance the final form of a music work is

40 intentionally created as it is performed, where audiences witnessing the performance expect to hear originality to the extent that the work is considered new in part, and therefore distinct from other performances of the same rāga. This novelty is specifically identified through the interaction of fixed and flexible features of the music, and it involves gradual expansion from a pre-composed bandish that occurs via thoroughly practised processes of recombination.

2.2 Jazz

For readers who are acquainted with Western music traditions, the jazz performance context may be familiar. Even so, in order to contextualise the research presented in

Chapter Four, thereby refining definitional issues relating to the performance context and the practice of improvisation in the jazz tradition, various structural features will now be summarised.

The twentieth and twenty-first centuries have witnessed a wide variety of genres develop within the jazz tradition, each exercising a distinct approach to the interplay between fixed and flexible features of traditional performance models. Indeed, individual responses to the jazz performance context are celebrated as an essential aspect of the tradition. Whilst acknowledging this diversity of approaches, given that differing interpretations have stemmed from a common tradition, for the purposes of this thesis the performance context and approach to improvisation that will be outlined will be a traditional one: the conventional centre of an expansive spectrum. The definition of traditional, in this context, will become apparent throughout the course of my ensuing analysis.

41 2.2.1 The traditional performance context

The foundation of the performance model in the jazz tradition are songs, often selected from a repertoire known as ‘standards’, a collection of compositions deemed standard by their favoured adoption by musicians as vehicles of departure in the jazz performance model.7 Berliner confirms that these pieces, ‘consisting of a melody and an accompanying harmonic progression, have provided the structure for improvisations throughout most of the history of jazz’ (1994, p. 63). Originally adapted from spirituals, marches, rags, and popular songs (Berliner 1994, p. 63), this standard repertoire also includes original compositions expressly composed by jazz musicians for performance and recording, later ‘standardised’ due to their popular usage by other jazz musicians.

Often these original compositions are adopted into standard repertoire because they are representative of important stylistic developments in jazz, and as such, are deemed important to learn and perform by later generations of musicians. Obvious examples include Miles Davis’ landmark modal jazz composition So What, and John Coltrane’s virtuosic Giant Steps. Compositions freshly created that adhere to the technical (often harmonic) features of particular song types are also considered part of this repertoire, even though they have not yet become standards.

Beyond these pieces customarily played by musicians, the repertoire chosen to perform can include any composition that includes a harmonic progression. The practice of adopting popular songs and Western classical repertoire into jazz repertoire has continued to the present day, witnessed in Charlie Hunter’s Bob Marley project Natty

Dread, the pianist Brad Mehldau’s performance of songs by the internationally

7 Keightley (2001) emphasises that standards, in addition to being musical artifacts, hold a significant cultural value that is greater than their functional role as vehicles for improvisation in jazz. I assert that their use in jazz further reinforces the power they already hold, as objects that move beyond temporal limits to become timeless cultural artifacts (Keightley 2001).

42 acclaimed ‘alternative rock’ band Radiohead, and pianist Uri Caine’s adaptions of works by Mahler, Mozart, Schumann, Wagner, and J.S. Bach.

A vital aspect of any composition chosen as a vehicle of departure is a cyclical form having a relatively short structure, often between 12 and 32 bars. This allows for the practice of improvisation to occur over a recurrent harmonic progression, a practice which DeVeaux and Giddins claim derives from concepts of form in African music where improvisation occurs over a repeated rhythmic drumming cycle (2009, p. 31).

In jazz, the melody and associated harmonic progression are known as the ‘head’, and when the harmonic progression is repeated cyclically during improvisations, this cyclical component is known as the ‘chorus’. Jazz musicians improvise ‘over’ one or more choruses, and in doing so they replace the original melody with new musical material. The harmonic progression over which this new musical material is played can theoretically be based on any cyclical combination of chords, but in practice the progression is often based on variations of traditional forms such as the 12-bar blues.

The 32-bar AABA form is another common structure, but unlike the blues, it does not imply a particular harmonic progression as such.

Evidencing the influence of European music theory, jazz chords are built on a triad.

Combinations of absolute pitch letter names and the chord types of Western music are used to descriptively outline and prescriptively notate these chord progressions.

Extensions are routinely added to this chord, such as the seventh, ninth, eleventh, and thirteenth, all existing in lowered and raised versions as dictated by individual chord symbols penned by a composer. Roman numeral labelling, derived from Western music theory, may be used to analyse chords diatonically, but is not used to prescribe chords for realisation in performance.

43 Once the melody of a given song has been played, the musicians who play instruments that provide the harmonic and rhythmic basis (bass, drums, and piano and/or guitar) continue to outline the harmonic progression, and musicians who were playing the melody alternately improvise, playing a new melody line that adheres to the passing chord progression, though melodic improvisation is also undertaken by piano and bass. This new melody line realises the passing harmonic movement according to major-minor chord-scale theory, just as the original melody line encapsulated the passing progression as its upper voice. Performances traditionally end with a restatement of the original melody, creating a cyclical close.

2.2.2 Improvisation

Whilst methods of improvising this new melody line vary from musician to musician, two main approaches with regard to pitch may be delineated. In the first, improvisation addresses each chord of the given progression individually. This translates to musicians needing to respond by playing new pitches related to the specific harmony of passing chords in a stylistically appropriate way. The density of notes in the new material will almost always be greater that the notes of the original melody, and in any case, the successful navigation of the passing chords requires a fluent, agile technique. In the second approach, where possible, the overall harmony is summarised to a single pitch centre, or if not, to a more limited number of key centres. This is possible when the individual chords of a progression can be analysed in relation to a single tonic, using the corresponding tonic scale(s) to improvise over the entire progression.

Regardless of how musicians address harmonic progressions, either as individual chords or as a summary of a more limited number of key centres, the pitch material sourced by jazz musicians when improvising can be conceptualised vertically or

44 horizontally, a commonly identified dichotomy in discussions concerning jazz improvisation theory. In the first, the vertical approach focuses on the individual pitches of each chord, which are viewed as the most consonant and thus emphasised. Other pitches are treated as passing tones, or avoided altogether. Improvising in this chord- oriented way encourages musicians to outline the specific identity of each chord, thereby reinforcing the underlying key centre. This is in contrast to the horizontal approach, which advocates a translation of chords to modes: stepwise octave-bound scales derived from major and minor scales. This chord-mode theory focuses on scale- oriented stepwise movement, thereby introducing a greater number of potential pitches for each key chord, though in doing so this inherently detracts from the focus on individual chord tones as pitch material choices. Whilst these two approaches can be identified as separate conceptualisations, it should be emphasised that musicians do not necessarily choose one approach over the other. Rather, musicians commonly integrate both strategies when addressing pitch choices whilst improvising.

The cyclical nature of jazz improvisation is characterised by the harmonic progression flowing with perpetual momentum. This is possible because the rhythmic aspect of the performance context is fixed, according to a set number of pulses per bar, represented in the time signature. Whilst it is possible for time signature changes to occur within one composition – theoretically, the metre may be organised according to any design – the ‘standard’ repertoire performed by jazz musicians most commonly contains compositions that are organised into unchanging meters of four, three, or two beats per bar, and less commonly into six beats per bar. As the compositional success of improvised sections may depend on fluency and momentum, which is more difficult to achieve when time signatures change regularly, such changes are not a common phenomenon in the jazz tradition.

45 2.2.3 The ensemble context

The interaction of rhythm, pitch, and harmony are materialised through the interplay of instrumentalists in the jazz ensemble. Whilst heterogeneity abounds as to the exact instrumentation of bands, the basic configuration is based on the division between the

‘rhythm section’ and the ‘front line’ or ‘soloists’, although this implied division of labour is somewhat misleading given that instruments in the ‘rhythm section’ often solo in the course of performances, and some ‘soloist’ instruments perform an accompanying role at times. These exceptions identified, the division will be maintained for this analysis.

The ‘rhythm section’ usually consists of a bass instrument, which is normally the double bass, but can actually be any instrument capable of providing the lowest voice of the harmonic progression. At times, the pedals or left hand of organs have fulfilled this role, as has the tuba, or indeed the left hand of the piano. When the piano is free from bass note responsibilities, it is often included in the instrumentation of jazz ensembles to provide harmonic accompaniment, abbreviated to ‘comping’ in jazz parlance. Instead of, and sometimes in addition to the piano, the guitar is also featured to provide harmonic support to soloists. Finally, the drum kit, and possibly a larger percussion section complete the ‘rhythm section’, interacting with the other rhythm instruments to propagate the harmony of compositions whilst creating a flow of perpetual motion, itself substantially improvised in detail, for improvisations to be performed over. The drummer often contributes to this rhythmic momentum by playing a swing-beat cycle, where each beat of one bar of music is further divided into two sub-beats, and played in an uneven ratio of approximately 2:1. In a metre of four beats per bar, this cycle is often notated for the ride cymbal as in Figure 1. The exact ratio can vary substantially: at different speeds, from drummer to drummer, and in fact from bar to bar of the same

46 drummer’s playing (Prögler 1995; Friberg and Sundström 2002). Furthermore, this rhythm is often varied and interspersed with other note values within the cycle.

Figure 1: Swing rhythm (simplified)

In addition to the common swing cycle and other related variations of this pattern, rhythmic configurations underpinning the momentum of performances also include cycles based on even subdivisions of the bar: ‘straight’ patterns, prevalent in sub-genres of jazz such as Funk, Fusion, and Latin Jazz.

The ‘soloists’ or ‘front line’ includes instruments such as the trumpet, saxophone, and trombone. A vocalist will also often lead an ensemble, and in the case of a piano trio, the pianist will act as leader with bass and drums completing the group. As in other aspects of the jazz tradition, diverse arrangements exist. Often a jazz quintet includes saxophone, trumpet, piano, bass, and drums, whereas a quartet would usually only include one ‘soloist’ instrument. In addition to the timbres created by these instrumentations, trumpet players often also perform on flugelhorn, cornet, or other brass instruments, and saxophonists often also play the flute or clarinet. If a guitarist leads an ensemble, they may replace other soloists, fulfilling the dual role of chord player and soloist as needed. In drawing these examples, I feel the need to emphasise once again that whilst these configurations are common in the tradition, there are no formalised conventions concerning instrumentation.

Solo guitarists and pianists often self-accompany by outlining a harmonic progression whilst simultaneously performing an improvised melody, and drummers continue Max Roach’s legacy in performing solo drum improvisations in concert.

47 Departing from harmonic reference, saxophonist Coleman Hawkins is identified as the first musician to record an example of a solo improvisation without reference to an underlying chord progression, recording Picasso in 1948 (Hawkins 1998), Porter (1998, pp. 72, 313).

Solo improvisation is not common in Hindustani music, where even if no rhythmic accompaniment is used, a drone always accompanies without an element of interaction.

Contrastingly, the unaccompanied soloist is prevalent in Egyptian music, the details of which I will now outline.

2.3 Egyptian music

As introduced in Chapter One, this dissertation examines Egyptian music considered al- mūsīqā al-’arabiyya (Arab music) by its adherence to the conventions of Arabic music, and I will now examine sources that consider conventions of Arabic music theory. This analysis initially proceeds from a historical perspective. I have chosen to begin with this retrospective theoretical evaluation, which contrasts to the modern-day perspectives of the jazz and Hindustani traditions presented above, in order to establish that modern-day

Egyptian performers conceptualise and discuss Arabic music theory in a manner that contradicts historical records. The exposition of this contradiction may seem irrelevant given that my dissertation is examining how musicians learn to improvise in Egyptian music today, but the resolution of this conundrum is essential when placed in the context that sees these historical records echoed by a cross-section of modern literature.

The opinion that contradicts both historical records and modern literature is epitomised by the often-declared freedom that Egyptian performers assert, claiming they are not bound by rules of performance. Similarly, their theoretical descriptions of musical elements and practices are often minimalist. However, my research detailed in

48 Chapter Five establishes that the improvisations of modern-day Egyptians adhere to stylistic parameters that are more involved and constraining than performers assert.

In addition to establishing that Egyptian musicians have streamlined their theoretical definitions in contrast to historical theory and current literature, this historical explication provides a context for the template-based performance model I witnessed in Cairo, even though performers profess they are not bound by such constraints. This historical analysis also enabled a correlation to be discovered between the historical record and modern-day literature, which only further reinforced my assertion that Egyptian improvised performance practice is undertaken within a detailed theoretical system. Ultimately I do not suggest that modern-day performance practice finds its exact theoretical origin in the historical records I analyse, but I do suggest that these historical records witness an organised performance practice that has evolved into a similarly organised fashion to the present day, even though these parameters are neither explicitly articulated nor readily identified by the vast majority of Egyptian musicians.

This exposition therefore focuses on those sources that provide a detailed examination of the components employed during an improvisation. Moreover, through this investigation I will provide a coherent portrayal of how musical elements interact in the course of improvised performance, thereby establishing how improvisation should be conceptualised in Egyptian music, balancing the apparent contradictions between historical theory and modern practice.

2.3.1 Taqsīm and mawwāl

An improvised performance in the Arabic tradition is known as a taqsīm (plural: taqāsīm). The word means ‘division’ in Arabic, which appropriately describes the

49 characteristic separation of phrases that occurs throughout a taqsīm. A taqsīm is described in the literature as a performance in which ‘The Arab instrumentalist is … accorded the opportunity to improvise his own creations …’ (Marcus 1993, p. 108).

Mawwāl (plural: mawāwīl), means ‘associated with’ or ‘connected to’, and is a genre of vocal improvisation.

Marcus asserts that the duration of an individual taqsīm is determined by ‘the amount of time allotted to the performer, and also by the performer’s mood at the time of performance’ (1993a, p. 108). Taqāsīm generally may last anywhere from three to ten minutes, and are sometimes, though rarely, longer. This length contrasts to the far longer performances of North Indian classical music that can last up to 45 minutes for vocalists and 60 minutes for instrumentalists. As such, musicians must pace the development of the rāga. Improvisations in the jazz tradition can be restricted or elongated. In the bebop period, the length of recordings was limited by the technology being used, and as a result soloists needed to restrict the length of solos in recorded contexts. Contrastingly, on stage, and with more recent technology, improvisation can last extensive periods of time in the jazz tradition, sometimes up to 45 minutes or more, as determined by the individual choices of performers.

Marcus stipulates that ‘Individual taqāsīm8 are not simply free-form products of the instrumentalist’s fancy; instead, the instrumentalist improvises according to a complex set of pre-established rules and conventions’ (1993a, p. 108). The nature of these rules and conventions will be analysed within the subsequent discussion that details the musical context in which improvisation takes place in Egyptian music.

A taqsīm may take place as a freestanding musical presentation, as part of a larger instrumental composition, or alternatively during a vocal performance where it serves as

8 Marcus (1993) does not differentiate between singular taqsīm and plural taqāsīm. Racy (2000) and others do make this distinction, as I have also.

50 an instrumental interlude. An unaccompanied instrumentalist can perform a taqsīm.

Solo instruments of the taqsīm genre include the ‘ūd (Middle Eastern fretless lute), the

European violin (kamanja), the nāy (a reed flute), and the qānūn (a Middle Eastern horizontal box zither). Alternatively, a taqsīm can be performed in a variety of accompanied settings, including a duo, a trio, within a takht chamber ensemble, or within a larger traditional orchestra. Takht ensembles consist of four main melodic instruments (‘ūd, nāy, violin, and qānūn) with one or more percussion instruments: a riq

(tambourine), Arabic tabla (goblet drum), or bendir (frame drum).

In accompanied settings, often improvisations are played over a single-pitch drone without rhythmic accompaniment, and therefore have relative temporal freedom. Where a repeated bass line or drone supports an improvisation, this is often provided by an accompanying ‘ūd or violin in smaller ensembles, or by a ‘cello or double bass in traditional orchestras. Alternatively, improvisations can also occur over a repeated, pitched ostinato, or within a repeated percussion cycle. When an improvisation is undertaken within temporal parameters, it is known as taqsīm ‘ala al-wahda (literally taqsīm ‘on the beat’). Vocal improvisations, mawwāl, often occur after an instrumental taqsīm or short instrumental composition (dūlāb). These serve as a modal introduction to the mawwāl, which itself often precedes the performance of a vocal composition.

Singers demonstrate their virtuosity during these performances, and if accompanied, instrumentalists summarise their phrases directly after they have been sung, a similar practice to that of melodic accompanists in the Khyāl genre of Hindustani music.

Returning to an analysis of the aforementioned rules and conventions alluded to by

Marcus (1993, p. 108), Shiloah draws attention to the difficulty of ascertaining an accurate concept of what Arabic music may have sounded like before 1900, due to a distinct lack of ‘any concrete musical document earlier than the first recordings’

51 (Shiloah 1981, p. 19). Acknowledging that the first serious document that expounds upon Arabic music theory was written sometime approximately in the 50 years preceding 967 (Shiloah 1981, pp. 29–30), he asserts that great care must be taken in relying on other documents. He claims interpretative challenges exist due to ‘vagueness of definition, inconsistency in the use of a fluid terminology, and the likelihood of mistakes made by copyists who did not always understand what they were copying’

(Shiloah 1981, p. 20). Shiloah’s point is taken; it is difficult to know what Arabic music sounded like before 1900, and most importantly, care must be taken when drawing upon the written theoretical sources that exist, as with any written historical source.

Providing more detail, Scott Marcus, one of the most prolific American writers on

Arabic music theory, also draws out various flaws in sources, though his contentions rest more on the details of these sources, such as flawed representations of the quartertone scale, rather than emphasising the generalised inadequacies that Shiloah points to

(Marcus 1989, p. 46). Importantly, Marcus was able to put any flaws in sources to one side and deduce how the aspects of Arabic music theory that are employed in performance situations, such as taqsīm, have evolved to the present day.

Marcus defines taqsīm in a dualistic fashion, referring to both the construction of performances and also pitch or modality as significant structural concepts. ‘A taqāsīm9 is multi-sectional, with sections separated by moments of silence’ (1993a, p. 108).

Marcus identifies pitch material, or rather the modality of each section, as the factor that draws the multiple sections of a taqsīm together. ‘The musical coherence of each section is achieved by the instrumentalist focussing on one melodic idea, usually a specific melodic mode’ (1993a, p. 108). Habib Touma describes the process in a similar fashion, noting multiple musical sections (‘phases’) separated by silence (1996, p. 39),

9 See footnote 8.

52 whilst also recognising the importance of modality in the performance process: ‘The taqsīm is an instrumental realization of the modal framework …’ (1971, p. 43).

Performer and ethnomusicologist Ali Jihad Racy also reinforces this modal aspect; he views taqāsīm performances as being ‘resolved in accordance with an established modal plan’ (2000, p. 308). Similarly, Racy describes taqāsīm performances as following ‘relatively flexible structural tracks’ (2000, p. 309) whilst also asserting that various publications stipulate that during a taqsīm, a performer is expected to follow a certain melodic shape, or rather follow the sayr (path) of the mode (2000, p. 309). He describes the sayr as a ‘brief description of how each mode is expected to unfold’

(2000, p. 309). Thus, whilst the various phrases of a taqsīm do indeed contribute to taqsīm performance structure, the literature agrees that modality is the central and most significant concept that motivates, characterises, and gives identity to each taqsīm performance.

2.3.2 Maqāmāt

The modality or more specifically the modes to which these writers refer are known in

Arabic as maqām (plural, maqāmāt), however the use of the term needs to be clarified as there are various interpretations regarding the interplay between maqāmāt and their role in performances such as taqāsīm. Modern literature indicates that in examples of modern-day Arabic music theory, ‘the vast majority of maqāmāt are conceptualized as single-octave structures that duplicate at the octave’ (Marcus 1987, p. 39) and in current pedagogy, maqāmāt are presented as Western scales when notated. However, Marcus and other writers emphasise that the maqāmāt are theoretical concepts of more complexity. This apparent contradiction begs the question: is this disparity due to a

53 change in performance practice, or is it the result of a simplification of Arabic music theory modern pedagogy?

Addressing this issue, Marcus hints at this complexity in his statement ‘Each maqām has a unique scale and special melodic features’ (1993a, p. 108, my italics), and

Ayari and McAdams echo this perception:

A maqām is more of a schema than simply a scale. It is a process of melodic movement, and an operational protocol of improvisation according to the models and forms of melodic and rhythmic organization. It is a complex system that implies the illustration of its structures in modal melodic development. On the continuum between abstract scale structure and tune families, the maqām is clearly closer to the tune families … (Ayari and McAdams 2003, p. 164).

Carmi-Cohen extends this type of definition by way of comparison, asserting that historically the maqām ‘are generally understood as melody types, similar to the Hindu rāgas and the Greek nomoi’ (1964, p. 102). Indeed, her research confirms that whilst

‘modern theory tends to treat them quite abstractly as scales in the Western sense’

(Carmi-Cohen 1964, p. 102), ‘the maqām must be regarded as a melody-type and not simply a scale’ (1964, p. 104).

Marcus (1993, p. 108), Carmi-Cohen (1964, pp. 102, 104) and Ayari and McAdams

(2003, p. 159) assert that each maqām is represented by characteristic melodic movement, rather than by a simple ascending and descending scale. Ayari and

McAdams refine this concept generally by stating that there are ‘essential melodico- rhythmic configurations that are emblematic of the maqām’ (2003, p. 159), and continue this conceptual thread in more detail: ‘The concatenation of repetitive musical motifs and melodic cells composes the internal fabric of the maqām’ (2003, p. 161). However, the fact remains that these procedural explanations contradict how maqāmāt are conceptualised in modern-day Arabic music theory, the details of which will be described in more detail henceforth.

54 Nettl and Riddle propose a continuum-based definition as the most useful conceptual medium. From opposite extremes on this hypothetical continuum, it is proposed that a maqām can be viewed as an object, or as a conceptual modal framework. Nettl and Riddle use maqām nahawand as an example, and suggest that from one perspective it can be viewed as ‘a weighted scale from which tones are selected for performance and composition’ (1998 p. 372). Maqāmāt are conceived from this perspective in modern Arabic music theory (Marcus 1987, p. 39). However, the definition is also extended to a point on the continuum where the term ‘maqām’ is viewed as a concept that expresses the established relationships that exist between each of the notes in the scale set and how these notes are combined in customary patterns during performance (Maqam World, 2001–2005). As such, the definition is still obscured.

Reinforcing this definitional intricacy, Nettl and Riddle clarify the situation:

Most accurately, but also with great difficulty, it [maqām nahawand] would have to be defined in terms of all of the performances, improvisations, and compositions which are regarded as being cast within it. All of the musical events from scale, ornamentation, and acceptable accidentals, to typical modulation patterns, norms of section lengths, relationships of all sorts among the sections (1973, p. 14).

This continuum definition places substantial emphasis on aspects that go beyond viewing the maqām as a scale, and it is clear that the opinion expressed in the majority of academic literature is that maqāmāt should not just be viewed simply as modes equivalent to Western scales, but rather as musical entities identified by a detailed procedural organisation exhibited in the course of performance.

Touma presents a more detailed description of this procedural organisation, the language of which is strongly influenced by his overwhelming philosophy:

In order to render a clear and comprehensible account of this phenomenon, it is above all necessary to avoid the definitions and technical terms used in

55 European music and to rid oneself of the opinions expressed by European writers on the subject (1971, p. 38).10

Summarising his analysis, each ‘maqām performance’ is described as having melodic passages, phases, tone levels, various melodic axes, central tones, and nuclei (Touma

1996, pp. 39–43).

The ‘melodic passage’ is described as a distinct musical statement separated by silence from other melodic passages, in which something new happens compared to previous melodic passages (Touma 1996, p. 39). Touma states that in each ‘melodic passage’ at least one ‘phase’ exists, which is music played ‘on one tone or within a certain tonal area’ (1996, p. 40). The tonal area is described as being established or constructed, of ‘tones that are organized around a melodic axis’ which results ‘when a central tone, which may be encircled by neighbouring tones, has been repeated at least three times’ (Touma 1996, p. 40). Touma concludes his detailed description by stating that the ‘supply of central tones’ form the nucleus of the maqām, and the ‘types of interval that constitute a nucleus determine the emotional content of the maqām’ (1996, p. 42).

This complex, involved and performance-centred description interprets a maqām as a system rather than as an individual mode: an overriding conceptual framework. Rather than listing a set of theoretical modal relationships, Touma is defining a performance structure. As such, his interpretation actually represents the maqāmāt as almost the whole continuum of definitions, except that he ironically does not refer to the maqām as a scale at all. He also does not refer to taqsīm during his description of ‘maqām performance’ (Touma 1996, p. 39). Whatever his reasons for doing so, it is clear that

10 Examples that may illustrate what Touma is referring to include Carmi-Cohen’s emphasis of ‘melody types’ (Carmi-Cohen 1964, p. 102), Ayari and McAdams use of the terms ‘musical motifs and melodic cells’ (Ayari and McAdams 2003, p. 161), Nettl and Riddle’s emphasis on ‘ornamentation’, ‘acceptable accidentals’, and ‘typical modulation patterns’ (Nettl and Riddle 1973, p. 14), and even Marcus’ contention that each maqām has ‘special melodic features’ (Marcus 1993a, p. 108).

56 Touma’s contention is that the maqāmāt only come to exist completely during performance. They cannot be abstracted to exist as scales because their complete characteristics only come to light in the course of performance. In this light, Touma’s interpretation and those that similarly represent the same space on the definitional continuum appear quite valid.

Interpretations such as Touma’s are based on the fact that this musical phenomenon cannot be easily portrayed in Western theoretical terms. Indeed as discussed above,

Nettl and Riddle support the broader continuum-based interpretation for maqām (1998, p. 372). Tending towards analytical segmentation that views theoretical components as being conceptually subservient to the act of performance, Westernised interpretations may well fail in coming to terms with the maqāmāt and taqāsīm. However, by not mentioning that a maqām can be represented as a scalar structure, Touma’s description contradicts current trends in Arabic theory (Marcus 1987, p. 39). By subsequently contextualising in a historical light all interpretations that define maqāmāt in broader terms, as well as those that view maqāmāt as scales, I will clarify that all these perspectives are important, and in doing so, I will derive a more conclusive understanding of exactly what maqāmāt are.

This definitional elusiveness can be further explained through an exposition of the transformations that have occurred in the theoretical representation of maqāmāt in

Arabic modal theory. Marcus’s dissertation (1989) and related articles (1987, 1992,

1993a, 2002a) thoroughly document the ‘periodization’ and evolution of Arabic music theory. A reflection upon these conceptual changes provides answers for why this definitional confusion exists between the literature and modern-day pedagogy and performance practice. Marcus suggests that ‘changes in the way theorists have conceptualised about the maqāmāt suggest that it is best to divide the modern era into

57 three periods: early, middle, and present day’ (1987, p. 35). From this historical framework, Marcus demonstrates exactly what changes have occurred, and also provides evidence of socio-cultural forces that have influenced these changes in Arabic music theory. His analysis provides an excellent chronology that explains why such definitional variety exists.

1840

Viewing the beginning of the modern era as beginning around the time when the Arabic scale was formally re-conceived into quarter-tone intervals, Marcus outlines the manner in which maqāmāt were described from this time onwards (1987, p. 36). He cites the methodology of Mashāqah, a theorist from the mid-nineteenth century (writing circa

1840), as a source for how maqāmāt were conceptualised in the early period. A maqām was described by giving a specific melodic line, naming each note using the Arabic and

Persian note names, from which ‘one could apparently ascertain the notes used and also the maqām’s characteristic melodic features’ (Marcus 1987, p. 36). This methodology was widespread in Turkey and Syria during this period, and in providing evidence for why this theory came to the fore, Marcus notes a strong Turkish influence throughout the Middle East in the musical culture of the early and middle periods, pointing out that

Mashāqah lived in Syria (1987, p. 44).

However, soon after this methodology was formalised in Mashāqah’s treatise

(1840) it fell out of use in the Arab world. Marcus asserts,

Mashāqah’s lines were no longer understood. That is to say, musicians and theorists no longer comprehended how to use these melodic lines as the basis for improvisation, how to see them as abstractions of a mode’s characteristic features (1987, p. 37).

58 Indeed, no procedural information seems to be contained in the descriptions given by

Mashāqah; the maqāmāt are represented as abstract melodic lines.

1930

Marcus’ analysis reveals that the next system that contributed to the evolution of maqām theory appeared in the 1930s, in the form of a method that integrated the concepts of scale and tetrachord11 to define maqāmāt. In this system, each maqām was described structurally by naming a scale, using the Arabic and Persian note names, and also by listing the various tetrachords of each maqām by note name. The characteristic melodic movement of each maqām would then be described exhaustively by outlining the order in which the various tetrachords should be performed (Marcus 1987, p. 37).

Marcus (1987, p. 38) draws on D’Erlanger, whom he asserts was strongly influenced by

Turkish musical culture given that D’Erlanger’s main informant was educated in Turkey

(1987, p. 44). D’Erlanger describes the melodic movement of a particular maqām, hijāz, in great detail (see Figure 2). To the reader, this description appears as a melodic formula that could only be realised in the course of performance. It is a complicated and detailed description, far more complex, as Marcus asserts, than anything conceptualised in the early period of Arabic theory (1987, p. 38). It is also at odds with modern-day representations (compare Figures 2 and 3).

Importantly, this style of conceptualisation pre-empts Touma’s in-depth description detailed earlier (Touma 1996, pp. 39–43). Even though D’Erlanger’s description does not mention the detailed components that Touma pins his interpretation on – melodic passages, phases, tone levels, various melodic axes, central tones, and nuclei (1996, pp.

39–43) – the substance of his explanation is certainly leaning towards a procedural

11 A tetrachord is a scalar succession of four consecutive notes.

59 formula that can only be realised in the course of performance. As such, Touma’s position on the definitional continuum certainly has historical precedent.

Figure 2: Maqām hijāz, described by D’Erlanger, notated by Marcus (1987, p. 38)

Figure 3: Maqām hijāz: modern-day representation as a scale

60 The maqām notated in Figure 3 has two forms, demonstrated in ascending and descending form above, as it is often employed in performance. This illustrates the modern-day tendency to notate maqāmāt by showing the component parts.12

Modern-day conceptualisation

In Marcus’ description of modern-day maqām theory (1989), he makes a vital point in the analysis of how modal theory has evolved into its current practice. Marcus establishes that in the early and middle periods, performances of maqāmāt did not start on the tonic note, but rather ‘the performance of many maqāmāt started in the middle of the scale or at the octave rather than at the tonic … the tonic would appear only at the end of the performance’ (1987, p. 40). In this explanation one can only assume that

Marcus is using the word ‘tonic’ to refer only to the lowest instance of the tonic note on a given instrument; he is not using the word to describe other instances of this note found in other octaves, and he is not implying that performer’s did not play other instances of the tonic note in the instrument’s range throughout performance (1987, p.

40). Putting this to one side, Marcus asserts that the functional relationship between maqām and its tonic (lowest form of tonic) was characterised by ‘qarār’, meaning

‘resting place’, carrying a ‘sense of finalis’ (1987, p. 40).

Today, the term qarār has now been replaced by asās, which means ‘fundamental note’ (Marcus 1987, p. 40). Marcus asserts that this is the result of Western influence, and as such, asās is a term that emphasises the lowest tonic note of the scale (1987, p.

41). He points out that this definitional shift has seen a parallel change in performance practice. He asserts that during the time of his research (the 1980s), he discovered that performers were being taught to start a performance around a maqām’s lowest tonic

12 This maqām is constructed of ajnās (sets): trichords (three consecutive notes) and tetrachords (four consecutive notes). Some maqāmāt also contain pentachords (five consecutive notes).

61 region (focussed around the asās), rather than elsewhere in the octave as in earlier periods when the term qarār was in common usage (Marcus 1987, p. 41).13

Continuing this line of argument, Marcus draws a clear causal link between the perceived superiority of Western classical music by many Arabs, and the subsequent adoption of Western musicological methods and concepts during the modern period, becoming solidified in current pedagogical practices (1987, p. 45). Following on from the trend originating with the adoption of Western concepts such as the fundamental nature of the tonic note, the process of Westernisation began to be solidified in the mid- nineteenth century, as illustrated by the gradual adoption of a form of Western solfège

(Marcus 1989, p. 124). It did take some time for the tradition to agree upon how the solfège note names would be modified to account for the quarter tone pitches, which

Arabic maqām would be chosen to represent the ‘natural’ scale in the solfège system, and which Arabic note would represent ‘do’. Eventually maqām bayyāti and the note rast (C) were chosen (Marcus 1989, pp. 125–130).

Accompanying this transformation, maqāmāt also began to be illustrated using

Western staff notation, doing away with descriptions using Arabic or Persian note names (Marcus 1989, p. 130) resulting in maqāmāt being notated and thought of as single octave scales in modern-day theory (1987, pp. 39, 41; 1989, p. 130).

Furthermore, unlike in preceding periods, a description of the characteristic melodic movement was no longer given (Marcus 1987, p. 39). Marcus notes a distinct irony in the fact that as Westerners came to study the maqām system, ‘wanting to focus on melodic movement’ as represented by early- and middle-period theory, Arab theorists

13 Indian music theory implies a main tonic, with the tonic above being a goal of performance development. It may be that in early- and middle-period performance practice Arabic musicians employed the opposite trajectory, where the lowest form of the tonic was their ultimate goal. As Marcus (1987) asserts, the octave-bound scale has changed this conceptualisation, given that performers now learn to start performances from the fundamental note, asās.

62 were ‘fascinated by the static elements of Western music theory’ and consequently drew on Western classical music to simplify Arab music theory in the modern period (1989, p. 45). Marcus suggests: ‘It was apparently felt [by Arab theorists] that discussing melodic movement would clutter an otherwise uniform and systematic music theory’

(1987, p. 45).

Current Arabic music theory pertaining to maqāmāt in detail

Following the sequence of events that have shaped Arabic music theory into what it is today, maqāmāt are now analysed theoretically as the product of an amalgamation between two or more consecutive sets of notes. These structural components are known in Arabic as jins (plural: ajnās) and accordingly a maqām is constructed by the addition of two or more (Ayari and McAdams 2003, p. 161; Marcus 1989, p. 308). Ajnās are sets of consecutive stepwise notes (trichords, tetrachords, and pentachords) that are identified by musicians by way of their association with various maqām, specifically in their role as root ajnās. For example, the hijāz tetrachord (see Figure 3) is known as such because it is the root tetrachord of maqām hijāz (Marcus 1989, p. 271). Marcus offers a thorough analysis of the principal ajnās in modern Arabic music theory, and identifies either 9 or 11, depending on a range of differing analytical perspectives.

However, by including a range of other rare maqāmāt in the analysis, Marcus suggests that there are actually somewhere between 17 and 30 distinct ajnās (1989, p. 271).

Marcus asserts that it was not until 1932 at the Arabic Music Conference in Cairo that the structural analysis of ajnās was formalised. It was apparent to conference members that ‘this is the first time that the Arab modes … have been analysed and broken down into genres’ (Marcus 1989, p. 278). Significantly, Marcus declares, ‘the

63 whole idea of presenting a structural analysis for each mode (absent in the early period) seems to be the result of Western influence’ (Marcus 1987, p. 44).

The separate ajnās that combine to form maqāmāt can form in three ways (Ayari and McAdams 2003, p. 163):

• Disjunct, where two ajnās connect by adjoining notes

• Conjunct, where two ajnās connect by a note in common

• Overlapping, where two ajnās share more than one note in common and join

across this shared pitch territory.

This is concurred with by all accounts that reflect modern Arabic theory, and similar descriptions of how maqāmāt are constructed can be found in Marcus (1989, p. 308),

Marcus (2000 pp. 37–38), as well as the most modern of sources, such as the extremely well-authored website Maqam World (2001–2005).

Theory and practice: what is reality?

I have established, through an analysis of Marcus’ detailed chronology (1987, 1989), that in modern-day theory maqāmāt are represented as modes built from the combination of two or more ajnās, notated in the same manner as Western scales, largely due to a Western influence. At the same time, recent sources investigating

Arabic music (Carmi-Cohen 1964; Marcus 1993; Touma 1971, 1996; Nettl and Riddle

1998; Racy 2000; Ayari and McAdams 2003) continue to reinforce that maqāmāt are structures more complex than scales, with inherent procedural rules of performance.

These attitudes echo the historical conceptualisations articulated by Marcus (1987,

1989) that emphasise the idea of an organised performance practice. Thus the tension that I identified at the beginning of this discussion is brought to the fore: why does this misalliance exist between the literature that emphasises maqāmāt as having melodic

64 parameters, and modern-day theory and pedagogy, which teaches maqāmāt as scale configurations?

The historical conceptualisations traced by Marcus’ chronology (1987, 1989) earmark notions of Arabic music theory that describe maqāmāt as a melody type or melodic formula. Remembering that these documents were reporting on an oral-aural tradition, it is perhaps best to view them as attempted representations of an evanescent performance practice, each presenting something of the total picture of what maqāmāt are. Correspondingly, modern works of literature that similarly emphasise that maqāmāt possess an inherent melodic protocol are also attempting to describe a performance practice, one that I argue has evolved from the historical sources identified by Marcus

(1987, 1989). These sources generally support the research I will present in Chapter

Five, which provides evidence of a melodic procedure encapsulated in the maqām concept that performers intuitively aim for. In asserting this, I must emphasise that the

Egyptian musicians I interviewed did not speak about their modal practice in these abstract theoretical terms. Most importantly, this musical practice was not learned in the explicit theoretical manner that readers interpreting literature which formalises these practices may construe.

Thus, accounting for all sources, I argue that maqāmāt must be defined in a way that incorporates the more complex details concerning a modal schema and performance plan, but any description must also account for the manner in which these other criteria that operate in practice are learned. Maqāmāt are not simply scales, but musicians most appropriately and successfully undertake any further description of their nature during performance.

As such, it is perhaps most appropriate to define maqāmāt in terms of those practices such as taqāsīm that feature them. That is, a maqām provides the pitches for

65 the improvised melodies of a taqsīm, an instrumental improvisation in which the melodic protocols of a maqām – absorbed performance practices rather than codified theoretical concepts – are brought to full realisation during the course of performance. I demonstrate in Chapter Five that this customary melodic practice is something that musicians internalise passively during their development. Indeed, Marcus suggests such a process in stating that, ‘For musicians, the individual modes are generally understood by the way they occur in practice, that is, in existing compositions and improvisations, new and old’ (2002a, p. 33). The full details of this process of transmission will be explored and detailed in Chapter Five.

In recognising that in practice a maqām is brought to life during a taqsīm performance, Marcus details an intricate modal framework that is demonstrated by compositions and improvisations. He details this ‘more complex and dynamic definition’ as

involving such considerations as the notes not duplicating at the octave, specific non-tempered intonation for some notes, a common progression or path for moving through the various regions of a mode’s scales, additional ajnās beyond those indicated in the simple scalar presentation of the mode, characteristic accidentals, and a set of standard modulations to other modes (Marcus 2002, p. 38).

2.3.3 Modulation

Modal modulation is recognised as ‘one of the most important aspects of Arabic modal practice’ (Marcus 1992, p. 173). Perhaps the most distinctive aspect of a taqsīm performance, this procedure allows performers to demonstrate their virtuosic modal fluency. Tracing its usage in Arabic music from the medieval period to the modern era,

Marcus asserts, ‘Some even argue that modulations represent the most important aesthetic moments in Arab music’s modal practices’ (1992, p. 173). Involving movement from one mode to another, the practice of modal modulation in a taqsīm

66 performance is an opportunity for the performer to draw the audience into their improvisation by engaging in expertly conceived preparation that eventually culminates in a satisfying modulation to another mode, potentially followed by further modulations to other related maqāmāt, before returning skilfully to musical material rooted in the original maqām’s modal context.

Specific relationships exist within maqāmāt families where the proximity of relatedness is determined by the similarity of the ajnās that combine to form maqāmāt.

Those maqāmāt that share common ajnās are considered closely related, and modulations of this kind are viewed as gradual due to the ease of movement between the two modes, as facilitated by the similarity between their ajnās (Marcus 1992, p.

178). Contrastingly, those that do not share such similarity are considered more distant, but the performer may modulate between them immediately to emphasise a sudden modal shift, juxtaposing their distinct tonal qualities (Marcus 1992, p. 178). If sudden modulations are not desired, less-related maqāmāt require carefully crafted steps using an intermediary maqām (Marcus 1992, p. 182). This facilitates discrete pitch modifications that in turn render ajnās more compatible and therefore suitable for modulation as determined by stylistic conventions. Modulations of this kind necessitate a greater level of expertise on part of the performer, and are undertaken in an extremely gradual manner. My research conducted with Dr Alfred Gamil suggests that extensive practice time is devoted to developing methods associated with making modulations of this kind, and a certain pride is taken when such modulations are made successfully.

Theoretical sources have varied over time with regard to articulated rules and conventions concerning maqām modulation. Whilst Marcus outlines sources from the nineteenth and early twentieth centuries that stipulate a necessity to modulate (Marcus

1992, p. 181), my research suggests that musicians such as Dr Alfred Gamil, Joseph

67 Tawadros, and Esawi Dagher do not feel they are rule-bound with regard to modulation; rather, they are only bound by what is pleasing to the ear. The details of these opinions are investigated in more detail in Chapter Five. Marcus puts forward an alternative contention. Whilst acknowledging some musicians, like my interview subjects, claim only to be guided by ‘good taste’ and no other rules, other musicians do articulate clear principles that govern modulation (Marcus 1992, p. 183). Summarised, these include a necessity to eventually return to the original maqām after modulations take place, and a knowledge of the relative proximity of maqāmāt as determined by the affinity of ajnās described in the preceding paragraph (Marcus 1992, p. 183). Other more specific rules concerning modulations between modes that belong to various incompatible pitch-based families, details that are beyond the scope of this thesis, are outlined in depth by Marcus

(1992, pp. 185–189).

2.4 Conclusions

The considerations outlined above concerning Arabic modal theory and the interaction between maqāmāt and taqāsīm go to the core of the modal framework that Egyptian musicians work during improvised taqāsīm performances. Considered in conjunction with the fundamental concepts related to improvisatory practice in Hindustani classical music and jazz, unique qualities have surfaced regarding the improvised performance context as practised in each tradition. However, whilst distinctive aspects of each tradition are clear to the reader, what is of greater interest is the similarities that may be gleaned by looking more deeply into practices that seem at first distinct.

Of prominence is the role of harmony in the jazz tradition, a musical construct non- existent in Hindustani music aside from the tambūrā accompaniment, and in Egyptian music beyond the use of accompanying drones or single-key ostinato bass lines. The

68 existence of the cyclical harmonic progression in jazz not only outlines form, but more importantly, directly impacts on the modal choices available to musicians when improvising. Whilst a perceivably limiting parameter, the principles governing a

Hindustani musician’s choices when developing a rāga during performance are also curbed, not by harmonic boundaries, but by the necessity to portray the identifying characteristics of a specific rāga accurately, whilst balancing personal originality with inherited tradition.

For Egyptian musicians, choices are seemingly free in comparison. However, a detailed modal schemata, whilst not known by musicians as a formalised published theory, is known by musicians by experience, and it directs not only their characterisation of maqām when improvising, but also governs the movement away from, and back to, the original maqām. Thus, whilst the subjects of this dissertation may at first seem to be three largely autonomous improvising traditions, what relates them are the genre-specific parameters that organise their improvised behaviour by shaping the interaction between fixed and flexible aspects of each tradition. The following three chapters present a detailed account of how musicians learn to improvise in each tradition, and as such represent an analysis of how these genre-specific parameters are incorporated into the musical behaviour of individual musicians.

69 Chapter Three Learning to Improvise in Hindustani Classical Music

In this chapter I will present a detailed explanation and analysis of what Hindustani musicians do in order to develop the ability to improvise. I examine both the learning environments in which they are trained and the methodologies that are employed within these enriching situations. In doing so, the nature of the improvisational process in

North Indian music is clarified. Whilst presenting historical practices, modern contexts will also be emphasised. I established in Chapter Two that vocal music provides the principle paradigm in North Indian classical music. As such, the aspects of pedagogy and transmission specifically derived from the practice of vocal teachers throughout this discussion can equally be applied to the instrumental context. I outline these similarities, whilst also outlining practices related to the pedagogy of instrumental improvisation towards the end of this chapter. It is crucial to assert at the outset that instrumental students most commonly begin their studies by singing, continuing this practice throughout their development, and much of the content of their lessons is derived from a vocal origin. Significantly, this content and other material is often vocalised before being played on an instrument.

Within the extensive collection of sources that investigate North Indian music, various writers have presented accounts of the tradition’s history, culture, and theoretical development. These include Meer (1980), Neuman (1980), and Ruckert

(2004). These sources do investigate pedagogical themes, though only in a general manner. This literature is supplemented by genre-specific studies such as those by Wade

(1984), and by Sanyal and Widdess (2003), which focus on the structure and development of particular genres in Hindustani music (khyāl and dhrupad respectively).

70 Given the rigorous nature and orientation of learning processes in the Hindustani music tradition, a number of other more specific studies have investigated the tradition’s pedagogy and modes of transmission. These include published works by Shankar

(1969), Stevens (1975), Shepherd (1976), Booth (1982, 1983, 1986, 1987, 1996), Alter

(1989, 1997), Scott (1997), Farrell (1997), and Schippers (2007). Khan and Ruckert’s work (1998) provides an insight into a teaching methodology in a modern context (at the Ali Akbar College of Music), whilst Kippen’s study (1988) provides an insight into the development of a particular school of tablā performance and pedagogy. Other studies have examined the nature of improvisation in the performance of North Indian music, such as research by Nettl (1974), Lipiczky (1985), McNeil (1993–1994), and

Napier (2004, 2006).

3.1 Primary learning environment: guru and shisyā parampara

One of the most prominent themes discussed throughout the literature on Hindustani music pedagogy is the traditional relationship between teacher and student, guru and shisyā.14 Shankar (1969), Stevens (1975), Shepherd (1976), Neuman (1980), Kippen

(1988), Scott (1997), Booth (1982, 1983, 1986), Meer (1980), Alter (2000), and Ruckert

(2004) are important sources that establish this relationship as the historical foundation of training in Hindustani music, many noting that it is a fading pedagogical structure.

As such, discussion of this relationship takes place in both present and past tenses.

Reflecting upon current teaching methods, Alter asserts, ‘Most musicians regard the idealised traditional teaching system, featuring the master-disciple relationship, as the basic model from which music educators develop their teaching methods’ (2000, p.

442). Whilst accounts of the changing status of this relationship indicate that it rarely 14 For musicians of Islamic background, the guru is known as ustad, and shisyā as shagird.

71 exists in its traditional form today, the relationship bears such philosophical significance to pedagogy in Hindustani music that it will be investigated in more detail before tempering the investigation with a portrayal of current pedagogical trends.

Before doing so, I will contextualise the guru-shisyā parampara (guru shisyā succession) in the greater context of Indian society. Shepherd suggests that this pedagogical relationship ‘is very similar to the guru kul (“teacher’s house”) system of education in ancient and medieval India’ (1976, p. 53). Slawek agrees, identifying the

‘gurukul household of the guru’ as the historical antecedent to the guru-shisyā parampara of North Indian music (2000b, p. 457). As an educational relationship, it is relevant not just to music and art, but extends from academia to spirituality, and even to sport. Describing the ancestral pedagogical model expressed in the gurukul, Shepherd writes that a boy’s education in reading, writing, and arithmetic started at home under his first teacher, before he progressed to the discipleship of a guru who accepted the boy into his home once he turned eight. At this time he was taught the Vedas15 and given formal religious training (1976, p. 53). Identifying an even wider sphere of influence,

Armstrong and Bates (2001, p. 194) go as far as suggesting that the influence of the guru-shisyā parampara in sport has extended beyond India, influencing the reverential pedagogical relationships in judo and karate, which they hypothesise, originated from the Indian martial art kalari payattu.

The guru-shisyā parampara, as the fundamental conceptual notion that defines traditional education in India, influences not only specific educational pursuits, but is an idea that imparts a philosophical authority onto Indian society. Reverence, mutual respect, and devotion are integral characteristics of the relationship between master and disciple. The filial piety that characterises the interpersonal mechanisms of the

15 ‘Divine Knowledge’: ancient Sanskrit texts that formed the basis of later Hindu scripture (Ruckert 2004, p. 95).

72 relationship require shisyā to submit to the authority of their guru (Neuman 1980, p.

46), a cultural behaviour that feeds and characterises the unquestioning imitation that is so important to Hindustani classical music pedagogy (Shankar 1969, pp. 11–14), which will be elaborated later in this chapter.

Returning to a purely musical focus, Shankar emphasises the overwhelming extent to which the guru historically guided a student’s musical training (1969 pp. 11–15) and

Ruckert reinforces this element:

The student’s attitude, conceptions, and technical prowess was both the product and the responsibility of the guru, and it was a relationship not to be taken lightly, since it involved a great commitment on both sides (2004a, p. 34).

As in the general historical antecedent, the commitment is embodied in the tradition of shisyā living with their guru (Shepherd 1976, p. 53) (Slawek 2000b, p 459) allowing disciples to be involved in every aspect of their guru’s life. Most importantly, the in situ aspect of their training allowed shisyā to benefit from being surrounded by music, overhearing lessons with other shisyā, while also witnessing their guru practising. Scott reinforces the extent to which this arrangement also allowed for the guru to supervise a disciple’s practice (1997 p. 463). Furthermore, Ruckert emphasises that without any other media from which to learn, a student’s relationship to Hindustani music was completely facilitated through the relationship with their guru (2004, p. 34). Indeed, this resulted in a shisyā’s musical environment being dominated by the guru’s specific style, especially prior to modern concert life and the development of recordable media.

Neuman writes concerning the love and devotion that exists between guru and shisyā, emphasising that the guru, if not already related by blood to the shisyā, becomes as a father to the disciple (1980, p. 45). Slawek notes this paternal closeness as an aspect that characterises his relationship with his guru Ravi Shankar (2000, p. 464).

Elaborating upon the closeness between guru and shisyā, Slawek emphasises that the

73 relationship is built on ‘an element of exchange’ (2000, p. 464) whereby the guru gives the shisyā a wealth of musical knowledge in exchange for devoted gratitude (2000, p.

464). Furthermore, describing the importance of the relationship from the disciple’s perspective, Neuman emphasises the extent to which a musician’s socio-musical identity is derived from their musical heritage – that is, with whom they have studied and to whom they are related (1980, pp. 43–44).

Extending the philosophical scope of the relationship, Scott, in his detailed analysis of Hindustani vocal pedagogy, describes the image of the guru as being based on two important metaphors in Indian culture: the guru is viewed as both ‘divinity’ and ‘father’

(1997, p. 2). Emphasising the dual relevance this relationship holds for the disciple – that is, spiritual and familial – Scott also refers to the guru being as a ‘god-parent’

(1997 p. 501). Kippen also reinforces the ‘spiritual power that has distinguished the … relationship, and which has made it so successful’, whilst also referencing earlier sources in the literature that speak of ‘direct communion’ between guru and shisyā

(1988, p. 107). Correspondingly, Shankar goes as far as to suggest that choosing a guru is more important than deciding upon a husband or wife (1969, p. 11), and is joined by

Shepherd (1976, pp. 55–58), Kippen (1980, p. 113), and Slawek (2000b, p. 459) in emphasising the importance of the corresponding ganda bandhan ceremony, in which once the disciple has proved to be acceptable to the guru, a red thread is tied around the shisyā’s wrist to symbolise the important bond that now exists between them and their guru (Shankar 1969, p. 12). It is within this fundamental relationship that traditional music pedagogy has taken place in Hindustani music: historically, it is the primary learning environment.

Within the literature’s discussion of this important relationship, Meer outlines a range of circumstances that may influence the choice of a teacher, such as geographic

74 location or familial connections (1980, p. 138), not to mention a guru’s willingness to accept a new shisyā. Furthermore, in a study of tablā playing and pedagogy, Shepherd asserts that musicians are drawn from all sectors of society: ‘from musician and non- musician families’ (1976, p. 51).

In more detail regarding children from music-practising families, Shepherd states that a child (normally a boy) is encouraged to play with instruments from an early age, as soon as he ‘can crawl’ (1976, p. 52). Commonly children express a preference to play a certain instrument at this early age, often that of their father or another family member, although sometimes they choose another instrument that has been experienced by attending concerts or interacting in other familial-musical spheres. Once they do make a choice, Shepherd states that they are then taught by a member in their ‘joint family household’ (1976, p. 52). Once the child is around eight years old, their family will choose a guru under whom the young disciple will continue their musical education

(Shepherd 1976, p. 52). For all shisyā, from musical families or otherwise, the necessity of a more focussed and all-encompassing musical environment eventually necessitates the relocation of the shisyā to their guru’s house.

From then on, and until he begins his professional life, he spends the best part of each day in his guru’s home – whether or not his guru is his father or other member of his own household (Shepherd 1976, p. 53).

Meer’s account agrees, noting that it is common for musicians who eventually enter into discipleships to be born into a family of musicians, and concurs that often early training is received from family members (1980, pp. 144–148) before shisyā enter into a discipleship with a family member or another prominent musician.

The hereditary component reinforced by the literature is unsurprising: those children born into a family populated by musicians benefit from near-constant exposure to music, and their transition to a musical career is a consequence of this environment.

75 However, not all musicians traverse an inherited path to discipleship. In this case,

Shepherd asserts that those not from musical families come to discipleship later in life and ‘often do not start studying classical music seriously until their teens or even much later in their lives’ (1976, p. 51).

Writing over 40 years ago, Shankar grieves that the traditional relationship described above is fading:

There was a time when classical musicians did not have to be so much concerned with the material side of life, for their artistry was widely proclaimed and the patronage of royalty and wealthy persons provided them will all the necessities of life. In turn, they were able to accept into their homes a number of disciples, so all their time could be devoted solely to music (1969, p. 14).

Shepherd concurs in more detail, stating in 1976: ‘Until two generations ago, a chela16 would live in his guru’s home along with the other chelas, but with economic and social changes this is not longer the case’ (1976, p. 53). Alter examines reasons for the gradual demise of the traditional pedagogical relationship between guru and shisyā in his thesis

(1989),17 via an examination of Indian socio-economic changes that have drastically altered the nature of how Indian musicians are able to live their lives. As predicted by

Shankar (1969, p. 13) and Shepherd (1976, p. 53), these changes have virtually prohibited the existence of the relationship in its traditional form.

Whilst various socio-economic changes have altered the ability of musicians to take on disciples and teach them on a regular daily basis, Alter describes the extent to which modifications to the traditional relationship have been attempted by government and private institutions such as schools and universities since the end of the nineteenth century (2000, p. 442). However, modifying a traditional system based principally on rote learning (Slawek 2000b, p. 464) to conform to classrooms, curriculums, published textbooks and the like is a complicated transition (Alter 2000, p. 442). Even so, Alter 16 Another term for shisyā that Shepherd states is in more common usage in 1976 (1976, p. 51). 17 For more detail, see Alter (1989), Chapter Six.

76 suggests that the adaptation of traditional teaching relationships to the institutional setting has become widespread (2000, p. 442), though as will be demonstrated, some adaptations are not considered successful.

3.1.1 Modern adaptations of the guru-shisyā parampara

Background

Of fundamental importance to these pedagogical adaptations are those persons Alter labels as ‘educational pathfinders’: those who were instrumental in founding music- teaching institutions and publishing various pedagogical materials (Alter 2000, p. 442).

Alter highlights various individuals who founded music colleges, such as Sourindro

Mohun Tagore of Bengal, Sayajirao III of Baroda, and Madhanrau Scindia of Gwalior

(2000, pp. 442–443), although he suggests the most significant contributions to the modernisation of Hindustani music were made by Pandit Vishnu Narayan Bhatkhande and Pandit Vishnu Digambar Paluskar. He asserts that the time at which they worked is often referred to as a ‘Renaissance’ of Indian music (2000, p. 444).

One of most important contributions documented by Alter is the establishment of music schools by Bhatkhande in Lucknow, Nagpur, and Bombay, inspired by nationalistic feelings within the new middle class and a desire to deliver music education to the masses in order to preserve Hindustani classical music (2000, p. 442).

Paluskar similarly established music colleges in Lahore and Bombay (Alter 2000, p.

442). Accompanying the establishment of these music institutions, various publications were authored elucidating the subjects of music and music theory. Bhatkhande’s method is identified as the most influential in the Marathi and Hindi languages,

Bhātkhande Sangīt-Sāstra: Hindustānī sangīta-paddhati (1910–32), and whilst Alter emphasises that debate abounds as to the validity of some of Bhatkhande’s theoretical

77 explanations, the publications are argued to have influenced countless musicians, students, and scholars (Alter 2000, p. 445). A significant aspect of Bhatkhande’s method was the adaptation of the traditional guru shisyā relationship to the institutional context. As Alter describes it, Bhatkhande’s music system ‘imitates the teaching style of the master-disciple tradition … the disciple sits at the master’s feet, and music is sung and explained’ (Alter 2000, p. 445).

Nevertheless, Alter asserts that while the desires of Bhatkhande, Paluskar and others were inspired by admirable intentions, the current state of institutionalised music education receives criticism (2000, p. 446). As early as 1980, Meer wrote disparagingly regarding institutionalised music education in Delhi, asserting that ‘the University does not really produce performing artists’ (1980, p. 147), whilst Ruckert in 2004 echoes

Meer’s sentiments, suggesting middle-class aspirations are a corrupting source of demand for institutional education: ‘The idea of music as a refinement for the young is pursued by many parents anxious to provide opportunities and polish their offspring, much like taking piano lessons’ (2004, p. 37). Banerjee stridently criticises the private music schools that offer this type of instruction, suggesting that ‘these music schools and institutions do not reflect strikingly high academic standards and values’ (1986, p.

20) and asserting that problems are caused by ‘the growing commercialization of music as a consumer commodity and hence its obvious impoverishment’ (1986, p. 20).

Banerjee does express more confidence in the ability of universities to provide more successful instruction (1986, p. 21), though they are yet to achieve ‘a scientific and viable methodology of teaching and instruction’ (1986, p. 35).

Whilst imperfections exist in the modernised variety of Hindustani music pedagogy, the following elaboration will detail the pedagogical processes now active in

Hindustani music, and in terms of relevance to this thesis, my investigation will

78 specifically pursue the issue of what musicians do when learning to improvise.

Furthermore, given that the guru shisyā system of transmission is maintained to be the most effective model on which to base teaching in the Hindustani tradition, I will examine it in more detail before analysing modern adaptations of this methodology.

Listening and imitating

From a pedagogical perspective, the importance of this relationship and the guru’s function is illustrated directly by the oral-aural imitative communication that students and teachers engage in. Meer strongly emphasises that all teaching methods in North

India are based on copying (1980, p. 139), and Shankar reinforces the extent to which the student relies solely on the guru’s guidance in the initial stages of learning, which he asserts lasts approximately five years (1969, p. 13). Scott also stresses this fact, noting that gurus have regularly prescribed extended periods of listening before beginning formal training, enabled much of the time by the tradition of guru and shisyā sharing a residence (1997, pp. 246, 281). Scott also points out other advantages afforded by the tradition of living with one’s guru, allowing for continued ‘learning by osmosis’ (1997, p. 379).

The process, listening and then imitating, is identified by all writers who discuss pedagogical processes in Hindustani music, though until recently it has not been documented in specific detail, perhaps because this process is assumed to take place in most cultures. Consequently, a detailed insight into the interaction between guru and shisyā in historical contexts is not available in the literature, other than as brief, general statements of fact (derived from memories or general histories). However, Booth (1982,

1983, 1986), Kippen (1988), Alter (1997), Farrell (1997), Scott (1997), Khan and

Ruckert (1998), Schippers (2007), and Stevens (1975) have documented the

79 pedagogical process as it occurs in modern-day contexts by examining a variety of teaching situations.

3.1.2 Detailed studies

Field research conducted by Alter provides a detailed analysis of the changing nature of the guru-shisyā relationship. Alter experienced and recorded a modified guru-shisyā teaching environment by observing the teaching of Pandit Amarnath, the foremost disciple of the highly respected Ustad Amir Khan (1989, p. 142). In this college situation, teaching was undertaken in classes of six to twelve, with individual instruction lasting around 30 minutes given in front of the whole group. A primary point of interest is that Alter’s research was undertaken in a geographically traditional setting in New Delhi, India, examining the teaching of a well-respected vocal teacher in the same city.

Even so, the educational setting observed by Alter (1989, 1997) was modernised, in a music and dance college, the Sri Ram Bharatiya Kala Kendra. As such, the guru- shisyā relationship’s sacred profundity, based on location in the guru’s home, as documented by writers such as Shankar (1969, p. 11), was not present in its traditional form. In the introduction to his thesis, Alter (1989) notes the paradox of trying to find traditional forms of transmission in this modernised setting. He also discusses the conflict that is often present between the pedagogy of these institutions and more traditional methods of instruction embodied in the guru-shisyā relationship as discussed above. Alter resolves this conundrum by stating that the teaching he observed is based on a traditional model employing ‘significant adaptive strategies in the institutional setting’ (Alter 1989, p. 2). The extent to which this teaching model represents traditional methods is hard, if not impossible to determine, given the scarcity of detailed

80 information that would allow for a comprehensive analytical comparison with the past.

Nevertheless, Alter contends that the teaching he observed was the closest to a traditional model that he was able to access, and it would seem from Scott’s research

(1997) discussed below that Alter’s contention is accurate.

Furthermore, Alter asserts that educational institutions such as that which he examined have made a significant contribution to the transmission of traditional music in the twentieth century (Alter 1989, p. 2), and in a subsequent article discussing this subject matter, he debates whether the reality of any pedagogical situation, past or present, ever matches the ideal relationship described in the literature (Alter 1997, p.

61). Banerjee concurs, and suggests that the disappointment that often accompanies discussions of the traditional relationship’s demise is misplaced because it ‘is largely based on an idealized image of the system, the preconditions to whose effective functioning were often not fulfilled in actuality’ (1986, p. 17). Banerjee argues that in reality,

this mode of instruction depends not so much upon an objective system as upon an individual, the guru or the ustad who, despite his possible personal idiosyncrasies, represents a musical system himself. There is hence no guarantee that this individual’s musical knowledge or his method of teaching would be always of the first order (1986, p. 17).

Yet Alter concludes that no matter the reality, it is an accepted fact that the most effective pedagogical process is in fact this ideal traditional relationship, however rare it may actually be (Alter 1997, p. 61). Scott also makes a similar point in the conclusion of his doctoral thesis, stating that even though the traditional pedagogical system has

‘exploded into a myriad of institutional forms and pedagogical formats … the guru- shisyā parampara remains the essential paradigm for music education’ (1997, p. 504).

Banerjee details this in practice, asserting, ‘there still exists in India the system of music instruction based in some respects on the traditional mode of the guru-shishya-

81 parampara’ (1986, p. 19). It seems that although it is acknowledged that it is near impossible for the relationship to exist today in a perfect form, the concept is maintained as the tradition’s underlying philosophical goal. In this manner, the mode of transmission that Alter found active in this environment was founded upon the circular procedure of teacher-student imitation, a direct adaptation of traditional pedagogical communication. More specific details derived from Alter’s research will be discussed subsequently.

3.2 Primary pedagogical methodologies

3.2.1 Imitation and vocalisation

Gregory Booth’s initial research in this area (1982, 1983) involved the observation of instrumental lessons given by teachers at the Ali Akbar College of Music in San Rafael,

California. After examining instrumental lessons given by Ustad Ali Akbar Khan

(sarod), Ustad Zakir Hussain (tablā), and Swapan Chaudhuri (tablā), Booth was able to derive analyses of their teaching philosophies and methodologies as he witnessed them.

Summarising his findings, imitation was found to be the predominant feature of all teaching methodologies (Booth, 1983, p. 4). In order to be able to interact imitatively with the teacher, Booth identified that students required, not surprisingly, a ‘high level of aural perception skills’ (1983, p. 4). Stevens (1975, p. 36), Scott (1997, p. 469), and

Alter (1989, p. 158) also emphasise this phenomenon. Furthermore, Booth asserts that the voice-based methodology used by teachers to facilitate the imitative process enabled students to develop superior aural skills: ‘the importance of vocal practice for all students to develop aural sensitivity to and fluency with rhythmic and melodic intervals and patterns’ (1983, p. 1).

82 This general point has been made throughout the literature particularly by Scott

(1997, p. 349) in his description of the time he spent studying at the college featured in

Booth’s research. In more detail, Booth discovered that whether teaching instruments or voice, all teachers made use of a ‘personalized, highly idiomatic musical language of which the student must develop a fluent command … acquired through constant exposure and practice’ (1983, p. 5). Booth notes that Ustad Zakir Hussain specifically emphasised that the use of a ‘highly sophisticated oral/musical language for both pitch and rhythm allowed students to internalize the patterns and sequences inherent in the music’ (1983, p. 5).

The theory that Booth presents is that students acquire performance fluency through an interactive process facilitated by vocalisation, internalisation (memorisation), development of aural perception, and imitation of performance in practice. Shankar presents a similar, albeit less formal summary of the osmotic learning process outlined by Booth:

The disciple spends many hours simply listening to his guru, and then he endeavours to fill up the frame of a raga with improvised passages born out of the compelling mood of the moment (1969, p. 57).

Booth concludes that instrumental teachers used vocalisation as a means of developing aural perception and sensitivity to the stylistic attributes of the tradition, and it is clear from his research that the teachers under observation explicitly articulated the importance of vocalisation as the most significant aspect of the learning process, even though they were teaching instruments. The use of vocalised syllables in Hindustani music pedagogy is formalised in syllabic labelling systems: sārgām18 (for pitch) and bol

(for rhythm/drum strokes).

18 Sārgām was introduced in Chapter Two.

83 Explaining that vocalisation is central to all forms of Indian music, Shankar emphasises the centrality of singing in Hindustani music pedagogy, for vocalists and instrumentalists alike:

All Indian classical music – raga sangeet – is based on vocal music, because the structural basis of our music is melody, which holds the prime place in our musical traditions … Every instrumental musician must undergo rigorous training of the voice, learning many fixed song compositions. This should give the artist a thorough insight into the ragas and make him more sensitive to music. The wind and bowed instruments are most closely associated with vocal music, for they imitate almost exactly the flow and expression of the human voice (Shankar 1969, p. 18).

Stevens agrees, asserting that the prevalence of vocalisation is due to the fact that ‘all

Indian music stems from, and to some extent, imitates, the human voice’ (1975, p. 37).

Stevens also suggests that singing is not intended to give students a broad musical education, as is the philosophy of multi-instrument education in Western traditions, but rather, as Shankar suggests, ‘the memorizing of songs is considered equal to the memorizing of instrumental compositions as a means of achieving … the ability to improvise within a raga’ (Stevens 1975, p. 37).

3.2.2 Imitation and sārgām

Specifically, the use of sārgām in the teaching process aids in the development of extraordinary aural perception, such that Hindustani musicians, instrumentalists and vocalists alike, often develop absolute pitch as an unintended and serendipitous by- product of the pedagogical methodology based upon sārgām and imitation. Booth asserts that unlike Western methods used to teach aural perception that ‘begin with visual stimuli [a musical score] which then must be translated by the student into a vocal, musical response’, the methods of vocalisation he observed in Hindustani contexts employ aural stimuli from the outset. As such students are forced ‘to use and develop their aural abilities’ (Booth 1983, p. 5). Booth identifies the pedagogical

84 sequence in which syllables are employed in response to aural stimuli as a more logical approach than the Western method (1983, p. 5). The primacy of the visual score in

Western music often leads to the neglect of aural perception, whereas in Hindustani music pedagogy the absence of visual notation reinforces excellent functional pitch recognition.

It is interesting to note that although Alter, Booth, and Stevens discuss the necessity for students to develop superior aural perception, few writers have focussed in any detail on components of the pedagogical process, such as sārgām, that contribute to the development of this aural perception. This is in part explained by Booth (1983, pp. 2–3,

5) who asserts that the overall methodology that he observed throughout his research was one in which musical components were not taught in isolation, but rather, presented holistically to each student who assimilated as much information as individually possible. As education continues over many years, students notice and incorporate into their own performance practice the more complex layers and attributes present within the music in varied and possibly idiosyncratic ways.

Furthermore, as emphasised by Stevens, the very nature of an oral tradition, continuously forcing a student to rely on their aural skills to perceive the information being transmitted in spoken or musical form, also contributes largely to the development of aural skills (1975, p. 37), such that a specific system such as sārgām should be contextualised within a totally aural environment. Thus, it is not surprising that various pedagogical components such as sārgām are not highlighted as separate entities, but rather, presented within an overarching system. Whilst sārgām is not singled out, its impact on those Hindustani students who are taught using this method is in no way lessened (Alter 1989, p. 166), and Western onlookers who admire its effectiveness in Hindustani pedagogy note its significance.

85 In accepting Booth’s research, it is important to note that it was conducted in a

Western environment where adaptations to the traditional oral teaching process were made in order to cater for Western students. For instance, Swapan Chaudhuri and Ali

Akbar Khan allowed notation to aid the memorisation process, and Swapan Chaudhuri did not insist on vocal recitation with non-Indian students (Booth 1983). Booth does not discuss the extent to which this change altered the process of transmission. Furthermore the fact that the teaching occurred within the environs of a music college in a Western country, where teaching was not undertaken as regularly as in historical circumstances, needs to be factored into an acceptance of Booth’s findings. Whilst the assessment can only be wholly useful in examining modernised educational settings, Andrew Alter’s research discussed forthwith suggests that the teaching documented by Booth is representative of an adapted traditional teaching style, albeit removed from its original geographic and socio-musicological context.

3.2.3 Improvisation and the role of prototypes

Giving abundant detail regarding what students actually imitate during lessons, Alter describes how the teacher he observed in situ, Pandit Amarnath, used three methods of instruction when teaching formally: model phrase instruction, demonstration and verbal instruction (1989, p. 158). They are not three separate stages of pedagogy, but are combined according to what musical material is being taught. In examining these methods of instruction, Alter reinforces Booth’s findings regarding the use of vocalisation in Hindustani music pedagogy, though in more specific detail. Witnessing

Amarnath’s teaching, he noticed that three types of vocalisation were used in the transmission of the musical material: (i) sārgām syllables (as discussed above), (ii) any

86 vowel sound (usually ‘a’), or (iii) text and text syllables (1997, p. 62). These correspond to actual styles of singing used in performance, as introduced in Chapter Two.

Alter asserts that the method of instruction and type of vocalisation employed correlate to a different pedagogical focus according to the type of musical material being transmitted, either: (i) learning a new rāga, (ii) learning a composition (bandish), or (iii) developing skills of improvisation (1997, p. 62). He stresses that this does not reflect a segmented pedagogical structure within the process of transmission, but is only a framework that is useful for discussion and analysis (1997, p. 62). These findings are well illustrated by a transcribed lesson given to three students, providing a valuable insight into this teacher’s methodology.

When introducing a new rāga, Alter found that verbal instruction was an important aspect of initial instruction, asserting that it is deemed to be more effective to verbally describe the particularly lowered, raised, or missing notes that are represented in the arohavaroh – the ascending and descending conjunct pair of scales associated with a particular rāga (1989, p. 161). Importantly, as simply singing the arohavaroh cannot properly represent a rāga, after the teacher describes the note content verbally and the students have briefly sung the arohavaroh, he then sings short model phrases that are derived from the arohavaroh. The students repeat these phrases antiphonally. Listening to the student’s immediate mimicked responses, the teacher continues once a satisfactory response is sung, offering repetitions or short verbal advice if a student experiences difficulty imitating his prototype.

At times lengthier explanations and advice were offered to the whole group who were observing the individual student being taught (Alter 1997, p. 62). A further detail that was observed was the process of singing longer model phrases and then subdividing them into smaller repeatable units, according to the student’s aural ability

87 (Alter 1997, p. 62). Alter emphasises that the process of singing model phrases antiphonally allows students to absorb and assimilate the nuances of a rāga more effectively than simply singing the arohavaroh (1989, p. 161). Scott recounts this exact process as it occurred in his own lessons with three teachers in various stages of their pedagogy (1997, pp. 408, 427–429). Alter also reinforces that in addition to teaching the rāga, students develop four areas of technique whilst engaging in the imitative osmotic pedagogical process, these being: voice quality (defined by vocal timbre, strength and range), breath control, intonation, and facility with sārgām (Alter 1989, p.

164).

Sārgām syllables were used by Amarnath and students to sing the short and longer prototype phrases, and whilst advanced students who have a competent command over a new rāga eventually move towards singing on a vowel (such as ‘a’), Alter points out that sārgām is never fully discarded as part of the learning process (1989, p. 61), and once it is mastered through practice at slow tempos, sārgām offers students a simple way to perceive the melodic outline of the teacher’s model phrases, helping to clarify difficult passages (1989, p. 166).

I have already highlighted Booth’s research (1983) that evidences the use of sārgām and vocalisation in an instrumental context. Similarly, Shankar (1969, p. 97–

152) and Khan and Ruckert (1998) also use sārgām to teach rāga concepts and related pedagogical exercises. Shankar’s text has been designed to introduce Western sitār students to Indian music, motivated by the declared unavailability of competent Indian teachers in the West (1969, p. 108). Khan and Ruckert present a snapshot of the mode of pedagogy at the Ali Akbar College of Music. All authors notate the arohavaroh, compositions, and exercises in sārgām using letters of the Roman alphabet. Their method is derived from a vocal origin and represents a vocalised practice that would

88 take place in instrumental lessons (Khan and Ruckert 1998, p. 8; Booth 1983, p. 1).

Khan and Ruckert explicitly advise that the notated material be sung as such (1998, p.

8), it being notated only to allow the vocalisation to occur in the absence of a teacher.

Alter observed that in addition to singing antiphonal model phrases, the teacher also demonstrated longer phrases not intended for repetition by students (1997, p. 62). He found that demonstrating in this way allowed the teacher to continue to transmit the details of subtle melodic features in a more complete phrase, such as emphasising important notes that should be highlighted, as well as the nature and melodic position of ornaments. This style of osmotic learning, where a student does nothing but listen to their teacher’s demonstration, is the key to the student acquiring proficiency in some of the less tangible but essential aspects of performance, extending to overarching performance practices such the slow, stepwise exposition of a rāga.

In reference to this, Napier emphasises ‘the difficulty in controlling both the paradigmatic element rāga and syntagmatic element (structure and teleology) of performance’ (2004, p. 170), an issue he highlights as being particularly challenging for musically trained foreigners learning Indian music. In order to resolve difficulties stemming from incompatible Westernised perspectives, Napier highlights Zuckerman’s experiences when learning from the esteemed Ali Akbar Khan, who Zuckerman asserts would continually remind students to let go of their ‘“western” analytical, intellectual attitudes towards learning, in favour of a more subtle state of listening and openness’

(Zuckerman 1996, n.p. in Napier 2004, p. 169). The reverential state of listening prescribed by Ali Akbar Khan is not only philosophically important as the centrepiece of the guru-shisyā parampara, but as Napier avows, it is valuable to assume this position as a melodic accompanist when learning how to control a performance’s trajectory (2004, p. 170).

89 3.2.4 Bandish: why are they taught?

Alter’s research emphasised that when learning a new rāga, a composition (bandish) was taught to students to establish and develop their understanding of a rāga. Khan and

Ruckert (1998) and Shankar (1969) also include this as part of their instrumental pedagogy. This pedagogical stage is also discussed by Meer (1980, p. 140), Neuman

(1980, p. 23), and Scott (1997, pp. 377, 406), and Scott indicates that some teachers introduce more than one bandish in order to broaden the expressive scope for students in future performances (1997, p. 474).

The importance of this pedagogical process is emphasised by Alter’s statement:

‘Each new bandish in a rāga further develops the student’s understanding of that rāga since each bandish contains elemental motivic structures which are characteristic of that rāga’ (1997, p. 67). Scott concurs, noting, ‘Virtually every teacher and student in this dissertation has alluded to the importance of the bandish as a blueprint for the rendering of a raga’ (1997, p. 377). Alter elaborates on this point, stating that:

The most important formulaic material for performance and instruction which this paper has identified is the bandish itself. Within the melody of a bandish lie important melodic formulas of the rag (1997, p. 81).

Similarly highlighting the importance of the bandish as a framework for improvisation,

Nazir Jairazbhoy, cited by Neuman, states ‘The composition serves as a spring-board for these extemporizations and a frame of reference to which musician’s periodically return’ (Neuman 1980, p. 23).

Alter suggests that a musician’s understanding of a rāga is continually enhanced as more compositions are learnt, noting that when revising a previously introduced rāga during lessons, Amarnath does so by teaching a new composition (1989, p. 167). Alter stresses that by this strategy ‘the student reviews a rag through the process of learning a new bandish in that rag, thereby expanding repertory and exploring the rag more

90 completely’ (1997, p. 81). That is, given that in the melody of each bandish ‘lie important melodic formulas of the rag’ (Alter 1997, p. 81), by learning multiple bandishes for each rāga ‘the full repertory of bandishes and not any single individual bandish acts as the catalogue for the various formulas which identify a rag’ (Alter 1997, p. 81). Stevens agrees:

By memorizing fixed compositions, the student learns many phrases that evoke the dynamic qualities appropriate to different ragas, along with certain key phrases that, so to speak, identify particular ragas (1975, p. 38).

Bandish contain melodic content as well as stylistic and expressive elements that are directly related to a specific rāga. Memorising such prototypes allow students to internalise these characteristic musical elements.

Alter asserts that the repetitive use of prototypical models – firstly in the form of the previously discussed model phrases, and then in the form of longer compositions – is a ‘critical aspect of learning Hindustani music’ (1997, p. 67). Scott also emphasises that ‘A composition is “good” as it conveys the “genetic imprint” of the raga in which it is composed’ (1997, p. 377). As such, in combination with the guru’s performance examples, it is considered to be the best model on which to form improvised sections of a performance (1997, p. 377). Summarising Alter’s findings, in the early stages of instruction Amarnath teaches a bandish to beginning students after they have gained familiarity with the new rāga, achieved by copying model phrases, witnessing teacher demonstrations, and receiving verbal descriptions. More experienced students may be taught a bandish as the initial step in learning a new rāga (Alter 1989, p. 167).

In the classes observed by Alter, the text was first dictated to students (who notated it without teacher assistance due to his often-expressed disapproval of this practice) and then the melody was taught through the use of repeated model phrases that were sometimes the length of one line of text and other times subdivided to make

91 transmission more successful (1989, p. 168). In comparison to previous model phrase instruction used when introducing a rāga, when learning a bandish, model phrases were repeated quite extensively because bandish are considered to be more fixed, and student accuracy was required (Alter 1989, p. 168).

The nature of bandish structure allows for cyclical repetition, which aids the process of transmission, whilst also allowing the teacher and student to sing in unison and antiphonally (Alter 1989, p. 170). Verbalised instructions dealing with subtleties of pitch and inflection were again used to make corrections in the student’s responses, and sārgām syllables were only used at this stage if particular trouble was experienced in apprehending pitch elements. Once a student is deemed ready, Amarnath introduced the rhythmic aspects of the bandish, a stage that was accompanied by tablā.

3.2.5 Improvisation pedagogy

Alter (1989, p. 175) reports that after competence had been gained singing model phrases and bandishes, improvisation skills were taught in a more focussed manner. He does point out, however, that improvisation skills are worked on throughout a student’s instruction, once again emphasising the holistic approach to pedagogy in Hindustani music. Booth (1983, p. 5) also emphasises the holistic nature of pedagogy, specifically regarding improvisation. Even so, assuming familiarity has been gained in the rāga being studied by learning the arohavaroh and copying the teacher’s model phrases, and once a student has achieved a satisfactory competence with the bandish material, the student is then ready to be introduced to a variety of pedagogical methods used to teach improvisation using materials derived from the bandish and arohavaroh.

92 3.2.6 Springboards: the mukhrā and model improvised phrases

In the lesson he observed, Alter discovered that improvisation skills were introduced and developed firstly by teaching students the structural importance of the mukhrā, a cadential figure derived from the bandish being performed, continually returned to during performance (1989, p. 176). That is, the starting point for learning to improvise was developing an understanding of the improvisation’s finishing point! This is not that surprising, though, given that an essential aspect of a successful performance is the repeated correct arrival at the sam19 after a period of improvisation. A performer not able to do this would not only be embarrassed, but would be regarded as deficient.

Each bandish contains a mukhrā ‘around which much of the improvisation grows’.

(Alter 1997, p. 81) Students need to develop competence in recognising where the mukhrā falls within the rhythmic structure so that they are eventually able to perform improvised lines that conclude with an accurately positioned mukhrā, before beginning new improvised phrases and repeating this cycle. In this way, the mukhrā cadential figure operates both as an anchor for improvised phrases, and as the pivot point of departure. Wade similarly refers to the mukhrā as a ‘springboard’ for improvisation

(1973, p. 443).

In more detail, Alter notes that teachers and students interact in ‘a pattern of model phrase antiphony alternating with unison singing on the mukrā’ (1989, p. 176) so that students may develop the ability to perform their own improvised phrases. In this pedagogical method, a teacher will sing an improvised model phrase, leading the student to then sing the mukhrā in unison at the correct phrase position, after which the teacher sings another improvised phrase. Reflecting performance structure, this cyclical process allows students to develop mukhrā placement competence whilst observing the

19 The first beat of the rhythmic cycle.

93 teacher’s improvised melodies. Students assimilate how improvised phrases are tailored to fit before the mukhrā falls, whilst also assimilating their melodic content aurally.

Alter (1989) notices a steady progression of ability, starting with the teacher conducting the student’s entry into the mukhrā, up to the point in time when they are able to anticipate it correctly without assistance. Students also eventually gain independence and begin to attempt to immediately copy the teacher’s improvised phrases.

This model phrase instruction, employed in a form that replicates performance, was witnessed by Alter as the method of instruction used to teach many kinds of improvisation (Alter 1989, p. 176). Interestingly, Amarnath never explicitly articulated which type of improvisation he was teaching at any given point in the instruction.

Rather, during formal instruction he taught the correct type of improvisation for each section of the hypothetical performance model being transmitted to the student (Alter

1989, p. 177). As such, and as emphasised by Stevens (1975, p. 39), these holistic processes provide students with an experiential as opposed to intellectual or theoretical understanding of where the various types of improvisation should be used during actual performance.

3.2.7 Tān

Alter noticed that at certain times Amarnath interrupted model phrase instruction so that a type of improvisation technique could be developed: exercises based on fast note patterns called tān (1989, p. 178), as introduced in Chapter Two. Scott also records this process being undertaken at this stage in the pedagogical processes that he observed

(1997, pp. 290, 294, 406).

Scott (1997, p. 440) and Alter (1989, p. 178) both confirm that these exercises, again taught imitatively through model phrases using sārgām syllables, exhibit

94 transferrable formulaic construction that allow them to be applied to different rāgas according to individual arohavaroh rules. Stevens (1975) describes a similar use of tān exercises in instrumental pedagogy, and Khan and Ruckert (1998) note the importance of tān in instrumental performance.

Indeed, Alter suggests that ‘The most evident process in which the student internalizes a language of formulas can be seen in tān exercises’ (1997, p. 81). Alter stresses that, unlike other forms of improvisation that are learnt by copying the teacher, tān exercises are designed to stimulate the student’s individual creativity (1989, p. 178).

He asserts that they represent a dual importance for students, being: (i) ‘useful in developing his or her technical and creative skills’, and (ii) ‘designed to enhance the student’s ability to formulate improvised phrases in a variety of contexts’ by learning ‘a series of formulas which may be recombined unconsciously or consciously during performance’ (1997, p. 68). Scott agrees, and emphasises that these tāns, like the bandishes discussed earlier, ‘sometimes carry important pieces of the “genetic”

(motivic) imprint of the raga’ (1997, p. 290).

Scott emphasises the additional importance of tān in developing vocal dexterity

(1997, p. 290) and Alter recognises that due to the rhythmic complexity involved in many tān exercises (which he analyses in detail), their practice develops a masterful control over rhythmic invention and creativity when improvising, a practice known as

‘layakari’ (discussed in Chapter Two as laykari) (1989, pp. 184–185). This rhythmic creativity is an aspect of instrumental performance, also formed by practicing tān

(Slawek 2000a, p. 200).

Advancing beyond practicing tān exercises to improvising tān phrases is a long- term pursuit. Scott suggests that whilst tān phrases are normally improvised in concert, such can be the potential complexity of the underlying metrical cycles (tāla) that the

95 phrases are sometimes pre-composed to fit securely (1997, p. 290). Alter makes the point that as students become more competent in their knowledge and performance of tān exercises (after a few years), they develop the ability to realise previously unnoticed patterns and underlying structures in their teacher’s improvised tān phrases. Once realised, this knowledge of phrase architecture, combined with aural dexterity developed through years of regular imitative aural transcription, allows them to ‘copy those patterns with considerable ease’ (1997, p. 78).

3.2.8 Systems of variation and recombination

The final sections of both Alter’s and Scott’s research makes their work the most useful in the literature concerning improvisation pedagogy in Hindustani music. By investigating tān phrases in more detail, they throw much light on the sometimes over- emphasised mystery surrounding the process through which students eventually develop the ability to improvise. As Alter explains it, in addition to being formulaically transferrable into different rāga contexts according to arohavaroh rules (1989, p. 178;

Scott 1997, p. 440), the sets of tān exercises which are taught to students all

‘demonstrate a common system for creating recombinative sequence patterns’ (1997, p.

78). Similarly highlighting this elemental characteristic of Indian improvisation,

Lipiczky emphasises that in addition to tāns, ‘There are several types of formulaic patterns … that can be applied to create extemporized passages’ (1985, p. 158).

Beginning with the most simple of these tān exercise sets, Alter (1989, pp. 179–

186) undertakes a demonstration of the inherent recombinative system, eliciting a more detailed understanding of this procedure. The transcribed tān exercise sets feature ascending and descending scale patterns built upon an original melodic cell. These sets of exercises can be varied by: (i) expanding the scale beyond octave limits, (ii)

96 reordering the line structure of the exercises, and (iii) adding extra variant lines.

Without excerpting Alter’s demonstration here, 20 the process of variation and recombination is practised initially in basic forms, such as gradually growing ascending and descending scale patterns, whilst the exercises prescribed to advanced students feature such complex variation techniques that the structure behind the process of variation is initially obscured, appearing quite random without Alter’s subsequent analysis.

Alter emphasises that as students learn and practice these tān exercises, they are learning a process of variation through recombination that allows for improvisation to be undertaken. Through the exhaustive practice of recombinative methods as demonstrated by the teacher, the student ‘enhances his or her ability to combine short phrases from various systematic variation patterns’ (Alter 1997, p. 80). Alter stresses that students are not learning a vocabulary of phrases per se, but a whole system of variation that involves the recombination of small, internalised phrases according to a learnt method of variation such that

in recombined form the new phrases may not resemble any of the systematic patterns from which they evolved and eventually, the student may become unaware from which pattern a small phrase has emerged (1989, p. 186).

Scott contributes a similar conclusion in the form of his teacher Mohan Singh’s feelings. ‘After one or two years of practicing a raga he never knew which tans were his own or which were his ustad’s’ (1997, p. 474).

Alter emphasises that whilst tān phrases are internalised through extensive practice, what is all important is the conceptual process of variation that can later be applied in any rāga context (1997, p. 78). Scott’s emphasis is similar, pointing out the valuable prototypical role that these exercises play in developing the student’s improvisational

20 See Appendix 1: Excerpt from Alter (1989, pp. 179–186).

97 conceptualisation (1997, p. 302, 360). Both writers confirm Stevens’ earlier research concerning improvisational processes:

The ideas for variation come from previously memorized tans. By repeating the same process of variation with many other simple patterns, he eventually arrives at an experiential (as opposed to intellectual) understanding (1975, p. 39).

Such experiential understanding enables intuitive improvised performance by enabling procedural automation.

Writing regarding the predominance of formulaic structures in North Indian music, of which tān exercises are representative, Lipiczky also concurs:

Internalizing these formulas and applying them to the repertory of ragas and talas form a large part of the Indian musician’s practice sessions. In performance, the musician, like a chess player, responds to ever-changing musical situations by spontaneously adapting and inserting these formulas, transforming the event into a fascinating and unrepeatable display of skill and taste (1985, p. 158).

Moreover, Alter makes a most, if not the most valuable point (1989, pp. 186–187).

Although learning this process is vital in facilitating improvisational ability, the nature of musical creativity is such that simply teaching an exercise does not allow students to develop the requisite subtle understanding. Inferring that a different pedagogical approach is needed, the subsequent pedagogical methods exposed by Alter’s research rest on how the teacher, whilst still engaging in model phrase instruction extended to intricate tān phrase improvisations, sings phrases of such length and complexity that the student, unable to replicate the example, is forced to rely on their own creativity and intuition to produce a paraphrased replica (1989, pp. 186–187). As Alter discovered,

the majority of the teacher’s model phrases stretch the student’s aural skills beyond their limit so that the student must rely on her own interpretative skills which have been built over years of listening to her teacher and receiving instruction from him (1989, p. 186).

98 By forcing advanced students to engage in a process of creation rather than replication, whilst still demanding adherence to the underlying conceptual structures and methodology that have been assimilated over many years of osmotic teacher-student interaction, this process allows students to eventually develop an improvisational ability that balances the demands of the tradition with individual creativity. Napier eloquently describes this process as it occurs in North Indian classical music as ‘a dialogue between the demands of reproduction, distinctiveness, intergenerational continuity and contemporary subjectivity’ (2006, p. 12). He concludes that: ‘A culturally appropriate understanding of improvisation in Indian classical music must include the concept of subtle variation’ (Napier 2006, p. 12). Such subtle variation is what the process as described by Alter and Scott represents.

Scott’s doctoral thesis (1997) already referred to at length above will now be examined individually. Scott presents a detailed and fascinating insight into a variety of pedagogical processes in the transmission of Hindustani vocal practice. Much of his research confirms Alter’s findings, and to avoid repetition similarities have already been indicated throughout the discussion of Alter’s research. What follows is a description of how Scott’s research elaborates upon various data collected by Alter, or where his data provides a unique outlook on this investigation of improvisation pedagogy.

Scott’s research was undertaken with teachers with whom he had studied in the past, or with whom he was studying at the time of his research between 1989 and 1997.

These teachers were Krishna Chandra Bandopadhyaya, Sushil Kumar Mukherjee, and

Mohan Singh Khangura. Unfortunately Krishna Chandra Bandopadhyaya passed away at the outset of Scott’s research, though by changing tack, Scott was able to derive pedagogical information from students of the late teacher, as well as his own recollections of past lessons. The strength of Scott’s research is threefold: the majority

99 of it was conducted over a relatively long period of time (eight years); it was conducted in two distinct contexts, North India and the USA; and the featured investigations into three pedagogical approaches provide ample detail. The way in which the combination of these attributes benefits Scott’s research is that, due to the comparison he is able to draw between the methodologies of three teachers, unique details are brought into focus.

3.2.9 Tihāī phrases

A third improvisational technique, and an area that is focused upon in Scott’s research, are tihāī phrases, as introduced in Chapter Two. Lipiczky suggests that the tihāī are the most elaborate of formulaic phrases learnt by Indian musicians (1985, p. 159). Slawek emphasises their importance to instrumental improvisation (2000, p. 198–202). As such, the prime aspect of performance application that must be mastered involves being able to calculate when to begin playing the tihāī so that by its third repetition, it coincides accurately with the beginning of the rhythmic cycle, or in some cases leads fluently into the mukhrā.

Drawing a similar thread to previous research, Scott concurs with Lipiczky that these phrases are an important structural devices which, like tān phrases, mukhrā, and bandish, provide a model on which to develop individual improvised phrases (1997, p.

289). Describing the familiar process of memorisation, Scott’s research extends to a description of practising tihāī phrases to the extent that they were second nature, such that he was enabled to ‘perform them verbatim or create new ones based on their underlying forms’ (Scott 1997, p. 440). Lipiczky agrees, suggesting that after arduous practice, the calculation necessary in performance of tihāī phrases ‘becomes second nature to trained musicians’ (1985, p. 150). This discussion of learning and using tihāī

100 phrases in performance extends the previous discussion concerning the function of internalised prototypes in Hindustani music pedagogy, and emphasises the development of procedural automation, facilitated by internalisation through repetition. Importantly, it also points to an assimilation of structural rules and overarching formulaic concepts, rather than just note-for-note renditions of learnt phrases.

Underlying and embedded within the extensive descriptions that Scott (1997) provides of his lessons is the assumption that the nature of instruction in Hindustani pedagogy is not segmented. Rather, it is a flowing process that mirrors the format and cyclical nature of performance practices in Hindustani music: all performance components are present and taught within the ebb and flow of the lesson. Furthermore, there is an absence of explicit discussion regarding ‘how to play’, particularly compared to the metaphorical descriptions of ‘how to play’ that occur in the Western classical tradition. This is replaced in the Hindustani context by a demonstration of ‘what to play’ followed by imitation of exemplar performances by the teacher. Banerjee coherently assesses this phenomenon:

It is not often that attempts at producing meaningful intellectual-analytical formulations to aid the student’s conception of Indian classical music and enhance his artistic sensitivity are considered necessary, to arrive at a comprehensive grasp and understanding of the rāga as a coherent musical idea. For the student, then, the knowing here is not even in the doing, but mostly in the copying (1986, p. 24).

3.2.10 Concentrating effortlessly

Extending the discussion of the manner in which pedagogy format reflects performance structure, Stevens’ research in this area, concurring with much of the previously presented investigations, emphasises that

Throughout the training of Indian musicians, one basic principle emerges: all instruction is organized so as to put the maximum pressure on the student’s attention, that basic faculty without which any sustained effort to perform music is impossible (1975, p. 39).

101 There is logic to this claim when it is viewed in the context of the extremely complex polyrhythmic layers that Stevens exposes in his analysis of some pedagogical processes specifically related to sarod performance. Through example, Stevens (1975, p. 36) demonstrates a three-tiered rhythmic process whereby students play a tān exercise containing patterns that are built from, for example, four notes. These are played using a repeated stroke pattern based on, for instance, three, four, or five. These two layers are played over a rhythmic cycle counted by foot taps at equidistant points of the cycle.

When played together this produces a poly-rhythmic texture, the difficultly of which

‘lies in maintaining the three separate patterns (pitch group, stroke, and foot tap) simultaneously’ (Stevens 1975, p. 36). Stevens’ assertion is that a musical tradition featuring such multi-layered complexity necessitates a total focus on the development of concentration skills.

Although Stevens’ theory is well formed, in the context of research exposed earlier in this chapter by Alter (1989, 1997) and Booth (1983), it seems improbable that

Hindustani pedagogues would delineate components of teaching so as to teach students to concentrate above other aspects of performance. Whilst shisyā would certainly develop the ability to concentrate effortlessly by mastering exercises like those singled out by Stevens, gradual holistic development is emphasised in Hindustani music pedagogy.

The analysis presented above concerning the prototypical importance of bandish, mukhrā, tān, and tihāī have been derived from research largely, though not solely, conducted into the pedagogy of vocal music, the dominant paradigm in Hindustani classical music. I have also emphasised that instrumental pedagogy, vocalised as it is

(Booth, 1983), is taught in a manner that is similar to vocal models, sharing many forms and associated methods of transmission, such as mukhrā, tān, tihāī, and laykari (Stevens

102 1975, Booth 1983, Slawek 2000a). I will now examine a range of methodologies that are specific to the improvisation pedagogy of various instruments.

Turning the focus towards the North Indian percussionist, I will examine the form known as qaida (Kaida, Kayda). This is conceptualised in various ways. Courtney

(1995, p. 35) determines its meaning as ‘rule’, expanding to suggest that ‘it implies an organized system of rules or formulae used to generate theme and variations’. In practice, qaida are a musical form learnt by tablā players, ‘important for both the performance and pedagogy of tabla solos’ (Courtney 1995, p. 35). These themes and variations are taught to students using the percussion mnemonic equivalent of sārgām syllables, bol. Like other prototypical forms in Hindustani classical music, qaidas are memorised by students after extensively imitating their teacher who may vocalise the drum strokes and demonstrate on the tablā (Booth 1996, p. 20). Pedagogically, qaidas have a dual purpose. They develop a student’s technique whilst simultaneously serving as ‘stereotypical models of the creative process’ as it pertains to improvisation (Booth

1996, p. 21). Booth elaborates:

As more and more qaidas are learned, students begin to perceive patterns in the ways that their teacher creates and manipulates musical material. Learning orally, through imitation and memorisation, the tabla students internalize these structural patterns which then become avenues of approach to the creative process. For students, these templates are paths by and through which they may exercise their own creativity (Booth 1996, p. 21).

This description of percussion pedagogy is reminiscent of the exposition I presented earlier in this chapter concerning the manner in which vocal and instrumental students are taught to improvise tān phrases and tihāī. All involve the simultaneous internalisation of traditional forms and associated methods of variation, by imitation and memorisation, which once absorbed empower students to develop their own creatively generated phrases in the course of performance.

103 Ravi Shankar’s manual for the sitār (1969), already mentioned earlier in this chapter, provides an insight into some other specific pedagogical techniques related to developing the ability to improvise on the sitār. The manual includes a series of exercises derived from Shankar’s own education in Hindustani music, although he reinforces that they only exist as an introduction as lessons with a teacher must be commenced (1969, p. 97). Even so, they provide a valuable insight into some of the pedagogical methodologies employed by Indian instrumentalists. Prescribing a cyclical process of practice until exercises are memorised and mastered, the manual begins with sapat and muchhana exercises, the most basic and important form of training for beginners (1969, p. 108). Like the introductory vocal exercises discussed throughout this chapter, they are based on the arohavaroh of a given rāga. Shankar also recommends a series of repetitive exercises to develop mizrab (plectrum) technique, that like the tān exercises previously explicated, are based on using ‘combination figures’ (1969, p. 113). Similarly, alankar and paltas exercises are suggested to develop more advanced technique, practised in a variety of rāga contexts (1969, p. 115), as are exercises for the improvement of laya (1969, p. 130) (laykari) and ornamentation (1969, p. 132). These elements of instrumental pedagogy also feature in Khan and Ruckert’s exposition of pedagogical techniques as taught at the Ali Akbar College of Music

(1998).

Shankar’s manual, like Khan and Ruckert’s, provides further evidence for the derivative approach that instrumental pedagogy takes from vocal music in North India.

Given the dominance of vocal music, as historical antecedent, this is logical, and what

Shankar (1969) and Khan and Ruckert (1998) reinforce is the significance of abstracted exercises that simulate the challenges experienced in the course of improvised performance (Khan and Ruckert 1998, p. 205). These exercises facilitate the imitative

104 teacher-student cyclical exchanges that have been accentuated as a fundamental aspect of North Indian music pedagogy throughout this chapter.

3.3 Riaz

The greater part of this chapter has examined the development of improvisational ability in Hindustani music from the perspective of what is taught to students by their gurus.

The analysis of this didactic process has been accompanied by the notion that students need to constantly reinforce and memorise the material that is taught to them. This process of improvement through individual effort is known in Hindustani music as riaz, and its importance is a common theme in much of the literature concerning Hindustani music pedagogy. As Alter concludes, ‘There is little doubt that becoming a musician in

India takes many years of concentrated hard work. All teachers stress the need for constant riaz and a dedicated approach towards study’ (1989, p. 188). This premise is deeply embedded in previously discussed research by Stevens (1975), Booth (1982,

1983, 1986, 1987, 1996), Alter (1989, 1997), and Scott (1997), who all emphasise that the long-term process of learning Hindustani music must be accompanied by long hours of repetitive riaz. Furthermore, Neuman (1980) presents it as his main focus whilst addressing pedagogical themes, and the majority of his chapter dedicated to ‘Becoming a Musician’ discusses this aspect extensively. Meer (1980, p. 140), Kippen (1988, p.

117), and Ranade (1999, pp. 15–20) also reinforce the central importance of riaz.

Speaking of dedication and discipline, Neuman affirms that riaz is of fundamental importance to the development of skill in Hindustani music (Neuman 1980, p. 32). Scott similarly retells stories of extensive practice undertaken by some of the tradition’s most important musicians (1997 pp. 173–183). The accuracy of these, and similar stories recounted in other research, may be of secondary importance to the underlying purpose,

105 which as Neuman says, is a moral teaching to be entrenched in the thinking of students

(1980, p. 31). Furthermore, Scott draws out a comparison between practice and spiritual devotion (1997, p. 182), also a prevalent opinion in the literature. He points out that the focus and concentration reserved for spiritual practices such as yoga are equally relevant to a discussion of riaz. Shankar also emphasises that devoted practise goes beyond musical improvement, asserting that the ability to practise has a spiritual importance founded in the concept of sadhana (1968, p. 22). Moreover, Neuman (1980, p. 41) and Scott (1997, p. 465) note that musicians frequently undertake a chilla: a vow to undertake 40 days of continuous riaz, restricting their intake of food and sleep during this period of time. Whilst practise that is arduous is viewed as a crucial aspect of

Hindustani music pedagogy, Alter emphasises that

becoming a musician means more than simply devoting oneself to endless hours of practice. It means becoming part of a musical tradition and contributing to the growth, development and continuation of that tradition (1989, p. 188).

3.4 The conceptual aims of pedagogy

The conclusions of Alter’s study rest on the fact that the pedagogical process in

Hindustani music is a reflection of both the performance model and the nature of the oral tradition. He points out that it is the very lack of notation that necessitates both the use of sārgām as well as exhaustive repetition during lessons and practice. The most significant pedagogical concerns are firstly, the importance of learning rules of performance through the assimilation of performance prototypes taught in the form of compositions, model phrases, and formulas contained within tān exercises; and secondly, the dominance of additive phrases in the construction of the teacher’s improvised phrases. Furthermore, Alter stresses that although formulaic processes have been identified, these are subsumed by musical creativity during performance, such that

106 they bear distant resemblance to those phrases practised. Perhaps the most fundamental conclusion drawn by Alter is that students ‘are encouraged to re-create conceptual ideas rather than repeat note for note the phrase Amarnath has sung’ (1989, p. 194). It is the imbibing of prototype concepts that allows for improvisation skills to be developed.

3.5 Conclusions

Underpinned by an ancient pedagogical philosophy embodied in the guru-shisyā parampara, the mode of transmission in Hindustani Classical music is fundamentally based on imitation. Musicians develop the ability to perform a rāga within a non- segmented methodology that addresses the constituent elements of traditional performance conventions holistically. This lengthy and arduous process begins by listening to one’s guru. From the outset this respectful act trains shisyā to develop their most valuable asset: the ability to listen intensely. When appropriate, vocal explanations and demonstrations of prototype musical material are performed for shisyā to imitate, corrected if need be by the guru, before further instruction is undertaken continuing this cycle.

The imitated prototypical musical detail is embodied in compositions and other exemplars such as tān, tihāī phrases, and mukhrā, which are transmitted to the shisyā aurally. Inherent to internalising these phrases is the acquisition of a formulaic system of recombination that allows musicians to eventually improvise novel phrases in the course of performing a rāga. Through this performance-mirroring methodology carried out over many years, gurus tailor the imitative performances of their shisyās to meet with genre-specific stylistic criteria. Essential to this pedagogy is the use of vocalised syllables, both sārgām and bol, whose existence significantly mirrors the absence of written notation. As such, these systems deliver an unparalleled level of aural acuity,

107 which over time allows shisyā to imitate prototypes of intense difficulty. By engaging in this long-term imitative process, reinforced through extensive individual practice, these archetypes become the internal reference from which novel variations are created in the moment of performance.

A self-reinforcing process that produces an outcome perfectly suited to improvisation enables this: by working towards imitating the aurally absorbed musical material, an individual develops an ability to eventually reproduce, within reason, anything that is heard. Concurrently, by imitating this prototypical material and associated formulaic methods of phrase construction, shisyā internalise the detailed stylistic parameters of Hindustani Classical music. Thus, during improvised performances, enabled by their ability to reproduce with ease, musicians spontaneously produce music that conforms to traditional musical parameters by re-presenting the internalised prototypes. In a given moment, inherent in this process of re-presentation is a degree of novel variation that distinguishes performances as original, and in as much as this portrays the interplay between the fixed and flexible features of Hindustani

Classical music, it also describes the improvisatory process therein.

108 Chapter Four Learning to Improvise in the Jazz Tradition

In this chapter I will present a detailed account of what jazz musicians have traditionally done in order to develop the ability to improvise, whilst also contextualising this learning process within modern educational settings. Throughout this exposé, I will also investigate the nature of the improvisational process in as much as it is defined by the manner in which jazz musicians develop the ability to improvise. The jazz tradition, although young when compared to other musical traditions, has benefited from extensive research including historical accounts outlining the tradition’s stylistic developments, sociological and anthropological studies, and personal and musical biographies that illustrate individual contributions to the tradition. Additionally, transcriptions of improvisations (in varying levels of accuracy) have proliferated. Texts that analyse the musical style of individual artists have existed since 1927 (Berliner

1994, p. 97) and have flooded the marketplace over the last 60 years.

Even though a vast quantity of information is available, only a sparse collection of sources exist that examine in detail what jazz musicians do to develop their improvisational ability by deriving information directly from musicians. In stating this,

I emphasise that I have generally excluded pedagogical literature and texts from this evaluation for the reason that, whilst prescribing what to do and thereby implicitly detailing an opinion concerning what musicians actually do, these sources are often derived from a single subjective opinion rather than a collected retrospective analysis. I assert that the latter is the most effective means of analysis, as multiple opinions clarify common practice. In other literature, occasional references and anecdotes concerning development practices are common, but they often lack detail. However, Berliner’s

109 study (1994) excels in this field of research with regard to its detail and scope. Other sources, such as Owens (2002) Prouty (2004), and Sudnow (2001), supplement

Berliner’s exhaustive study.

4.1 Childhood in the jazz tradition

Introducing the notion, albeit informally, that a child’s knowledge of their environment is absorbed aurally before birth, Berliner’s first chapter outlines an oral-aural process at work in the jazz tradition. Stimulated by a variety of musical experiences, the musicians he interviewed began their education in the jazz tradition ‘surrounded’ by music, undertaking a musical education that one respondent described as ‘environmental’

(1994, p. 22). Berliner argues that children were traditionally provided with performance models, early training, and initial performance opportunities in places such as the home, church community, school, and the greater social network (1994, pp. 22–

24). In his analysis, the church not only figures as an important encourager of young musical talent in these ways, but is also highlighted as giving practical assistance to young musicians by providing access to instruments (1994, p. 25).

From within this early learning environment, young musicians expand upon their skills by undertaking a variety of activities, described in detail by Berliner. These may be summarised as: listening habitually to recordings around the home, thereby starting to develop knowledge of jazz repertoire and performers; developing preliminary instrumental technique by taking lessons on various instruments; learning performance formats and techniques by experiencing live performances in a variety of contexts; developing an understanding of jazz style and musical detail by mimicking adult musicians; and putting all these components into practice by undertaking amateur performances (1994, pp. 22–31). Those musicians interviewed by Berliner highlighted

110 these activities as important developmental factors, and their views are supported in detail by any number of biographies and autobiographies.

The data drawn from Berliner is detailed, and it is drawn from an unusually large number of informants. It represents a portrayal of memories from childhood and early teenage years. Moving from one memory and musician to the next, the ultimate conclusion of his first chapter is that jazz performance, and more specifically improvisation, is enabled by frequent aural exposure and practice during childhood.

This is an unsurprising finding. Berliner presents the information in a way that reflects the environment experienced by his respondents, and he does well to find and demonstrate some structure in what is largely an amorphous learning process.

Throughout his research he found that children absorbed from their environment whatever musical details they could, and primarily that which surrounded them determined the substance of their learning. He hesitates to organise the data into a more rigid theoretical architecture, but rather leaves a valuable impression of the organic musical saturation experienced by those musicians who participated in his study.

4.2 Informal networks and jam sessions

Berliner’s study discovered the existence of an informal music-education network that assists young musicians ‘to acquire the specialized knowledge upon which advanced jazz performance depends’ as they end their time at high school, or sometimes before leaving school (Berliner 1994, p. 35). Unofficial apprenticeships with older musicians

(Berliner 1994, p. 37), a peer information exchange system structured through practice sessions known as ‘woodshedding’, and performance activity undertaken in ‘jam sessions’, provide the oral-aural framework through which knowledge is disseminated.

Berliner emphasises that

111 For almost a century, the jazz community has functioned as a large educational system for producing, preserving, and transmitting musical knowledge, preparing students for the artistic demands of a jazz career through its particularized methods and forums (1994, p. 37).

At this point in Berliner’s study some socio-anthropological issues are raised that provide an important insight into the traditional learning environment of the jazz tradition: competitiveness as part of the education system, and group interaction as part of the learning process (which reflects the performance model). This has also been emphasised by Monson (1996). ‘Cutting’, the process of ‘blowing so and so out’

(Berliner 1994, p. 44), a term charged with visceral and violent overtones, describes one important aspect of the jam session. Musicians, often playing the same instrument, would spar with one another, the intent being to prove one’s technical prowess over the other musician. Berliner emphasises that this attitude underpins the competitive environment in which musicians developed their ‘chops’ and earned their reputation

(1994, p. 44).

These occurred virtually anywhere groups of musicians were to be found: after hours at professional clubs, at musicians' homes, or at recording studios. As Berliner

(1994, pp. 51–55) emphasises jam sessions were performances where musicians ‘paid their dues’, demonstrating their ability in front of their peers whilst reinforcing their personal style. Berliner (1994, pp. 51–55) also describes how musicians gathered together to play and share information, compare approaches, and reinforce their own knowledge of repertoire, theoretical information, and instrumental technique. Another aspect of this activity would often be hearing musical processes or witnessing instrumental techniques that may influence the future development of an individual’s style.

Berliner (1994, pp. 51–55) is not detailed about developmental specifics; he emphasises that this process did not occur in a structured or taught manner, but rather

112 through what could be described as situational osmosis. Correspondingly, the critical feedback received in jam session was less likely to be specific or accurate, and more likely to be implicit or at least inexplicit. Moreover, the ‘instruction’ offered by the jam session was likely to implicitly communicate whether a musician was up to the challenge. If not, musicians would know that they needed to rectify this shortfall by getting some ‘chops’, though they would rarely be left with details concerning the means whereby an aspirant musician could achieve this.

Like Berliner, Ake (2002) stresses the educational importance of these sessions. He crucially points out that they have largely been left out of the pedagogical structure of many modern-day university degree jazz programs, as they largely do not fit into the pervading educational philosophies of the conservatorium environment, previously specialised in teaching notated Western art music. Ake asserts that modern-day tertiary jazz programs are focussed on theoretical knowledge and individual practice, rather than on the group learning and creative interaction common to previous generations of musicians (2002, pp. 262–265).

4.3 The jazz performance format and approaches of mastery

Engaging a more specific focus, Berliner examines the way in which musicians have learnt to approach traditional vehicles of departure: songs and related chord changes.

Interestingly, he draws upon the opinion of neo-traditionalist Wynton Marsalis, who sees jazz as an exclusive craft, and also the more inclusive insights of distinctive saxophonist Lee Konitz. Highlighting the multiplicity of interpretations that jazz musicians have taken when performing within their tradition, Berliner compares the opinions that Marsalis and Konitz hold on how to manipulate the songs that are used in the improvisational process.

113 As introduced in Chapter Two, the expansive repertoire of jazz songs that are used as the springboard for improvisations is drawn from a history of ‘spirituals, marches, rags, and popular songs’ (Berliner 1994, p. 63). Correspondingly, the underlying harmonic models that many of these standard songs share are the frameworks on which melodic improvisations are based. Although possibly an opinion that is too conservative for many of his contemporaries, Marsalis’ view, that jazz is a traditional music where musicians are required to meet various stylistic expectations to be considered to be performing ‘jazz’, undeniably represents one shape that jazz has taken today. As

Berliner quotes Marsalis: ‘Jazz in not just, “Well, man, this is what I feel like playing.”

It’s a very structured thing that comes down from a tradition and requires a lot of thought and study’ (Berliner 1994, p. 63). On the other hand, Berliner compares this with Konitz’s more flexible view: that ‘Jazz tunes are great vehicles. They are forms that can be used and reused. Their implications are infinite’ (Berliner 1994, p. 63).

Seemingly disparate, the point at which Marsalis and Konitz agree is summarised in

Konitz’s advice to learners, who he says must ‘become familiar with these tunes and their frameworks before taking any liberties in playing variations or in improvising’

(Berliner 1994, p. 64). While the final musical products performed by Marsalis and

Konitz differ stylistically and are derived from different ideologies, the question as to whether or not either style of interpretation is valid is of secondary importance to the underlying message: that knowledge of, and adherence to the song form is paramount.

As such, it is vital for jazz musicians to master the many harmonic structures that have been derived from these songs, or indeed, any song or structure that is chosen as a model of variation.

114 4.4 Learning the jazz harmonic context

4.4.1 Memorising the melody

Berliner describes the process through which musicians eventually master harmonic structures by drawing a developmental continuum. This process starts with learning the melody (the original prototype) and evolves to a personalised rendering of the melody.

Individual expressiveness is at the very least, and sometimes first, created through the use of techniques such as vibrato, microtonal intonation, slides (portamenti, glissandi), and vocal mimicking. After any or all of several intervening stages, the process reaches the

highest level of intensity, transforming the melody into patterns bearing little or no resemblance to the original model or using models altogether alternative to the melody as the basis for inventing new phrases (Berliner 1994, p. 70).

The manner in which Berliner describes this process is not reductionist – that is, it is not progressive and there is no universal starting point and no final knowledge destination.

Musicians enter the continuum at any point, or commonly at multiple points, and assimilate as much information as they are able to, engaging in ‘the combined operations from interpretation to improvisation’ (Berliner 1994, p. 70). Contrary to this characteristically individualised process, publications by jazz pedagogues such as Coker

(1986) and Baker (1989) have attempted to formalise the process. They prescribe a progressive approach that incorporates theoretical information and recommend a series of practical exercises. Whilst these texts share a selection of pedagogical themes, the variety and inherent subjectivity of their suggestions attest to the multiplicity of options recounted by Berliner.

The ultimate goal for musicians who aim to improvise melodically is to have adequate flexibility manoeuvring through the harmonic progression, such that they are able to play a new melody that realises the underlying harmonic progression. Berliner

115 describes this goal as playing a new melody for the song, and quotes Konitz who asserts that the process of interpretation through to improvisation can ‘carry musicians more than halfway to creating a new song within the framework of another melody’ (Berliner

1994, p. 70). Indeed, Konitz’s improvisatory style is a particularly fine example of this very phenomenon.

Within the diversity of learning approaches, one step, normally the first, would be to memorise the melody. Berliner portrays the general procedure as autodidactic, where musicians learn the melody aurally from recordings or live renditions, or visually from a score (Berliner 1994, p. 64). Scores are valuable as a short-cut to the time-consuming process of aurally transcribing the melody from recordings, and of course, very useful in the absence of recordings altogether, inspiring non-music-readers to learn to read in order to gain access to this information. However, it must be emphasised that learning the melody by ear without the benefit of a score is by far the most preferred method, given the improvement this generates in a musician’s aural acuity, even though it may initially seem to take more effort for those skilled at reading scores.

Apart from the practical benefits gained through the process of memorisation, the importance of memorising the melody is threefold. Firstly, by memorising melodies aurally, musicians more thoroughly learn essential aspects of style that evade effective notation: timbre, subtleties of timing, articulation, and inflection. This stage of learning is crucial. Secondly, in professional practice musicians are required to perform these standard songs from memory. In doing so they are afforded a third and more important benefit: by memorising the melody they absorb a prototype that illustrates a voice of the harmonic progression. By memorising a melody, a musician assimilates one voice part of the song’s harmonic movement. In addition, they become accustomed to the structure of the phrasing, which similarly often reflects the song’s harmonic progression. This

116 provides the musician with one example of how to voice lead from one chord to the next, which opens their mind to the melodic possibilities available through the improvisational process by imparting a sense of both phrasing and harmonic tension and release. It also potentially allows future improvisations to adhere to the ebb and flow of the original melody.

Rather than viewing this developmental stage (memorising the melody), as a retrospectively designed learning process, it is in fact a direct by-product of the aural jazz tradition, as is the improvisational process itself. Given that musicians often only had access to music on recordings or in live settings, they were only able to learn the music aurally. An aural learning environment naturally lends itself to experimentation to a far greater extent than learning visually from a fixed score, and as such, the improvisational process is best viewed as a sequential extension of the aural learning environment. As trumpeter Warren Kime summarises in Berliner’s research:

After I had been playing the melodies straight for a while … I started making little embellishments around them. Gradually my embellishments became more extensive, and eventually I learned how to improvise (Berliner 1994, p. 71).

Interestingly, Kime’s anecdote mirrors the process through which ‘improvisation’ became a central practice in the jazz tradition, and the melodic elaboration that he describes is a method that is still used by jazz musicians when improvising today

(DeVeaux and Giddins 2009, p. 43).

4.4.2 Benefits of aural memorisation: aural acuity and improvisational ability

Another issue Berliner raises at this point is the fact that performers, especially singers, choose their own tonic, leading many young instrumental players to realise that it is necessary for them to be able to play melodies in more than one key, and perhaps all twelve keys (Berliner 1994, p. 66), so that they are able to work with a range of band

117 leaders and singers. The process of learning melodies in this way, although extremely daunting for beginners, has the positive outcome of enabling a much more detailed understanding of a song’s melodic structure. Also, given that learning songs in multiple keys is most often and most effectively undertaken by ear, the memorisation of the melody is consequently strengthened because musicians tend towards learning melodies in terms of intervallic relationships and relations to chords, especially the tonic, rather than as note names.

This has the additional benefit of contributing substantially to, and empowering a musician’s sense of pitch (Berliner 1994, pp. 66–67). Detaching from reliance on a visual stimulus, causing this strengthening of aural acuity, also importantly impacts directly upon future improvisational ability. McPherson (1995, p. 118) reports a clear correlation between an ability to play by ear and an ability to improvise: the benefits of the jazz tradition’s practice of learning and playing aurally is paid clear testament by

McPherson’s research.

This issue of pitch is reinforced by Berliner, who outlines the advantage that those who possess what he labels as ‘perfect pitch’ have over musicians who do not possess the ability of pitch memory-recognition (Berliner 1994, p. 71). On this issue, Berliner’s assessment dwells on the popular mythology of perfect pitch, known in the literature as

‘absolute pitch’. However, the alternative ‘relative pitch’ – the ability to aurally recognise intervallic relationships between notes horizontally and vertically, and replicate these relationships (once a tonic is known) – is equally useful. It allows musicians to identify intervals and chords directly. Furthermore, absolute pitch can have a negative effect on musicians who play transposing instruments, given that the pitch names on their instruments do not adhere to concert pitch note names, whereas relative

118 pitch allows them to realise pitches and intervals without having to relate them to a concert pitch.

4.4.3 Individual processes of memorisation

Knowing the melody and then developing this knowledge to the complete memorisation of a song’s harmonic progression is a rigorous and intensive process. Like all memorisation, it takes an enormous amount of time and determination. For some musicians, the process is aided by an excellent sense of pitch, but for most, until pitch skills are developed, it is a slow and arduous process. Of Berliner’s respondents, pianists, bassists, or other instrumentalists with functional piano skills are identified as having an advantage in both understanding functional harmony and recognising chords and progressions aurally (Berliner 1994, p. 72). Furthermore, Berliner highlights a number of personalised systems that musicians use to memorise harmonic progressions, describing a series of ‘kinetic and visual conceptualizations’ such as visualising harmonic forms as states of changing tension, or visualising chords as finger maps on instruments that outline the voicing of chords (Berliner 1994, p. 72).

Berliner emphasises that the aural and visual processes that many musicians engage in when memorising a song’s melody and harmony are often supplemented by learning

Western functional harmony. John Coltrane is well known for having studied numerous harmony textbooks (Porter 1998, pp. 149–150) and Berliner describes a similar autodidactic process being undertaken by some of his respondents, including pianist

Kenny Barron (Berliner 1994, p. 73). Other musicians who Berliner interviewed indicated that their theoretical knowledge of harmony was learnt in high school, taught in the Western classical tradition (Berliner 1994, p. 73), and Miles Davis21 viewed the

21 Davis took an interest in Western art music over the course of his career, being introduced to the music of Igor Stravinsky at Julliard (Szwed 2003, p. 44) and to the music of John Cage by Gil Evans, who was

119 training he received in Western classical music, encountered during his time studying at the Julliard School of Music, as a valuable educational experience (Szwed 2003, p. 44).

Berliner emphasises that awareness and comprehension of functional harmony have ramifications beyond simply understanding how one chord flows and interacts with neighbouring chords. They allow musicians to extract various sequences of chords and categorise these progressions according to their function. They are then able to recognise similar sequences in other songs and develop methodologies (both learnt from other musicians and originally conceived) for improvising through such chord progressions. Berliner’s research uses the example of the ‘ii V’ progression to illustrate how a series of chords can be extracted and recognised in other songs (Berliner 1994, p.

70), and he also identifies other harmonic prototypes: the ‘rhythm’ changes and the blues progression, both of which contain internal examples of the ‘ii V’ prototype and other chord series (Berliner 1994, p. 76). By noticing the repeated use of similar chord progressions, musicians can practice improvising over these harmonic contexts abstracted from the whole. Berliner also calls attention to the fact that by recognising

‘parallel cases’ of chord progressions as a way of summarising harmonic information, musicians are assisted in learning the harmonic structure of songs by reducing the number of steps involved in the memorisation process. The progression is reduced from single chords to a collection of prototype progressions (1994, p. 78).

4.5 Memorising and transcribing improvisations

Improvised sections of music are also memorised by jazz musicians, normally after the melody has been memorised, although some attempt memorisation of the improvised parts of recordings before the melody. Berliner describes how musicians commonly also a Stravinsky devotee (Szwed 2003, p. 69). Through this process, Davis, like other jazz musicians, gained an understanding of recent developments in Western art music, particularly with regard to harmony and compositional techniques.

120 start by learning subsections of a solo before frequently memorising whole solos by ear

(1994, pp. 95–98). Notations (transcriptions) are sometimes made so that analysis may be undertaken visually, though often solos are only memorised and not notated.

Frequently musicians sing the improvisations before playing them on instruments, while at other times they may bypass singing, copying them directly onto instruments.

Emphasising how singing these solos can greatly improve aural skills, Berliner reinforces that this process takes many hours of repetitive and laborious practice (1994, p. 96). In addition to reproducing the series of pitches and rhythms accurately, tested in real time against the played recording, the process of copying solos may also extend to the emulation of every nuance of a player’s tone, inflections, vibrato, articulation, and dynamics (1994, pp. 97).

The importance of transcribing solos (notating) in the development of improvisation skills in the jazz tradition is a regular anecdote and a common prescription for students. Supporting Berliner’s description of students learning solos and practising them in written and memorised form as daily studies in technique, form, and style, Thomas Owens also notes the role that transcriptions can provide in retrospective analysis (Owens 2002, p. 293). Although Owens and Berliner both explain the ways in which Western notation fails to capture the entire essence of a musician’s performance, transcribing remains a widely prescribed suggestion. Jazz educator David

Baker represents this common teaching maxim, recommending it to students in his ‘Jazz

Pedagogy’ (Baker 1989, p. 173). Yet, although transcribing remains a common prescription in teaching manuals, jazz textbooks, and private lessons, there is a distinct lack of information in the literature regarding exactly what benefits it may have over and above those gained by aurally copying and memorising. Even the benefits of memorising improvisations by ear are not well documented.

121 The popular belief is that the process of transcription or memorisation by ear allows musicians to acquire a growing vocabulary of phrases to be used during improvisation.

Although Berliner – like Baker, Owens, and numerous other jazz educators and writers

– implies that the process of transcribing and memorising solos is especially important for ‘Getting your vocabulary straight’ (1994, p. 95), the actual process that occurs as students move from learning solos to improvising without directly performing excerpts from memorised solos remains quite clouded.

Descriptions abound of how undertaking this process allows students to absorb the essence of jazz style, much like how repeatedly copying an American accent from television would improve an Australian’s ability to speak with this accent. This aspect of the process, in as much as it applies to absorbing the stylistic nuances of the jazz tradition, is obvious from the outset given the easy comparison to developing different accents with the voice. Yet, the literature is sparse when it comes to a description of how the processes allow students to develop actual original melodic material for solos.

Students may well be justified in wondering exactly how the correct phraseology will find its way into their playing. Is it through the process of memorisation and repetition, or is a more deliberate practice methodology required as part of the transcription or memorisation process? Is there another process at play that allows students to develop improvisational ability through the process of memorising, transcribing, and practising improvised solos?

Berliner (1994, p. 101) briefly discusses how students can derive ‘discrete patterns’ from solos for use in their own playing, building up a repertory of patterns and phrases.

However, his account of this process is presented more as a temporary measure for young students beginning to improvise, a process that gives them something to play during solos. It is ambiguous as to how students bridge the gap between developing a

122 repertoire of derivative phrases for improvisation and improvising their own melodies that encapsulate the sound of the jazz tradition but are concurrently unique melodic formulations.

Prouty (2004) offers some evidence regarding how this process aids in the development of improvisational ability, suggesting that practising extracted melodic units as patterns in all twelve keys aids students in the development of harmonic fluency. He writes that the ability to spontaneously transpose phrases through different key centres ‘is considered to be a hallmark in separating beginning players from advanced ones’ (2004, p. 9). Yet, whilst Prouty offers more insight into this process, the leap from performing memorised patterns to improvising an original solo that adheres to jazz style remains clouded.

Do students develop the skill through careful analysis, studying the intervallic, modal, and harmonic content of prototype phrases, applying these structural parameters to the composition of their own phrases in practice sessions and then spontaneously in course of performance? Or does the ability develop through aural osmosis whereby the compositional principles that represent jazz stylistic parameters are absorbed and employed intuitively by musicians when improvising?

4.6 Towards procedural automation

Even though various personal systems are developed to make the process of memorisation of both song structures and improvised solo content easier, Berliner’s analysis reveals that the common thread to all experiences is actually many years of regular daily practice. This process is required in order to gain a functional control of the knowledge needed for jazz performance, to the extent that recalling the information

123 in the course of performance becomes, as one of Berliner’s respondents described, ‘like breathing’ (Berliner 1994, p. 82).

This concept – that rigorous practice elevates a musician’s performance to be as though it is automatic – is an important juncture in this analysis of skill development for improvised performance. Indeed, this elevated capability that a musician achieves may be described as procedural automation. A similar notion was introduced in Chapter

Three where I detailed the case of Hindustani musicians gaining improvisational fluency by developing an experiential understanding through repetitive practice of tān and tihāī exercises. The significance of procedural automation to improvisational fluency in both Hindustani classical music and jazz compels further analysis. Sudnow’s explanation of his own learning processes and performance practices provides a more detailed insight into this concept (1978, 2001).

4.6.1 Phenomenology and automatic processes

In Ways of the Hand: The Organization of Improvised Conduct, (2001; re-written and revised since its original 1978 publication), Sudnow provides a detailed description and analysis of how he developed the ability to improvise through a study that hints at a loose philosophical connection to Merleau-Ponty’s Phenomenology. Although specifically examining piano playing and related piano-specific improvisational skills,

Sudnow’s study investigates some ideas that are relevant to any instrumentalist attempting to develop the ability to improvise in the jazz tradition.

Supporting Berliner’s account of the process towards procedural automation,

Sudnow describes how what he identifies as ‘at-hand’22 actions were developed through repeated practice over many years of certain finger and hand position pathways, allowing him to develop an effortless technique supplied by a vocabulary of these at- 22 A term borrowed from Merleau-Ponty’s philosophy.

124 hand movements (2001, p. 53). He also describes in more detail how the processes of improvisation interacted and became like other parts of his body’s self-knowledge, ‘in a terrain nexus of hands and keyboards whose surfaces had become known as the surfaces of my tongue, teeth, and palate are known to each other’ (2001, p. 127). In this description Sudnow is suggesting a form of intuitive functional muscle memory that interacts in the spontaneous moment of performance to realise improvisations, though his analysis lacks detailed cognitive evidence.

4.6.2 Speech and improvisation analogies

Investigating how actions can be carried out in a manner that is seemingly automatic,

Sudnow also touches on a widely discussed analogy between language and music, briefly introducing concepts such as ‘paralinguistics’ and ‘a sentence structure’

(Sudnow 2001, p. 56) as he notices them appearing in his playing. However, in this territory his assertions are shallow and he does not enter into any detailed analysis. This may have actually saved his investigation from making the same mistakes as much other research that attempts to view music as a language, missing the vital point that music does not communicate objective propositions, and as such, does not serve the same purpose as a language (Culicover 2005, p. 233). Yet, numerous writers in the field of jazz research put these problems of language and music comparisons to one side, and continue to draw out more specific aspects of this comparative discourse. Relevant to this discussion are those that focus on speaking and improvisation.

As Sudnow hints, there are structural elements in his own improvisations that are comparable to those found in language. Berliner (1994, p. 92) extends this retrospectively to a comparison with learning to speak:

Just as children learn to speak their native language by imitating older competent speakers, so young musicians learn to speak jazz by imitating

125 seasoned improvisers. In part, this involves acquiring a complex vocabulary of conventional phrases and phrase components, which improvisers draw upon in formulating the melody of a jazz solo (Berliner 1994, p. 95).

I will examine this assertion in more detail in order to determine its validity.

At the outset, I will advocate that a comparison between developing the ability to improvise and learning to speak is appropriate. However, Berliner’s theory must be clarified and developed before it can be used to describe how musicians learn to improvise. Firstly, I will specify that the statement: ‘acquiring a complex vocabulary of conventional phrases and phrase components’ (Berliner 1994, p. 95) is not a valid comparison to how children learn to speak. I will propose an alternative. Secondly, I will stipulate exactly what the statement that ‘improvisers draw upon in formulating the melody of a jazz solo’ (Berliner 1994, p. 95) needs to refer to in order to work as an acceptable comparison with learning to speak.

Linguist Paul Culicover suggests,

Comparable improvisation is of course what is done during the course of producing language. The jazz musician’s intuition is that the way it is done in jazz is that the musician learns a ‘vocabulary’ of ways to realize certain forms that satisfy the various constraints … practiced until they become automatic (2005, p. 235).

On close inspection this claim subtly contradicts Berliner’s summary, as Culicover does not imply that the musician memorises a ‘complex vocabulary of conventional phrases and phrase components’ (Berliner 1994, p. 92), but rather ‘learns a “vocabulary” of methods to realize certain rules and forms that satisfy the various constraints’. Noam

Chomsky’s revolutionary work of the 1950s provides further insight.

Investigating transformational-generative grammar, Chomsky asserted that the role of rote-learnt sentences in ordinary language usage was small (MacKenzie 2000, p.

173). When comparing this to improvisation, the rote-learnt sentences would represent the acquired conventional phrases that Berliner describes (1994, p. 92), and as such, this

126 part of his comparison between speech acquisition and the development of an ability to improvise requires modification. As Rita Aiello summarises Chomsky’s theory,

the child subconsciously abstracts the rules of language structure and utilizes them to create his or her own novel sentences. Children do not just repeat the sentences they hear, but create new sentences by applying the rules they have abstracted from their language (Aiello 1994, pp. 41–42).

Taking Chomsky’s theory and applying it to music acquisition, Aiello states that

In order to learn and to create music, children abstract some of the rules of their musical culture and use them creatively. Just as they learn spontaneously the rules that allow them to originate an unlimited number of sentences, so they seem to learn the musical rules that allow them to create novel sequences in music (1994, p. 42).

This is a succinct summary of how an analogy between the acquisition of speech and improvisational ability may hold true, whilst also supporting Culicover’s assertion

(2005) that musicians actually internalise a vocabulary of solutions that enables them to intuitively realise improvised melodies that adhere to certain rules and forms, rather than a vocabulary of phrases.

Putting this problem with Berliner’s claim to one side, and examining the second half of Berliner’s assertion: ‘improvisers draw upon [ ____ ] in formulating the melody of a jazz solo’ (Berliner 1994, p. 95), I suggest that in place of the ‘____’ that I have inserted in Berliner’s original sentence, this should be replaced by ‘the rules and forms’, as highlighted by both Culicover (2005) and Aiello (1994). As such, inserting another sentence between the two that I have extracted from Berliner (1994, p. 95) helps to further clarify Berliner’s assertion, such that the whole statement reads more accurately as a comparison between learning to speak and developing the ability to improvise.

Just as children learn to speak their native language by imitating older competent speakers, so young musicians learn to speak jazz by imitating seasoned improvisers. In part, this involves acquiring a complex vocabulary of conventional phrases and phrase components. By internalising this vocabulary of conventional phrases and phrase components, musicians subconsciously abstract the rules and forms of jazz structure and style, utilizing this knowledge

127 intuitively when formulating the melody of a jazz solo (My revision of Berliner 1994, p. 95).

Thus, my contention is that practising transcribed solos and their derived phrases enables an improvisational ability to be developed, and by comparison to the linguistic theory of Chomsky adapted to a musical context by Culicover (2005) and Aiello (1994),

I suggest that this occurs through the assimilation of structural rules, allowing musicians to spontaneously generate phrases, rather than reorganising rote learnt phrases during performance.

This is not to suggest that similarities at the phrase level do not exist between multiple improvisations by the same musician (Owens 1974, Solis 2004, Weisberg et al.

2004). However, the analogies I presented above between improvisation and linguistic theory suggest that phrase level similarities should not be taken as proof for the existence of a paradigmatic process of acquirement through which a storehouse or vocabulary of such phrases is developed. Equally, I assert that writers such as Knox

(2004) are erroneous when they suggest that aspiring improvisers should develop a vocabulary of phrases so that this collection can be reorganised when improvising.

Finally, my contention does not exclude the possibility that musicians play learnt phrases from time to time, as this does occur, especially in the form of licks and historically referenced musical quotes from other musician’s solos (Tirro 1974).

However, this does not represent the complete creative process whereby improvisation is facilitated by an assimilated vocabulary of structural rules that define specific musical cultures.

As a concluding thought to the current discussion of analogies between improvisation and speech, a useful extension of these insights is found in the research of

Canadian music educator Jan Steele. Steele’s research (2004) summarises how a second language speaker can acquire improved fluency. She emphasises the now well-known

128 pedagogical tenet: that when developing the ability to speak in a second language, after a foundation has been taught explicitly, the most effective means through which learners can gain fluency is when they ‘engage in meaningful interaction in the second language’ (Steele 2004, p. 26). Emphasising a movement beyond didactic teaching,

Steele implies a type of oral-aural-osmosis. Reflecting this onto Berliner’s research, he would appear to concur in stating that only when musicians immerse themselves in the tradition, assisted by ‘fluent speakers of jazz’, are musicians able to grasp the ‘cadences of fluent native speakers’ and ‘construct complex musical statements’ (Berliner 1994, p.

102).

Contextualising all opinions presented above, the most accurate picture of the improvisatory process is one in which musicians spontaneously perform phrases that are intuitively composed in the moment, based on absorbed parameters derived from melodic prototypes arduously practised to the point of internalisation. Whilst all opinions agree that the improvisational process is an achievable by-product of the memorisation and transcription process (whose role in the development of an ability to improvise in the jazz tradition is therefore paramount and undisputed), it is vital that students are aware that their goal when transcribing is not to collect a vocabulary of phrases.

4.7 Jazz in the university and college curriculum

4.7.1 The history of jazz education

It has been demonstrated that jazz musicians have traditionally developed their ability to improvise in a variety of learning environments, beginning in childhood and extending beyond their professional debut throughout their career. Their education may begin from within their family or social group, be encouraged by opportunities presented at

129 church, school, or other community organisations, and at times their skills are refined through study with a private instrumental teacher. Their skills have then traditionally been developed in informal socio-professional networks such as jam sessions, whilst their development, by necessity in confronting the complexity of challenges that confront a jazz improviser, must always be stimulated by arduous autodidacticism. This traditional model has undergone significant transformations over the last 60 years as learning experiences have increasingly incorporated an institutional component.

Beginning in the 1940s, the learning environments in which jazz musicians engaged expanded into tertiary education. Berliner (1994, pp. 55–57), Ake (2002, pp.

262–265), Prouty (2005), and Marquis (1998, p. 117) outline how and when jazz education was accepted into a handful of university programs between 1945 and the

1960s. Ake identifies the Berklee School of Music in Boston (now the Berklee College of Music) as the first institution teaching jazz, opening its doors in 1945, followed in

1947 by North Texas State Teachers College (now the University of North Texas) which offered the first degree program in jazz (dance band) music from 1947 (2002, p.

262). Courses in jazz education proliferated henceforth.

Prouty asserts, however, that this ‘coherent, logically flowing institutional narrative’ (2005, p. 85) is flawed with respect to the impact it has on the account of jazz education history. He emphasises that highlighting ‘the creation of jazz education programs in Texas and Boston in the late 1940s … as being an important temporal marker in an evolving historical narrative’ (2005, p. 85) is injudicious in as much as the birth of legitimate jazz education is thus aligned with the growth of tertiary jazz degree programs. In doing so the efforts of individuals who developed and taught jazz in a time before institutions allowed jazz to be studied formally are ignored or misrepresented

(Prouty 2005, p. 99).

130 However, other writers do identify individuals. In his historical summary, Ake identify the efforts of Len Bowden who held various positions teaching jazz at college level and within the defence force from 1919 (2002, p. 262). Similarly, McDaniel reinforces that W.C. Handy held a bandmaster position from 1900 teaching what would be later called jazz at the ‘Teacher’s Agricultural and Mechanical College for Negroes at Normal, Alabama’ (1993, p. 115). Even so, Prouty’s contention is that highlighting the efforts of these individuals within a grander institutional narrative relegates their efforts to the ‘pre-historic’ in the chronology of jazz education (2005, p. 82).

Nevertheless, in addition to raising the efforts of these and other individuals to a status whereby they are recognised as important to other chapters in the story of jazz education (2005, p. 99), Prouty also proposes a total realignment of the jazz education historical narrative so that jazz education and the jazz community are linked in a cultural and historical perspective (2005, p. 88). Prouty points out that isolating institutions in the hitherto accepted narrative has disconnected the institution from the jazz community (2005, p. 100). Consequently, important historical advances in jazz, such as George Russell’s Lydian chromatic theory, the career-spanning musical developments of many musicians, and the work of jazz artist-educators such as David

Baker and Barry Harris have become disconnected islands in jazz history, rather than bridges between the jazz community and the educational institutions that contribute much to the formal training of jazz musicians today (2005, p. 100). Prouty’s conclusion is that jazz education history must be ‘inclusive of the historical traditions of jazz as well as the historical narratives of educators within academia’ (2005, p. 100).

131 4.7.2 The nature and content of jazz pedagogy at the tertiary level

I will now engage in an exposition and analysis of the tertiary environment encountered by students of jazz today. This evaluation will analyse the content and procedures used to instruct jazz students in modern contexts, and will draw conclusions concerning the appropriateness of this pedagogical structure, specifically as it relates to developing improvisational ability.

As I outlined in Chapter Three, institutionalisation has been comparatively successful in some Hindustani classical examples because they have adapted the traditional didactic model and methods into a modern tertiary context. Indeed successful institutionalisation has allowed the instructive guru-shisyā parampara, which has gradually been a less viable didactic relationship, to survive in a modified form.

Accordingly, given the relative parity between traditional and modern learning environments, my analysis of institutionalisation as it relates to learning to improvise in

Hindustani classical music has not required a detailed critique.

However, in the case of the jazz tradition, given that autodidacticism has historically been a far more prominent feature of developing improvisational ability I will undertake an appraisal of didactic tertiary jazz programs. I will ascertain if methods used to teach improvisation differ from methods employed before institutionalisation took place, while establishing if these methods are successful in developing improvisational ability. In doing so, the importance of imitation, the principal means through which musicians learn to improvise in the jazz tradition will be reinforced.

In this area assertions made by Prouty are significant to the focus of this critique because they expose weaknesses inherent to the tertiary environment in which many students of jazz learn today. Specifically, this weakness is due to the fact that academic degree programs, in a self-reinforcing manner, reflect the aforementioned historical

132 narrative that imposes a separation between the jazz community and academic institution. The deficiencies that this imposes upon the curriculum of tertiary jazz programs are summarised by Prouty, in which his claims are based on the contention that

jazz education draws upon distinct canons of musical study, those of the jazz community and of the academic institution, and that the tension between these two systems imparts a profound influence on the construction and application of teaching strategies (Prouty 2004, p. 1).

This tension emerged when jazz programs were superimposed onto a pre-existing academic framework that had been developed specifically to produce Western art music performers. The structure of this pedagogical system was designed to teach a notated musical tradition in which performers realise pre-composed scores. Also inherent in this methodology is a structured curriculum of progressive sequential development combined with expectations that methods of evaluation can be applied to a large group of students (Prouty 2004, p. 13). As such, judging and grading academic work and musical performances tend to occur according to an established scheme in which performances are compared to an ideal interpretation of stylistic and technical perfection.

However, Prouty emphasises that

Whilst the demands of the academy are satisfied through a sense of structure and curricular/pedagogical regularity, the aesthetic demands of the jazz tradition for individuality and intuition exert a powerful pull on educators and students (Prouty 2004, p. 13).

The institutional environment does not resolve this tension, and unfortunately students are often left confused as to what they should be aiming for when they perform. As

Prouty illustrates:

Students whose playing is criticized for being too technical complain that this is what they are taught. Those whose playing does not meet a certain standard for technical or stylistic appropriateness likewise see such evaluations as too rigid,

133 disregarding the individualism long cherished as a marker of identity in jazz (2004, p 13).

This phenomenon is also reinforced in Ake’s examination of institutionalised jazz education (2002). Quoting Berklee-trained saxophonist Branford Marsalis, Ake emphasises the problem:

Berklee has its own system of doing things, the Berklee way, the Berklee method. They basically say that when you write things that are theoretically against the Berklee method, then they’re incorrect. Even if they sound great, but theoretically it’s wrong, so it’s wrong. Which is not the purpose of music. Music theories are just theories (Ake 2002, p. 144).

Marquis, in her account of jazz’s transition from marginalised art form to accepted academic pursuit, one that has been facilitated by the existence of colleges such as

Berklee, correspondingly inquires, whilst jazz has ‘won a place for itself at the center of

American culture … Now, swaddled inside the velvet cage of academic music, can real jazz survive?’ (Marquis 1998, p. 122).

The specifics of what Prouty, Ake, and Marquis assert rest in the details of methods used in institutionalised jazz programs to instruct students in improvisation.

Summarising a myriad of curricula, Prouty identifies the major areas of instruction that most syllabi conform to: (i) repertoire and associated harmonic principles (2004, p. 3), and (ii) improvisation pedagogy and pitch structures (2004, p. 6).

By standardising compositions into a sequential syllabus in which repertoire is taught to students in a progressive manner, tunes are organised with respect to their relative harmonic complexity from easiest to most difficult. Students are usually taught blues-based and modal tunes first, because ‘These types of harmonic structures, it is argued, provide an easier vehicle for students to improvise’ and ‘From this starting point, students then move on to more harmonically challenging tunes, such as those based upon simple diatonic cadential patterns’ (Prouty 2004, p. 3). By the later stages, students arrive at pieces that provide the greatest level of harmonic complexity, namely

134 ‘tonalities based on melodic/harmonic minor scales and their related modes … the

“upper end” of the standard bebop foundation of the improvisational language’ (Prouty

2004, p. 4).

The contention that Prouty holds regarding this arm of tertiary jazz curricula is that the whole decision-making process as to what tunes are studied and at what stage each is studied is entirely subjective. The sequential order of repertoire and related harmonic forms that are studied are chosen by educators, based upon what they consider appropriate for students to learn. In doing so, Prouty (2004, pp. 3–4) asserts that some tunes (inevitably) are left out of the sequence, some harmonic progressions are emphasised at the expense of others, and the subjective opinion of individual educators has an unjustifiably biased impact on the education of tertiary students. Indeed, Prouty exposes his own subjective bias in noting the absence of some tunes and harmonic progressions. However, his point is taken.

Perhaps of a more fundamental concern, by undertaking this sequential standardisation of repertoire and placing a focus so heavily on harmony, the importance of ‘more esoteric, intangible aspects of jazz performance’ such as those largely learned through imitation are de-emphasised ‘in favour of technical harmonic competence and lessening the amount of individual creativity in jazz performance’ (Prouty 2004, p. 4).

Not only has this reprioritisation been carried out on individual tertiary music campuses, but the main American tertiary music accreditation body in the USA, the National

Association of Schools of Music, has set curricular standards that embody these very principles (Prouty 2004, pp. 5–6).

This has had the effect of spreading the issue nationally: curriculum conformity is necessary for accreditation. On a practical level, jazz historian James Lincoln Collier, as highlighted by Prouty, neatly articulates this danger. Collier argues:

135 With students all over the United States being taught more or less the same harmonic principles, it is hardly surprising that their solos tend to sound much the same. It is important for us to understand that many of the most influential jazz players developed their own personal harmonic schemes, very frequently because they had little training in theory and were forced to find it their own way … (Collier, in Prouty 2004, p. 4).

Individualism growing out of autodidacticism is at odds with the required conformity faced by the many students who study in modern North American jazz degree programs. By shifting the educational emphasis from learning to teaching, correspondingly disempowering the previously proactive autodidact in favour of the teacher, institutional methodologies are at odds with the nature of learning in the traditional jazz community that ‘places its emphasis on learning rather than on teaching’

(Berliner 1994, p. 51).

Furthermore, as tertiary students progress in learning tunes and their associated harmonic structures, various pedagogical approaches are taught that aim to enable improvisation over these harmonic progressions. By far the most common approach focuses on manipulating pitch, teaching students to relate scales to chords. This methodology is easily taught to students, and is further enhanced by play-along books with compact disk accompaniment. An example is the Jamey Aebersold play-along series, which features standardised melodies (heads) and solo sections where prescribed scales are notated into the bars below their corresponding chords.

Given that modes are clearly prescribed for each chord, students who are taught these chord-scale relationships, and who also commonly use play-along books in private practice sessions, are often directed to only one mode and chord combination. Sourcing this kind of information aurally from original recordings would solve this limitation, but given that this is a far more arduous task than simply reading from a score, students often treat the Aebersold books as an historically comprehensive record from which they make decisions on harmony, melody, tempo, and style (Prouty 2006, p. 328).

136 Furthermore, presenting students with a limited choice of notes promotes a style of improvisation that is scalic (horizontal), and students often discover that whilst they are playing the ‘right’ notes, their improvisations lack the sound inherent in jazz style. Ake succinctly summarises the crux of this problem:

Yet while this pedagogical approach does succeed for the most part in reducing ‘clams’ (notes heard as mistakes) and building ‘chops’ (virtuosity), it presents a number of major and related drawbacks as beginners strive to play idiomatically within various jazz contexts. Most obviously, this approach ignores the sonic and rhythmic conceptions [inherent to jazz style] (2002, p. 123).

This is not surprising, given that the method through which they are taught to improvise, according to chord-scale theory, differs substantially from traditional methods of development experienced throughout jazz history prior to the arrival of chord-scale theory and play-along books, as discussed at length earlier in this chapter.

Furthermore, Prouty (2006) extends criticism of the Aebersold play-along series and associated chord-scale theory, emphasising the extent to which

Some critics … argue that Aebersold’s recordings lead to a sense of stagnation, limiting the opportunities for real interaction with other musicians. Rather than simply providing a means of interaction, they suggest that such recordings actually inhibit eventual interaction with live musicians because students focus solely on harmonic and technical aspects (2006, p. 328).

This chord-scale theory, identified by Prouty as a theoretically based approach, is compared in his analysis of jazz improvisation pedagogy to practice-based approaches

(2004, p. 6). These approaches are the single most important component of tertiary jazz programs because they act as a bridge between jazz education history and modern jazz pedagogy, and as such neutralise the potentially negative impact caused by chord-scale theory.

In such approaches, educators directly draw materials used to teach improvisation from sources such as recordings and transcriptions of improvisations (Prouty 2004, p.

6). In highlighting this more positive aspect of modern pedagogy, it is important to

137 emphasise that in past learning environments solos were memorised or transcribed by individuals operating as autodidacts, rather than in teacher-mediated circumstances

(Berliner 1994, pp. 95–98). This caveat established, Prouty reports that teachers recommend that students memorise melodic motifs – licks, patterns, phrases, or clichés of style – from historical sources (Prouty 2004, p. 8). However, in reference to my earlier assertion concerning the role of assimilated structural rules in improvisation, I will now determine the extent to which these aspects of pedagogy advocated positively by Prouty (2004, pp. 6–9) conform to the historically informed portrayal of improvisation in jazz that I have established by comparison to linguistic theory.

Echoing Berliner’s research (1994, p. 101), Prouty suggests that this practice-based method is an important arm of pedagogy because it allows students to develop a vocabulary of building blocks to be used as a source of ideas in their own improvisations (2005, p. 8). Furthermore, by learning another player’s melodic motifs, students are taught to understand an individual player’s style by deconstructing improvised phrases to be the sum of smaller melodic units (Prouty 2004, p. 9). Prouty suggests that from this point students may even be required to write their own phrases in the style of various master musicians (2004, p. 9). Suggesting that the intention of exercises such as these is to develop an understanding of jazz melodic style through analysis, Prouty reinforces that in doing so students are provided with models on which

‘to build their own musical vocabulary, to understand the ways in which jazz musicians have historically constructed melodic units’ (2004, p. 8–9).

Stemming from his analysis of practice-based pedagogical methods, Prouty derives

‘two major generative approaches to improvisational patterns’, the first being an attempt to ‘link pedagogy with the historical traditions of jazz improvisations’, and the second

‘a more structural orientation, portraying patterns as musical/technical constructs, in a

138 sense, abstractions of common melodic practices’ (2004, p. 9). While both intentions certainly serve as important connections to past learning environments, the second is of vital importance in as much as it grants students access to the inner workings of jazz style.

Furthermore, the ‘more structural orientation’ allows the chord-scale theory that they may have learnt at an earlier stage of the education process to be transformed into a functional melodic idiom via an analysis and application of principles contained within these important melodic prototypes. The process articulated by Prouty enables students to assimilate vital compositional and structural rules that have been internalised after years of arduous practices, which then allows improvisation to take place according to jazz stylistic parameters. As such, this modern adaptation of historical methods conforms to my narrative of the improvisational process.

Prouty offers a picture of tertiary jazz education as one that needs reform. Whilst activities such as transcription, as taught by university and college teachers, do link the jazz curriculum to past learning environments, the homogenisation of style and technique in graduating musicians continues to be a major criticism that is levelled at tertiary jazz programs. Not surprisingly, the formal standardisation of repertoire, harmonic/modal theory, and subject content, not to mention the subjective influence exerted by single teachers on multiple generations of graduating jazz musicians, contribute inescapably to this standardisation.

However, while it may be argued that jazz education should return to a historical learning model where autodidacticism was central, it remains a fact that jazz pedagogy has changed from its original state for the vast majority of musicians. Some level of formalised didactic education has been evident in jazz education since the 1940s. In order to resolve the tension caused by applying an incompatible pedagogical structure

139 on a largely autodidactic practice, the outcomes of jazz programs could be realigned to the extent that they reflect the character of the jazz tradition. This would be a model that appreciates originality and innovation derived from autodidacticism, as opposed to the conformity that current pedagogy extracts from graduating students. As suggested by

Prouty, jazz education should be re-evaluated to the point that jazz educators ‘re- envision the academic study of jazz as an extension of the traditions of jazz performance, rather than as an isolated pedagogical system’ (2004, p. 1).

4.8 Conclusions

Jazz musicians have traditionally acted as autodidacts when developing the ability to improvise. From childhood, familial and social networks have provided osmotic learning environments through which musicians have first been inspired to experiment with music making. Taking experiences from these environments as models, musicians have developed the ability to improvise by engaging in self-motivated learning patterns facilitated by arduous long-term practice, known colloquially as ‘woodshedding’.

Within these individual practice sessions, musicians commonly first learn the melodies of those compositions viewed as ‘standard’. Whilst priming them for professional actives, learning these melodies and related chord progressions by ear facilitates the development of heightened aural acuity specifically with regard to pitch identification, manifested either in the development of absolute or relative pitch, or both. The development of these aural skills has been shown to be an important precursor to eventual improvisatory ability: a correlation exists between being able to play by ear and improvisational ability (McPherson 1995).

In addition to learning melodies by ear, musicians extend this practice to learning the improvisations of other musicians, enabled specifically by the aforementioned

140 development of aural pitch acuity. Though this is a practice which at first poses a significant challenge, by habitually aurally memorising and transcribing these solos, musicians begin to internalise a wide variety of prototypical solutions by imitating examples of how others have realised the underlying harmonic progressions of standard repertoire embodied in the creation of a new melody.

Whilst a variety of harmonic progressions exist within this ever-growing canonical repertoire, similar diatonic functionality is noticed between songs, and the reappearance of related harmonic challenges allows musicians to internalise model approaches, as absorbed through their transcribing activities. These prototypes are often truncated into sub-phrases known as licks, and the repetitive practice of these short-form patterns, often in multiple key centres, aids in the internalisation of the melodic characteristics of these models, whilst also developing fluent technique that is specifically related to traditional performance practice.

Additionally, this practice-derived knowledge is often supplemented with theoretical modal and harmonic knowledge that allows musicians to understand how progressions function harmonically. This theory is learnt by means of individual research and also through informal peer networks in which developing musicians share information. Combined, these methodologies arm musicians with a practically and theoretically derived understanding of how to realise a harmonic progression when improvising.

Initially in peer-moderated ‘jam sessions’ and eventually on the professional stage, musicians draw upon the internalised knowledge of phrase construction to spontaneously formulate new melodies in the course of performance that adhere to the stylistic parameters of the jazz tradition. Over the many years of detailed personal practice, much of which is dedicated to internalising prototypes and developing

141 personal melodic solutions based on these traditional models, improvisational ability is developed. Of vital importance is the aural nature of the autodidactic process, which eventually enables musicians to copy by ear anything that is reasonably perceivable.

During performance, musicians draw upon their internalised vocabulary of prototypical phraseology to play by ear a new melody that conforms to stylistic and technical parameters of the jazz tradition.

Today, this traditional manner of developing improvisational ability has been widely supplemented, if not replaced wholly for some music students, by tertiary music programs. The ramifications of this change have been discussed at length, and criticisms of teaching improvisation in this way are directed largely at the homogenisation of style that standardised curriculums and evaluative methodologies exacerbate, the over- pervasiveness of chord-scale theory, and the de-emphasis of individual approaches in favour of adherence to institutionalised standards of performance.

Whilst criticism accurately accentuates the degree to which modernised pedagogy has departed from traditional learning, in practice those musicians who study in modern settings benefit from a network that rarely exists in informal settings today.

Furthermore, whilst institutions can offer students an organised delivery of theoretical and historical information albeit inherently subjective and homogenised, the fact remains that in order to develop the requisite aural acuity, technical virtuosity, and internalisation of prototypes, musicians still need to engage in long-term individually motivated practice. As Berliner stresses

The jazz community’s traditional educational system places its emphasis on learning rather than on teaching, shifting to students the responsibility for determining what they need to learn, how they will go about learning, and from whom (Berliner 1994, p. 51).

142 Tertiary programs can only supplement individual effort, and as such, the traditional autodidactic approach that characterised the jazz tradition remains as much a part of a musician’s practice today as ever.

143 Chapter Five Learning to Improvise in the Egyptian Music Tradition

Ali Jihad Racy and Ann Elise Thomas have identified a variety of developmental pathways that musicians traverse in the Arab world. While my research focuses solely on Egyptian music and more specifically on learning to improvise in this genre, their work will be used as a general reference given that Egyptian music is part of the greater

Arabic music tradition. My more specific focus will proceed from this foundation.

5.1 The learning environments and pathways to becoming a musician

5.1.1 Childhood in Egypt and the greater Arab world

Racy (2003) identified five phases of development that summarise the process of becoming a tarab23 artist in the Arab world. These episodes consist of:

1) the appearance of talent, usually during childhood; 2) musical obsession, accompanied by struggle against family and cultural barriers; 3) family and societal recognition of budding talent, and in some cases reluctant acquiescence to the novice’s musical desires; 4) training of some sort; and 5) the undertaking of a performance career (2003, p. 18).

Racy is quick to point out that these stages may overlap, come in other orders, coincide with each other, and also may not apply to every musician (2003, pp. 18, 23). Yet, as a guiding principle, these phases provide a suitable framework from which my own research will expand.

23 Conceptually, tarab describes the ecstatic state that performer and audience enter into during heightened moments of intensity in Arabic music, especially as experienced in improvised taqāsīm performances. Racy labels performers as tarab artists, reinforcing the importance of achieving tarab in the course of performance.

144 Identifying that certain children show interest and respond quickly to the musical challenges of a given culture is a familiar and logical entry into studies of musical development. Indeed, the process whereby children are introduced to a musical tradition and its medium of expression, through to the point where they find themselves actively involved in the pursuit of developing the ability to perform in this context, is quite often a seamless one, at least initially. Esawi Dagher, an Egyptian violinist I spent time interviewing and informally studying and playing with in 2008, speaks about his childhood foray into the musical world with undertones of divine destiny and inevitable familial inheritance.

5.1.2 The family environment

Dagher, whose father and older brother were violinists, was encouraged to make the transition to violin by his brother once he demonstrated an interest in music as a child, initially playing the ‘ūd. For his elder brother Abdo, the now-famous elder statesman of

Egyptian violin known today as the ‘Malik At-Taqāsīm’, king of improvisation, the decision to play the violin, let alone pursue a career as a violinist, was met with fierce resistance from their father during his childhood, who ironically owned a music store and was also a violinist (Dagher 2000). Yet, the initial opposition that met Abdo Dagher is not surprising given that the life inherent in pursuing a musical career in Egypt can be difficult. Such a career may be financially unrewarding and uncertain, and some consider it to be a disreputable profession. Despite family resistance, Abdo Dagher left his father’s home at the age of 13 and began a difficult journey from Tanta to Cairo.

Negative attitudes encountered by Arabic musicians concerning their careers and associated lifestyles are also highlighted in research by Racy (2003, pp. 21–23) and

Thomas (2006, p. 149). Attesting to these notions of struggle and societal rejection,

145 another musician told me (off the record) that his family’s religious leader asked him how his sister would get married if he continued to pursue his apparently degrading lifestyle as a musician. In my conversations with Esawi Dagher, the type of struggle as endured by his older brother was not mentioned specifically; perhaps his older brother’s efforts made his path easier. In any case, Esawi and Abdo Dagher’s stories also illuminate the notion of inherited musical talent, a further point emphasised by Racy

(2003, p. 19).

Thomas is somewhat more pragmatic when discussing the inheritance of musical talent, emphasising that at particularly young ages it is rare for an Egyptian child to learn music from anyone other than a family member (2006 p. 150). Consequently, children will not be introduced to the same depth of beneficial soundscape in their youth unless they are born into a musical family; the notion of inherited talent is thus tempered in her analysis. Extending her point, it becomes clear that Egyptian children born into musical families have a practical advantage that allows their developing minds to engage in the music-making process earlier, more regularly, and in a more formalised manner than children who do not have musician parents.

In addition to not benefitting from a musical environment at home, the children of non-musicians rarely receive training until they are teenagers. As Thomas points out, the chief music institution in Cairo, The Higher Institute for Arabic Music admits 14- year-olds as its youngest students (Thomas 2006, p. 151). Students entering the institution at this age do not necessarily have any experience with musical education, familial or otherwise. One of my research participants, Dr Alfred Gamil, did not grow up with musician parents and was one such student who entered at 14, full of enthusiasm, though lacking experience.

146 Illustrative of the clear advantage gained by being born into a musical family, nāy24 player Mohammed Abdula Fouda told me during interview that his life as a musician began at the age of seven. Circumstances favoured this, given that his father was also a nāy player, and he fondly recounted stories of learning to play from his father in Upper

Egypt (the rural south). Like many musicians sharing this upbringing, and in agreement with the common tendency to link musical ability with an inherited talent (Racy 2003, p. 19), Fouda views his progression to becoming a professional musician as a hereditary endowment rather than as the consequence of practical convenience, as emphasised by

Thomas.

Emphasising the value placed in Egyptian society on the inheritance of musical talent, young vocalist Ahmad El Haggar spoke during interview of being from a long line of musicians. This inheritance stems from his father, the now-famous vocalist Ali

El Haggar, his uncle the composer, also named Ahmad El Haggar, and his grandfather, the late singer and composer Ibrahim El Haggar. Ahmad is following a vocal tradition for which he has received long-term support and encouragement from his family, embodied in regular Thursday and Friday singing sessions at his grandfather’s house from the age of three. This is obviously a markedly positive situation when compared to

Abdo Dagher’s struggle, and it represents one of numerous exceptions admitted in

Racy’s discussion (Racy 2003, p. 23).

5.1.3 The rise of Westernised environments

Nāy player Mohammed Antar’s childhood trajectory was initially more Westernised.

When interviewed in 2008, Antar described beginning piano lessons in 1982 at the age of four, facilitated by his encouraging parents. In doing so he was not following an

24 As introduced in Chapter Two, the nāy is an end blown bamboo flute played through out the Middle East and parts of Africa.

147 inherited musical pathway per se, even though his twin brother is also a musician. After studying Western music for eleven years he began learning the nāy, answering a desire to pursue Arabic music in a traditional setting. Similarly, ‘cellist Mahmoud Bedir said during interview that he began playing the ‘cello at the age of 14, following Western classical style pedagogy. After six years of lessons, Bedir transferred to playing in an

Arabic style, teaching himself and beginning improvising soon after.

In the mode of Antar and Bedir’s musical development, the education and career of

Egyptian multi-instrumentalist Dr Alfred Gamil also reflects the modernisation of music education that took place in Cairo after the 1932 Arabic Music Conference, as detailed in Chapter Two. Gamil, like Antar, is not from a family of musicians, although he described during interview growing up in a musically literate and supportive family. He showed an early interest in music, even going as far as fashioning his own makeshift

‘ūd out of an old box. Recognising his obvious interest in music, his parents enrolled him at the age of 14 in The Higher Institute for Arabic Music, and Gamil attended as many as five to six classes after school each week. Like Antar, Gamil represents an exception to the notion of inherited talent. This is particularly notable given the extent to which Gamil is a respected member of the Egyptian music community, and given the extent to which he has been a central figure in music education at The Higher Institute for Arabic Music for more than 30 years.

I interviewed Egyptian-born, Australian ‘ūd player Joseph Tawadros in 2007, and in doing so discovered that he is a fascinating case study in musical development within the Egyptian tradition. Even though neither of his parents are musicians and he grew up in the inner-eastern suburbs of Sydney, Australia, Tawadros managed to become well- versed in Egyptian music from a young age. The sounds of Egypt were a constant soundtrack to his familial atmosphere, enabled mostly by his mother’s love of

148 traditional Egyptian song. Even though Tawadros was more that 14,000 kilometres away from his native Cairo, his mother regularly played recordings from the radio that she had collected while still living in Cairo, as well as other recordings of famous singers such as Umm Kulthūm and Mohammed Abdel Wahab. Tawadros’ geographic isolation did not hamper his development, and even though his path was by necessity unique, it was not as distinctive as one may expect. The details of this will be elaborated upon later in this chapter.

My fieldwork in Cairo and Sydney demonstrated that the appearance of talent, usually exposed during childhood, is certainly stimulated in Egyptian children from musical families. They have a clear practical advantage. Yet, my research also demonstrates that other pathways can originate during childhood that allow children in

Egyptian society to develop outside of the familial paradigm, such as in the case of

Antar, Tawadros, and Gamil, contradicting the idea of the perceived superiority of inherited musical capability. In both cases, some kind of training takes place once an interest and ability has surfaced in childhood.

5.2 From childhood interest to training

5.2.1 Apprenticeship-like relationships

When an initiation into some form of music making has been made during childhood,

Egyptian musicians have followed various modes of progression, and it has been suggested that these paths have been changing quite markedly over the last 50 to 100 years. Racy suggests that apprenticeship-style learning environments were once the most common way of learning traditional music and that prior to World War I vocalists often gained their training in traditional Arabic music by singing the Sufi (Islamic) liturgies (Racy 2003, p. 25). He also notes that learning chanted renderings of the Koran

149 (Quran) was an efficient training arena for vocalists (2003, p. 26). Interestingly, during an interview in 2008, violinist Esawi Dagher told me that before to moving to Cairo, he had spent some time in 1960 performing on violin with sheikh singers in Upper Egypt.

He notes this as an essential part of his musical education. His elder brother Abdo also benefitted from the same experience, as Sufis provided him with shelter and sustenance after he left his father’s home (Dagher 2000). When interviewed in 2008, Mohammed

Antar suggested that playing with Sufis was a common training ground for musicians of

Esawi and Abdo Dagher’s generation who were born after World War I. This is beyond the time Racy suggests this process went into demise.

Thus, my research indicates that the chronological parameter established by Racy should actually be expanded so that these types of apprenticeship-like learning environments in Egypt, specifically with religious singers, are seen as an important learning situation for musicians at least until 1960, and perhaps later. The fact that the

Dagher brothers were engaging in this type of learning environment in rural areas outside inner-city Cairo may indicate that the trend away from these apprenticeship learning environments as highlighted by Racy may have taken a longer time to fade out of practice in rural Upper Egypt due to a greater conservatism or slower modernisation.

However, given Abdo Dagher’s prominence as the leading improvising violinist in

Egypt today, the significance of this learning environment should be emphasised. His role in mentoring multiple generations of violinists, including his own grandson, extends the heritage and influence of religious singers to the present day. Of course, this situation differs slightly from Racy’s assertion in that he specifically wrote about vocalists learning their trade in the midst of Sufi singers, whereas the specific example highlighted in my research evidences this process with instrumentalists.

150 In addition to asserting that vocal apprenticeships with religious singers began to fade out of practice around the time of World War I, Racy also states that the process of young instrumentalists learning their craft from an accomplished senior musician, centring on the small takht25 ensemble, also faded out of practice at the same time (Racy

2003, p. 26). However, my fieldwork suggests that this is not the case in Egypt.

Learning environments guided by a master vocalist or instrumentalist are still an important learning context for many musicians. Young vocalist Ahmad El Haggar undertook a familial apprenticeship under his grandfather and father’s care, as did nāy player Mohammed Fouda, who learnt directly from his father. Of those musicians I interviewed who did not live in musical families as children, all identified important pedagogical relationships with senior accomplished musicians that facilitated their development, in such a way that replicated aspects of the apprentice-style arrangements.

This type of learning environment was fundamental to Mahmoud Bedir and Dr Alfred

Gamil, learning directly from within aforementioned violinist Abdo Dagher’s ensemble.

Bedir told me that his education in Arabic music started six years after he had begun formally learning the ‘cello, gaining his education in Arabic-style performance through attending regular lengthy rehearsals at Abdo Dagher’s apartment in Cairo within

Dagher’s small Arabic chamber orchestra.

During my time in Cairo I witnessed this ensemble rehearsing with musicians from a variety of backgrounds and with differing levels of experience, all benefiting from

Abdo Dagher’s wealth of practice-based knowledge. Dagher’s grandson was also by his side throughout these rehearsals, imbibing his grandfather’s musical style during lengthy rehearsals by ear. In addition to passing on his training with Sufi singers,

Dagher’s role as mentor is enriched by the time he spent performing in the most famous

25 As introduced in Chapter Two.

151 ensemble in Egyptian music history, accompanying the national singing icon Umm

Kulthūm as a violinist in her orchestra (Dagher 2000). Dagher now passes on this traditional training to his own ensemble.

Reminiscent of, and inspired by the time he has spent playing in Abdo Dagher’s orchestra, Dr Alfred Gamil now leads his own ensemble, Qithara, fostering the development of young musicians in a small ensemble, apprenticeship-style environment. Australian-Egyptian Joseph Tawadros has made numerous trips to Cairo, actively seeking to engage in these learning environments with master musicians such as Gamil, Esawi Dagher, and Abdo Dagher. In 2008 I was able to witness this process in context. While not formalised apprenticeships as such, these arrangements are described by Tawadros as vital to his education in traditional Egyptian music.

5.2.2 Have these apprenticeship-like relationships been replaced by institutionalised training?

Racy suggests that during the twentieth century, apprenticeship training gradually lost its appeal. He cites, as the main causes for these changes, a change in the interpersonal trust dynamics between masters and pupils, a growing lack of interest in the older repertoire known by the elder generation of musicians, and a rise in the conservatory mode of instruction (Racy 2003, p. 26) Yet, in Egypt apprenticeship learning situations have continued to be an important aspect in the development of many musicians, even though the introduction and market-place prominence of music education institutions have certainly contributed to a changing dynamic in the mode of transmission for practitioners of classical Egyptian music.

Rather than diminishing the importance of these apprentice-style learning opportunities, I argue that for those musicians who are able to see the enormous benefits gained from these opportunities, the proliferation of music education institutions has

152 actually served to increase the value that musicians place on these apprentice-style opportunities to sit and learn in situ from masters of Egyptian music. Indeed, many of the musicians I observed interacting with senior instrumentalists in apprentice-like relationships were concurrently studying at a formal institution such as those described by Racy. This is certainly the case for Dr Alfred Gamil and Mahmoud Bedir, who studied formally at the Institute and also informally with non-tenured musicians. Ahmad

El Haggar continues this practice as a current student of the Institute. Uniquely, Joseph

Tawadros was studying in an Australian university music department between his trips to Cairo. Why these musicians, who train in institutions, still place a high value on the benefit they experience from master-musicians in apprenticeship-style relationships will be demonstrated below.

Before doing so, as a point of summary, the research presented above suggests that in Cairo the tradition of apprenticeship-style relationships continued beyond World War

I, and continues in a modified manner in current practice. The situations as experienced by the Dagher brothers, Fouda, Bedir, Tawadros, and El Haggar attest to this fact.

Generations of vocalists learning in situ from master singers, and instrumentalists learning their skill from master instrumentalists in small ensembles such as the takht ensemble, represent a community of musicians passing on details of performance practice from generation to generation in an oral-aural cycle. My findings in Cairo confirm Thomas (2006, pp. 151–152), who recently researched current educational trends in Arabic music.

153 5.3 Reinvention and adaptation of past learning environments

5.3.1 Apprenticeship-like relationships are still essential

Thomas’ research attests to the first aspect of my findings in that she tempers the perceived trend asserted by Racy that suggested apprenticeships have faded out of prominence. Thomas stresses that whilst these apprenticeship situations have decreased both with regard to the regularity and intensity of instruction, the more traditional modes of transmission do still exist, albeit in new, more flexible and less formal settings

(Thomas 2006, pp. 148–149). Interestingly, my research suggests that these apprenticeship-like relationships in Cairo may not even have decreased to the extent suggested by Thomas, but rather, have simply become modified to fit in and around the changing educational setting in which musicians find themselves. Indeed, as I noticed from my time with musicians in Cairo, Thomas also noticed

relationships established in the institutions frequently extend beyond the walls of the institute and provide opportunities for participation in other learning environments, including informal sittings and ensemble rehearsal and performance (Thomas 2006, p. 149).

Racy argues that aspects of these relationships actually exist within institutions: ‘In some respects, conservatory training illustrates the traditional modalities of instruction’

(Racy 2003, p. 28). Furthermore, he continues: ‘Many conservatory teachers are practicing musicians who had taught apprentices before’ (Racy 2003, p. 28). While

Racy’s theory seems plausible, Thomas’ experiences suggest otherwise and my own research confirms this fact with regard to Cairo. The traditional modes of transmission that Racy asserts still exist within the formal training offered by institutions were not witnessed throughout my fieldwork in Egypt. I discovered that musicians partake in traditional modes of transmission outside the formalised institutional framework that

154 seemingly stifles these types of relationships due to the irrelevant and limited curriculum of the academies. These shortfalls will be detailed below.

Moreover, the notion asserted by Racy – that the teachers of modern-day academies are practicing musicians experienced in both performance and in the teaching of apprentices – was certainly not evident from the interviews I conducted in Egypt. This deficiency is one reason why the important apprenticeship-like relationships exist outside of The Higher Institute for Arabic Music in Cairo. Students and even one of the teachers that I interviewed highlighted the perceived poor quality of teaching and unfocussed curriculum at The Higher Institute for Arabic Music. Master musicians are rarely associated with an institution, often because they lack formal qualifications. As a result students only have access to qualified substitutes within the academy.

Ironically, qualified teachers were described as having theoretical knowledge but lacking practical experience. Some were even said to be limited in their ability to even play the instrument that they were teaching, even though they had the necessary qualification allowing them to teach the theory encapsulated by the curriculum contents.

Moreover, the content of the curriculum that is taught suggests very strong reasons, detailed subsequently, for why students seek out apprentice-like relationships outside of the institution’s walls.

5.3.2 Arabic musical content in institutionalised tertiary training

As a vocalist studying at The Higher Institute for Arabic Music, Ahmad El Haggar told me that much of what he was taught focussed on developing solid technique. Breathing and posture were discussed as lesson items that he viewed as the most important aspects of what he was learning. Interestingly, the manner in which he was instructed in these areas was through Western exercises and repertoire such as arias, lieder, and vocalise.

155 El Haggar has been taught to view these techniques as valid aspects of vocal training, and he sees them as the most important aspects of his training at the Institute, even though previous generations of Egyptian singers, including his own father and grandfather, did not learn to sing through these means.

These developments signal a changing focus towards vocal technique in the next generation of Egyptian singers and a separation of technique from stylistic context.

From my perspective as an Australian researcher familiar with the Western conservatorium model, I was surprised to find that so much of the curriculum focus was on Western-derived training. So little seemed to be designed to further students’ training as Arabic musicians, especially in the areas of traditional vocal and instrumental improvisation. This is not to suggest that Western-style vocal training detracts from their overall development as singers. However, this change in pedagogy would certainly cause changes in voice production and timbre, and without the experiences encountered outside the institutional parameters, their ability to develop an Egyptian style of singing would be substantially impeded.

Thomas also found a similar conceptual disparity in the curriculum studied by young musicians at other institutes in Jordan and Cairo, asking,

Why were students specializing in Arab music required to know how to write four-part harmony? Why was an ‘ud player required to know how to play Bach inventions on the piano? Would these skills actually serve them in their professional lives? (Thomas 2006, p. 145).

I was struck by the same incongruity and questioned El Haggar along these lines. His response was that he would not be able to learn to develop proper technique if he did not study in this Western style. This is perplexing, given that this curriculum is pedagogically unrelated to the learning environment of previous generations of Arabic singers, including his family members. El Haggar does profess a desire to sing in a style affected by Western music, admiring the popular and art music of the West, and he also

156 asserts that due to the heritage in which he was born he has received an adequate training in traditional music. However, this only explains his reasons for feeling comfortable studying in this environment. It does not account for why The Higher

Institute undertakes the training of singers and instrumentalists in this manner, especially those who have not benefited from the generous training that El Haggar’s heritage has provided him. This is a significant point, as the 14-year-old students who enter the Institute often do not have the same wealth of experience of El Haggar.

Further attesting to this anomaly, Thomas also discovered in the course of her research in Jordan and Cairo that there was no instruction regularly offered in Arabic- style violin performance. Questioning various administrators as to why this was the case, their answer was reminiscent of El Haggar’s response to me:

violin students who had been exposed to Arab music all their lives need only to develop a general level of technique on the violin by studying the Western curriculum and they would be able to transfer this technique to play Arab music (Thomas 2006, p. 145).

However, Thomas discovered that this was not the case, and only a small number of violinists were actually able to make this transition even though the main professional occupation awaiting Egyptian violin students was to perform in an orchestra performing

Arabic-style music (Thomas 2006, pp. 145–146). This incongruity is articulated in important technical differences that violinists encounter when performing in either tradition. From the Egyptian perspective, this relates to an ability to pitch quartertones on the violin fingerboard, achieving vocal-like portamenti by sliding between notes, and rendering ornaments appropriately, as well as performing with an idiosyncratic width and speed of vibrato. These differences necessitate a specific muscular development such that transferring between genres is not as effortless as implied by Thomas’ subject

(Thomas 2006, p. 145).

157 As well as the relatively extensive Western-styled instrumental and vocal instruction, students at major institutes such as The Higher Institute for Arabic Music study a variety of other subject areas derived from Western music education. Dr Alfred

Gamil described the various aspects of curriculum when I interviewed him in 2008. He is a particularly comprehensive source of information as he studied at the Institute for seven years from the age of 14 and has taught in various capacities since graduating around 30 years ago.

The curriculum studied in the first few years of early high school-aged study (14–

16) includes solfège, theory, and some classical Arabic song repertoire. The theory and solfège subjects are studied in Western formats and are transferred for use in learning and analysing a selection of Arabic music. Following initial training in these areas, students are required to specialise in voice, an instrument, or in composition studies.

Western-influenced curriculum and pedagogy is continued. At the time of his interview,

Gamil taught solfège at The Higher Institute for Arabic Music, and he demonstrated an impressive fluency when singing repertoire in solfège syllables, including a thorough command of translating Arabic repertoire into modified solfège.

Highlighting underlying problems with education policy at the Institute, Gamil recalls that he wanted to study violin and applied to do so when he was required to choose a specialty after initial training. However, at this time he was strongly encouraged to specialise in vocal studies by the dean of the Institute, who Gamil says had developed a liking for his voice. Gamil views the situation that unfolded with regret and cynicism, believing that the dean was looking for talented students to teach and therefore poached him in this fashion, discouraging his violin specialisation. Since graduating, Gamil has made his performance career as a self-taught multi-

158 instrumentalist, primarily as a violinist. He speaks in humble self-deprecating terms about his vocal ability.

Gamil was also quite critical of the general mode of teaching at the Institute, in light of the fact that he was forced to study something which he had little interest in undertaking and also the inadequate amount of Arabic material taught to students. He maintains that the material is quite often too easy and is homogenised for use in teaching all instruments. Ahmad El Haggar, a current student at the Institute, concurred with these sentiments. Gamil explained why this was the case, asserting that it is not appropriate to study Arabic song in a progressively structured manner. While Western music is easily organised according to a rising scale of difficulty, thereby providing students an opportunity to learn through increasingly difficult repertoire, there is not the same progression of difficulty in Arabic song, since the melodies of Arabic music generally focus on lyricism as opposed to technical difficultly. As such, the repertoire cannot sustain a progressive curriculum.

Rather than examining ways that the institution’s curriculum structure may be developed to allow traditional repertoire to be incorporated in a more appropriate fashion, at a tertiary level Western repertoire is chosen to develop the technique of students in ways that extend beyond the technical needs and the aesthetic qualities of the very music that students are attempting to master. Whilst Western music may offer them a wider variety of technical challenges, as is often claimed by those making use of it to teach, it seems illogical for an institution that trains musicians to perform in a society focused on Arabic music to emphasise Western repertoire and technique so intensely. It is difficult to understand how institutions can justify their existence given that an agenda other than one that strives to train Arabic musicians seems to be underpinning their educational philosophies. Perhaps there is an intention at work to

159 train musicians for a wider, more cosmopolitan or international musical field, though the balance seems to weigh heavier in favour of Western training. Educators seem content to leave the majority of education in Arabic music to activities conducted by students outside of the institution. This state of tension has existed since the 1932

Arabic Music Conference held in Cairo. The conclusions of the conference identified a series of goals that appear somewhat contradictory: ‘revival, preservation, reorganisation, and development’ (Thomas, 2006, p. 9). The inherent difficulty in reviving a musical tradition so that it may be preserved, whilst concurrently reorganising and developing it, has clearly been encountered in Egyptian institutions such as The Higher Institute for Arabic Music.

5.3.3 The tertiary curriculum and the (in)ability to improvise

I found a continuing disparity of logic through my interaction with another student of the Institute. After meeting this young violinist playing in a traditional takht chamber ensemble, I had reason to meet him again when my translator, Joseph Tawadros, visited him to try out a violin that the young violinist was selling. This student, fortunate enough to possess the ability to already work as a professional ensemble violinist, was so taken by Tawadros’ ability to improvise on the violin that he could not believe that

Tawadros (who specialises professionally on ‘ūd) was not a professional violinist. This young violinist professed an inability to improvise, even though he was a student of the

Institute. However, his inability to improvise is not surprising given that the institution he attends does not teach the skill at all (Thomas 2006, p. 147). The inadequacies of the tertiary-level curriculum in this music academy are certainly exposed by this interaction, and are further accentuated given the fact that Tawadros developed his skill to improvise geographically detached from Egypt, growing up in Sydney, Australia.

160 5.3.4 Why is improvisation not taught?

Before examining how musicians actually learn to improvise, I will examine reasons for why improvisation is not taught, beyond the Westernisation of tertiary curricula. The most prominent reason discovered by Thomas (2006, p. 187) was that educators view various aspects of improvisation as unteachable. In this regard I discovered the same opinion held by some of my interview subjects such as Esawi Dagher, who initially asserted that he could not teach improvisation.

I start with a discussion not of improvisation per se, but of style, the very thing most likely to be vitiated by an emphasis, for example, on European classical voice production and articulation. Thomas explains that when performing taqāsīm (or mawwāl) musicians require not only familiarity with the maqāmāt (and the associated modal schema26), but also ihsas, translated as a ‘feel for the idiom’ (2006, p. 187).

Esawi Dagher explained this to me as the ‘Eastern sound’ during an interview in 2008.

Developing this feel for the idiom depends largely ‘upon the minute details of melodic phrasing and ornamentation present in the performance’ (2006, p. 187), known as tafasil. By rendering the tafasil appropriately, musicians endow their improvisations with an accurate feel, ihsas. They stylise their improvisation as a taqsīm.

Thomas found that these details are considered unteachable. They are considered to be too complex and impossible to reduce to an abstracted form for transmission in lessons (2006, p. 187). It was suggested to Thomas that the complexity of the tafasil was due to the fact that they are originally derived from vocal practices such as Koranic recitation, later translated to instrumental performances as reproductions, variations, and new musical creations modelled on such prototypes (2006, p. 188). Consequently,

Because tafasil are so intricate and ephemeral, teachers consider them to be unteachable. They are not represented in notation, thus they can’t be pointed 26 As discussed at length in Chapter Two.

161 out visually to the student. They are fluid and improvisatory in nature … (Thomas 2006, p. 188).

The underlying opinion is that even if the tafasil could be isolated as motifs and compiled as a

vocabulary of musical motifs, something like the catalogues of ornaments compiled for French harpsichord music of the eighteenth century, it would be impossible to make rules for their recombination. The improvisatory nature both of the invention of tafasil and of their integration into melodic phrases and longer passages make it difficult to reduce this complex process of inventing taqāsīm into explicit rules or guidelines. It is this improvisatory aspect that leads some musicians to conclude that it is unteachable (Thomas 2006, p. 190).

Unfortunately, this opinion is influenced by a curriculum that focuses on notation and solfège, an emphasis that embeds a mode of learning unsuited to teaching these more sophisticated aspects of style, thereby excluding them from the pedagogical paradigm.

Whilst other aspects of taqsīm content such as mode identity, note choice, phrase length, melodic direction, and the inclusion of cadential figures may be amenable to being taught to students, as I will detail later in this chapter, the rendering of stylistically appropriate ornamentation and phrasing encapsulated in tafasil is that which leads many to deem improvisation unteachable. However, the question remains: how then do musicians develop the ability to improvise, especially those improvisers who have studied at these Egyptian music academies such as The Higher Institute for Arabic

Music in Cairo that do not teach the skill at all? Furthermore, the additional question remains: is the assertion that taqsīm cannot be taught accurate? In my fieldwork I discovered that some teachers do believe they can teach taqāsīm, whilst others additionally believe that taqsīm has a structure that allows musicians to format their improvisations to adhere to modal and stylistic guidelines. The details of these methodologies, and also an analysis of how musicians develop the ability to improvise will be investigated below with reference to my own fieldwork conducted in Cairo during 2008.

162 5.4 What do musicians do in order to develop the ability to perform taqāsīm and mawwāl?

As the preceding examination of music pedagogy in Egypt has established, an increasing number of musicians attend an institution such as The Higher Institute for

Arabic Music in Cairo to receive formal training and a qualification. However, it was concurrently demonstrated that the curriculum of this institution in particular played a surprisingly unimportant role in the development of a musician’s ability to play Arabic music, and, of particular relevance to this thesis, the institution played a near-negligible role in developing the ability to improvise. Thomas’ research (2006) found a similar trend in institutions throughout the greater Arabic music world.

Nevertheless, those musicians who are eager to learn how to improvise engage in a range of learning environments that help achieve this ambition. As the study of current music pedagogy in Cairo demonstrated, seeking out and engaging in informal apprenticeships continues to provide the main learning environment that supplements any formal training a musician may receive. Indeed, many of the older generation of musicians in Cairo received no formal training whatsoever, relying on familial connections and apprenticeship-like situations to learn Arabic music, and more specifically to develop the ability to perform taqāsīm. Whilst examining in more detail this type of learning environment, the following section will also expose several other pedagogical situations. Concurrently I will focus on demonstrating the interplay between the environments in which musicians learn, the mode of transmission active in these environments, and the details of exactly what they do when interacting in these learning situations.

The performers I interviewed during 2008 approach the process of improvised performance with varying levels of formality and structure. Violinist Esawi Dagher and

‘cellist Mahmoud Bedir declared simply that taqsīm has no rules; Bedir went as far as

163 suggesting that if it starts to have rules then it is worthless. The ‘ūd player Joseph

Tawadros also emphasised during interview the freedom that was inherent in taqāsīm. It became clear that most musicians were hesitant to summarise rules of performance for taqāsīm, particularly regarding structural parameters. The notion provoked immediate resistance when I introduced the concept.

However, there seemed to certainly be codified patterns of performance according to hierarchical prototypes. Traditional Egyptian improvisations adhere to stylistic conventions largely by rendering appropriate tafasil. Additionally, interview subject

Mohammed Antar suggested that there are still more specific melodic requirements concerning the treatment of maqāmāt that must be realised in the course of performance for a taqsīm to be considered stylistically correct. This theory is discussed in detail below. Whatever the case, the primary aspect that is beyond debate is that a taqsīm is an improvised melodic line 27 that is stylised by the inclusion of appropriate tafasil demonstrating ihsas (feel for the idiom). The manner in which musicians develop the ability to improvise with appropriate tafasil will be the next section’s initial focus.

5.4.1 Aural imitation of prototype melodies: vocal songs

At the time of my first interview with Esawi Dagher, he insisted that taqsīm was not rule-bound, and that he could not teach someone to improvise. Elaborating slightly on this point, Dagher suggested that if a musician wanted to learn to improvise, they needed only to listen to a vocalist reciting traditional religious material. In suggesting this approach, Dagher prescribed to musicians wishing to learn how to improvise in an

Egyptian style a form of imitative behaviour by means of aural copying. As discussed

27 Performances sometimes have rare moments of harmony on instruments that allow for this, such as the ‘ūd and qānūn. Taqāsīm are performed in a variety of contexts, either as a solo performance or accompanied by percussion, a pitched ostinato, a pedal drone, or a combination of these accompaniments.

164 previously, a large portion of Dagher’s in situ training was in this context, playing violin alongside vocalists performing religious material.

Reflecting upon Dagher’s initial assertion, that taqāsīm have no rules, it became apparent that Dagher felt that a rule was only a rule if it was explicitly formalised.

Nevertheless, by stating in response to my question that all someone needs to do is to listen to religious recitations (and in turn imitate them), this is in fact prescribing adherence to a set of unspoken, unwritten rules that the vocal performance is emblematic of. As the advice from Dagher shows, there are certainly prototypes that performers imitate and aim for during performances. This represents a type of rule- based performance practice, even though the rules are inherently implicit and absorbed osmotically in the manner prescribed by Dagher.

During our interview Dagher demonstrated a taqsīm in maqām hijāz. A parallel audio comparison of his performance with a recitation of the adhān (Islamic call to prayer) in maqām hijāz illustrates the manner in which religious vocal recitations and songs serve as aural blueprints for his instrumental taqāsīm. I have taken Dagher’s taqsīm and mixed this with a recording of an Egyptian recitation of the adhān. This is especially pertinent in Dagher’s case as the call to prayer from the local mosque is heard inside his apartment on a daily basis. The example begins with Dagher’s violin taqsīm, and after 18 seconds the adhān recitation commences. The vocal recording, recited by

Mohammad Omran, is taken from a collection of Islamic liturgies (Omran 2011). Play

Audio Example 1.28 A parallel comparison of the two recordings demonstrates the close stylistic similarity between the character of the vocal recitation and the expressive techniques employed by Dagher.

28 In providing this sound example I am neither regarding the adhān recitation as music, or attempting to create a piece of music by mixing the two sounds. I am only placing the two sounds, adhān and music, side-by-side, so that comparison might be made.

165 The timbre of Dagher’s performance shares a dense, raw quality with the vocal recitation. The ornamentation, swells, portamenti between pitches, and the style of vibrato that Dagher employs all have a clear vocal origin, as he asserted. Moreover, the overall trajectory of the performance is structured in a similar manner, with phrases gradually increasing in length and intensity of expression, characterised by the more vigorous inclusion of ornamentation, swells, and vibrato. Whilst the violin possesses a wider overall pitch range, the range of phrases in both recordings increases as the recording continues, with phrases that focus on higher pitches appearing as the recordings continue.

In addition to working with a number of religious vocalists, Dagher engaged in other periods of in situ osmotic and mimetic training throughout his development as a violinist. These engagements included performing with mizmār29 players and dancers in

Upper Egypt, the importance and effectiveness of this training evidenced by his being able to demonstrate to me some of the melodies he had learned under these circumstances during our interview, some 50 years later. Upon his arrival in Cairo in

1960 he was engaged to accompany dancers such as Zeinat Olwi, who performed as the entertainment at social functions such as weddings and the like. Zeinat Olwi was one of the leading belly dancers in Egypt during Dagher’s time in her troupe, and featured as a dancer in several films. Dagher also spent time working in other traditional orchestras.

In these situations Dagher was taught the music to be performed aurally, in a call- and-response fashion, mimicking a model of performance provided by vocalists or musicians. Interestingly, he did not mention spending time working with other violinists, nor did he describe engaging in a process in which he imitated other violinists. He did mention that when he was a young child, he paid his older brother to

29 The mizmār is a double reed conical shawm, an oboe-like woodwind instrument used extensively in the folk music of Upper Egypt.

166 teach him traditional melodies aurally.30 However, given the nature of takht ensembles and traditional orchestras, it is likely that violinists where among those musicians

Dagher imitated in order to learn the repertoire.

Specifically, the traditional melodic forms that Dagher absorbed aurally throughout these periods of professional engagement contained important musical elements that have been highlighted by Thomas as tafasil (2006, p. 187). Connecting Thomas’ theory to observations of Dagher’s practical experience, the effect of performing with adequate tafasil is described by Dagher as equivalent to achieving an ‘Eastern sound’, something that he identified as needing to be present in a musician’s taqāsīm. Identifying other prototype vocal melodies exhibiting tafasil, Dagher made mention of famous popular singers such as Umm Kulthūm. He asserted that their secular songs, like those of the religious vocalists whom Dagher worked with early in his career, represent important archetypes that encapsulate and represent the Eastern sound.

During our interview, as a way of demonstrating the importance of singers such as

Umm Kulthūm and the extent to which their singing embodied the inheritance of

Egyptian music, Dagher played the first verse of the Umm Kulthūm song, Sahran Le

Wahdi (‘Sleepless by Myself’). While playing this composition, his performance consisted of weaving between phrases played only on the violin and phrases where my translator joined Dagher, singing the words to the song. The apparent stylistic similarity between Dagher’s performance that is excerpted on Audio Example 1, and his performance on Audio Example 2 would have borne an even stronger resemblance had the singing not identified these transitions between the solo violin and duo phrases. In giving this example, Dagher was able to illustrate just how closely his improvised

30 The manner in which Esawi relayed this story regarding paying his own brother appeared to be a comment on the elder brother’s entrepreneurial nature, rather than on a generalised Egyptian practice.

167 phrases are cast in the same stylistic and modal framework that is embodied in Umm

Kulthūm’s famous melodies. Play Audio Example 2.

Throughout my time in Cairo, it became exceptionally clear to me the extent to which Umm Kulthūm and other vocal models figure prominently in the day-to-day life of Egyptians. Most taxi drivers I travelled with would either be playing songs by Umm

Kulthūm, those of other singers such as Mohammed Abdul Wehab, or vocal renderings of the Koran and other sacred recitations. Melodies also drift out of busy street-side cafés and from open apartment windows throughout bustling Cairo. Taxi drivers know the words; my translator would sing with taxi drivers as we travelled from one interview to the next, more often than not a song by Umm Kulthūm. Remarkably, her funeral in

1975 attracted over four million mourners, the streets of Cairo overflowing with her devotees. Today, over 35 years after her death, her musical influence is still strongly present in Egypt and throughout the Arab world. Given this aural saturation it is not surprising that Esawi Dagher learned by ear, and can still play, so many songs by Umm

Kulthūm.

5.4.2 From vocal song to instrumental composition

All of the musicians that I interviewed identified knowledge of compositions – such as those made famous by singers such as Kulthūm and Abdul Wehab – as important to their familiarity with Arabic style. ‘Cellist Mahmoud Bedir says that he was able to make the transition from years of Western classical playing to improvising in an

Egyptian (Arabic) style partly because of his pre-existing knowledge of traditional

Arabic song, which he absorbed after many years of listening and transferred to his instrument by ear. Dr Alfred Gamil also identified his inherited knowledge of Arabic song as the key to his perceived pre-existing aural knowledge of the different maqāmāt

168 and their melodic treatment, going as far as saying that ‘I have the maqāmāt already in my ear, I am Egyptian … since I was in my mother’s stomach’.

When making mention of traditional vocal music, it is important to point out that much of this repertoire contains lengthy instrumental introductions, interludes, and improvisations, and a traditional chamber ensemble accompanies the vocal soloist. As a result, whilst absorbing important nuances of content including maqāmāt (modal) functionality, details of ornamentation and other means of expression contained within the vocal melody, musicians also grasp details of style by listening to and emulating instrumental players. This contributes in a significant way to their ability to represent the stylistic parameters of the music in their own instrumental performances. Even for

Australian immigrant, Egyptian-born Joseph Tawadros, these traditional vocal songs and instrumental parts, derived from imported recordings and films, played a vital role in his development as an improviser. It was in this transported Egyptian environment within his inner-city Sydney home that Tawadros came into contact with the canon of traditional Egyptian music.

The process he engaged in epitomises the process of listening, absorbing osmotically, and eventually imitating past masters of the Egyptian taqsīm tradition.

Tawadros remembers first becoming interested in playing the ‘ūd after seeing films starring musicians such as Mohammed Abdul Wehab. Inspired to pursue Arabic music, he began a long-term yet informal period of intensive listening. Typically of his place and generation, Tawadros spent much time as a young boy playing computer games – atypically with the sound off – whilst listening to traditional Arabic music, including but not limited to that which his mother had made recordings of in Egypt.

This repetitive gaming process accompanied by frequent listening enabled

Tawadros to learn a plethora of traditional music, attested to by the fact that by the age

169 of 12 he had learnt the lyrics to songs that were written in a style of Arabic that was a far stricter, more formal form of Arabic than the colloquial version he spoke at home. In fact, Tawadros says that he did not even know the full meaning of the poetry of the songs that he had memorised, reinforcing the extent to which he had been thoroughly saturated with this music. This informal but intensive listening process is seen by

Tawadros as the reason he was able to play traditional songs quite easily once he began playing the ‘ūd. Retrospectively, he also views his knowledge of traditional vocal and instrumental music as the reason for his ability to compose with an intuitive sense of form and structure.

Furthermore, as his knowledge expanded through these frequent informal listening sessions, and once he had learnt the names of Arabic modes, Tawadros was able to recognise what mode was being employed in a given melody given only a brief listening. Joseph says that his brother, percussionist James Tawadros, is also able to recognise the mode of a given melody due to years spent listening to traditional music, a noteworthy feat given he does not play a melodic instrument. As Joseph Tawadros told me, learning this traditional repertoire allows musicians ‘to become familiar with the scale, to know how it is used’. Furthermore, not only did learning these traditional songs allow Tawadros to absorb a stylistically appropriate method of using the traditional maqāmāt melodically, both in his compositions and improvisations, but also to know ‘where they modulate to’.

Tawadros comments,

I think its just immersing yourself in the music; once you immerse yourself, then you start seeing similarities, and then you start seeing things, and then your brain starts picking up on it.

170 By listening, Tawadros also learnt how to ‘close a phrase’, that is, employ traditional cadential patterns known as qafla31 during a taqsīm performance. As he says, ‘You’ve got to know how to finish the mode … because the audience is expecting that finish so they that they can yell out [engaging vocally in the performance]’. Tawadros has continued this practice of informal regular listening sessions into adulthood. His car is littered with old cassettes from Egypt that provide a constant Egyptian soundtrack as he drives, perpetually inundating his aural environment. I, too, have benefitted from this osmotic learning environment. Travelling with him around Sydney for short periods of time over the last ten years, I would find myself humming and whistling the melodies of these songs well after I had left his company.

5.4.3 Instrumental compositions as prototypes

Unlike Esawi Dagher and Joseph Tawadros who are autodidacts,32 Dr Alfred Gamil comes from a generation that has experienced institutionalised education, in Gamil’s case since the age of 14. This is not to deny the importance of any autodidactic aspect of

Gamil’s musical development, but unlike Dagher and Tawadros, Gamil interacted in the formal educational environment of The Higher Institute for Arabic Music as both student and teacher for 37 years. Notwithstanding the fact that Dagher, Tawadros, and

Gamil have developed through substantially different experiences, Gamil reinforced much of the information I received from Dagher and Tawadros regarding how the ability to improvise is developed.

When speaking about the repertoire of compositions, Dagher’s emphasis, representative of his professional experience, was on vocal performance, both sacred and secular. Dagher clearly views the singing voice as that which carries important 31 Pronounced ‘aff-la’. 32 Even though Tawadros has studied Western music at a university in Australia, he has learned Egyptian music, by necessity, as an autodidact.

171 Egyptian musical genetics. Furthermore, the narrative portrayed by Dagher implies that the stylistic attributes and prototype phraseology that instrumentalists need to imbibe in their playing in order to perform Egyptian music adequately are derived largely from being immersed in traditional vocal repertoire. These songs embody the characteristic style, and they are important models for both composed and improvised music in this tradition.

On the other hand, whilst acknowledging the importance of vocal compositions, Dr

Alfred Gamil’s focus during our interview was on instrumental compositions; Dagher had only introduced these briefly in mentioning a repertoire of miniature forms known as dūlāb (plural: dawālīb). Gamil provided more detail, describing a dūlāb as ‘a very small form composed from one idea or two maximum’. That is, a short melody that encapsulates the essential sound of a maqām.

The dawālīb are employed as a way of introducing the melodic characteristics of maqāmāt: the stylised manner in which the pitches of a maqām should be realised melodically. Gamil stipulated that there are dawālīb in every maqām, the composers of which are said to be unknown. He also confirmed that musicians throughout much of the Middle East know the most famous dawālīb, although some rare dawālīb are said to have disappeared from common knowledge, as have related rare maqāmāt. In addition to teaching musicians the pitch characteristics and melodic potential of maqāmāt, learning dawālīb also guides musicians and vocalists in the art of realising maqāmāt during improvised performances.

This is clearly demonstrated in an example that Gamil gave: vocalists often begin a performance by singing a dūlāb before launching into a vocal improvisation. This is so that, as Gamil summarised, the vocalist is able ‘to have in the ear the maqām’. Touma describes a similar process in the takht ensemble whereby a dūlāb is played by the

172 ensemble as an introduction to a vocal piece (1996, p. 106). Touma also emphasises that dawālīb are based on a maqām row, and are also used as practice pieces by instrumentalists (1996, p. 106). Learning and practising the dūlāb repertoire introduces vocalists and instrumentalists to some of the melodic potential and characteristics of maqāmāt. In turn, as experience is gained in multiple dawālīb, musicians develop the ability to render some of the characteristics of maqāmāt correctly when they eventually come to performing a taqāsīm, however simplistic their early performances may be.

From this developmental stage, Gamil asserted that when learning maqāmāt, students then spend the most time working on instrumental compositions which extend their technique beyond that encapsulated in the dūlāb, an aspect of pedagogy also reinforced by Marcus (2000, p. 33). The pieces to which Gamil refers are longer compositions categorised traditionally in forms such as the samai, a genre that has four sections separated by a refrain, and the longa, a spirited dance in duple time. These instrumental forms respectively introduce and conclude the strophic muwashshah, a vocal genre. However, instrumental compositions such as longa and samai are commonly learnt as separate entities during practice. Pieces learnt by instrumentalists are also derived from instrumental introductions to popular songs such as those sung by

Umm Kulthūm.

Like the miniature dūlāb, these longer compositions expose musicians to the melodic characteristics of each maqām, but in far more detail than allowed by the length and simplicity of a dūlāb. The focus on learning these compositions, organised into genres by common characteristics, evidences an increasing emphasis on learning structure, both on a small scale in terms of phrase construction, and on a larger scale, in terms of overall melodic trajectory and song arrangement. The notion of there being ‘no rules’ appears to be increasingly inaccurate in practice. The structural uniformity

173 apparent between songs of the one genre, and the manner in which these disparate genres combine to form an overall Egyptian music identity indicate the operation of rules, be they implicit or explicit.

Moreover, Gamil suggested that the more advanced compositions are likely to modulate between related maqāmāt and in doing so explore methods of modal modulation away from and back to the original maqām. During an interview he demonstrated how modulation practices are encapsulated by compositions, playing a piece whilst announcing the modulation points as they took place, thereby illustrating the educational role pieces such as these play in developing a working knowledge of these processes. Gamil asserted that most pieces have modulations and that learning this repertoire gives students a functional knowledge of modulation syntax that can be transferred to taqāsīm, a clear example of the operation of an implicit rule.

Tawadros also emphasised that by practising and performing these compositions, students advance their instrumental technique and gain a procedural aural knowledge of stylistic characteristics, maqāmāt, and patterns of modulation. As musicians become increasingly fluent in rendering maqāmāt correctly and understanding procedures of modulation, the procedures and style absorbed by learning these instrumental compositions is eventually translated to improvising their own stylistically appropriate melodies.

Whilst acknowledging the role of traditional compositions, it is also important to emphasise that there are aspects specific to traditional taqāsīm that can only be absorbed by learning the improvisations of other musicians. Ahmad El Haggar has engaged in the process of learning the improvisations of other musicians from an early age. During one of our interviews he demonstrated this knowledge by performing a replication of one of his grandfather’s lengthy vocal improvisations, a mawwāl. This originally improvised

174 performance had been recorded and later memorised by his grandfather who then taught it to El Haggar by singing each phrase. When performing his own memorisation/version?? of the mawwāl El Haggar asserted that he was not performing it exactly as his grandfather had improvised it, given that some of the details (tafasil) were near impossible to recreate as exact replicas. Instead, his performance was as close a copy as he could achieve, rendered after many hours of singing it with his grandfather as a young boy. By undertaking this mimetic activity he has been able to absorb many of the musical layers in his grandfather’s improvisation, including traditional vocal timbre, detailed ornamentation procedures, melismae, expressive vocal techniques such as swells, traditional vibrato practice, appropriate realisation of maqām, traditional phrasing elements, and cadential figures (qafla). All of my interview subjects undertook similar mimetic activities and these details are outlined below.

5.4.4 Autodidactic preparatory strategies stemming from personalised conceptions of taqsīm and mawwāl performance

Extending this examination beyond the role played by learning prototypical compositions and improvisations, I will now examine the ways in which personalised conceptualisations of taqsīm performance give rise to autodidactic preparatory strategies, while detailing how these two elements interact to give rise to unique performances. Comparing the individual case studies of three diverse interview subjects will form this analysis: Dr Alfred Gamil, Mohammed Antar, and Joseph Tawadros.

Dr Alfred Gamil

Dr Alfred Gamil’s more formalised educational background led me to expect him to make more conclusive statements regarding whether or not taqāsīm were rule-bound, but his initial reaction was similar to Esawi Dagher’s and emphasised the perceived

175 inherent freedom of taqāsīm. Yet, upon deeper contemplation, he was, unlike Dagher, able to give a detailed portrayal of the improvisatory process and articulate more complex musical concepts employed during taqsīm performance, such as modal modulation.

Regarding taqāsīm, Gamil broached the topic as something that is ‘a little difficult’, not attempted in his first or second year of playing (aged 14 and 15). He identified his approach to improvisation as being based on a constantly expanding collection of phrases that form a personal improvisational vocabulary or phraseology, developed over many years. Constantly listening to and imitating other instrumentalists and vocalists have allowed him to develop the phrases that make up this collection. From this material – taken from the radio, recordings and even live performances – individual phrases from compositions and improvisations are memorised. During this process he engages in what he describes as a subconscious process of modification and elaboration, to the point that the phrases end up being original melodies in his personal phraseology.

He also described transposing derived phrases into new tonics, and changing the tonality of the phrase completely by replacing the original maqām with another.

Gamil said that when he improvised, he was not ‘improvising completely’, but often referring to his flexible improvisation vocabulary in new contexts. However, he was also quick to point out that he certainly was not performing a pre-fashioned composition, but something between the two points on this continuum. In this way, when Gamil improvises he is guided aurally to combine phrases that are familiar to him, derived from his personal phraseology, and he performs them in new contexts, attached to fresh non-derivative material or in other modified forms. As such, the taqsīm thus performed is a novel musical artifact.

176 The context from which Gamil sourced phrases for practice time processing also extended to informal situations. One such network centred on a street in downtown

Cairo named Mohamed Ali Street. This café-lined boulevard, now a bustling furniture retail strip, is still occupied by the cafés and music stores that Gamil spoke of, although the black suit, white shirt-clad musicians are no longer as prominent a fixture as they were in the past. Gamil detailed a social classification system whereby musicians of different standards and reputations sat at certain cafés, networking and securing work with peers of a similar musical ability. This network of cliques, demarcated in the oblique and often unspoken manner familiar to the organisation of other social sets, represented a musical hierarchy in Cairo’s music scene. Upward mobility and the work this secured were desirable outcomes of this social interaction.

Here Gamil met musicians who were experienced professionals, learning phrases from their improvisations, absorbing maqām theory, learning compositions, and generally benefiting osmotically from the experience. Marcus similarly describes musicians learning how to render maqāmāt in their own improvisations by engaging with the taqāsīm of other musicians

from recordings or by listening to and watching fellow musicians in performance. After listening to such a session, young musicians might come away with one or two new ideas that they will incorporate into their own improvisations (1998–2002a, p. 34).

Through the informal street-side café network, Gamil engaged in a learning experience that allowed his improvisation phraseology to grow within traditional parameters.

I will recount a first-hand experience of one such sidewalk session in Cairo, where I was fortunate enough to engage in a similar learning situation. Following a late-night interview, I stopped with my translator at a café not far from Mohammed Ali Street.

Accompanying us was Yacoub Habib, the now-retired Egyptian trumpet player who had been facilitating the interview earlier that night. As we sat at the café Habib happened

177 upon a table of his colleagues, and we were immediately invited to join in conversation at their table. After seeing my saxophone, one of the musicians at our table identified himself as a fellow saxophonist, and over the course of the evening I sat with him discussing how Egyptian saxophonists manoeuvred the quartertone aspects of traditional maqāmāt that do not come easily to fixed-scale instruments such as the saxophone. He demonstrated a number of alternate fingerings, and it was fascinating to discover that the Western methods I had started to use were in fact the very same that my Egyptian counterparts employed. As we sat, I observed him playing improvised phrases, and while he demonstrated various ornaments and other expressive techniques, other musicians arrived to join in the network around us. Interestingly, in the course of this interaction I discovered that Egyptian saxophonists use reeds that are softer than my regular choice. This became apparent when the saxophonist I met experienced some difficultly producing a sound on my set-up. He explained that softer reeds allow him to produce vocal-like glissandi between notes, whilst also aiding in the fine-tuning of quartertones via embouchure modifications.

Exposing other more informal modes of learning, Dr Alfred Gamil’s self-diagnosed

‘addiction’ to practice (seen in good humour by him as a by-product of his addictive personality) motivates him to host jam sessions at his apartment in Faisal, Cairo, that frequently last until the early hours of the morning. I was able to attend some of these jam sessions, witnessing Gamil interact, and interacting myself in this informal process in which musicians share information and experience new approaches to the challenges inherent in traditional improvisation. After welcoming Joseph Tawadros and me to his home, offering refreshments upon our arrival, our discussions about music led Gamil to play examples. Before long he invited us to join him and we began improvising together, trading phrases in a triangular musical interchange, one soloing while the other

178 two accompanied by playing an ostinato or sustained pedal note. The three-way jam session would pause when he felt the need to elaborate verbally regarding an aspect of his performance, and then it would resume. This process was repeated whenever we visited Gamil, and Tawadros in particular has visited him on numerous trips to Egypt.

Gamil described these jam sessions as a vital facet of his musical education, and he humbly suggests that he learns a lot from these experiences, even though many of his guests are students from The Higher Institute for Arabic Music. They come to imitate and learn from him in this more informal setting.

Regarding the practice of modal modulation during a taqāsīm, Gamil detailed a certain level of premeditation as it applied to modulating between modes. Using the metaphor of fashioning keys that would open doors to other maqāmāt, he described the necessity of developing a vocabulary of methods that would facilitate the fluent modulation between modes during a taqāsīm. He also emphasised that it was important for performers to have a variety of possibilities (keys) at their disposal so that improvisation in one maqām could move from the original maqām (through a doorway) to a variety of new maqām contexts. Gamil also used the metaphor of a kitchen, full of ingredients and varied manners of combination (recipes), to describe the process. He has spent many hours of practice time experimenting with and developing new methods of modulation between maqāmāt, expanding his own repertoire of possibilities.

Gamil emphasised that modulations in taqāsīm have no rules, but rather, a musician should be guided only by their ear and should never offend an audience’s ear by undertaking a badly conceived modulation. Of course this is a guiding principle, and the practice of modulation would certainly be viewed as rule-bound if the details of what exactly constituted offending an audience’s ear were elaborated upon and formalised in

179 a clearly articulated manner. However, other than this concern, he professed to having no restrictions when he modulates.

In defending the use of the ear as a guiding principle in making modulation choices,

Gamil insisted that the rules of Western music theory were actually no different. Even though the rules of Western music have been written down, their original conception was based on the aesthetic desires of certain generations, that is, the ear and tastes of past generations of Western musicians. In doing so, Gamil correctly pointed out that the only difference between the two musical traditions is that these rules have been written in the West. The fact remains that the acceptability or ‘offensiveness’ of modulating from one maqām to another is not a subjective evaluation, but rather the execution must take into account traditional practices, well known by audiences and musicians alike.

Implicit as they are, these practices are governed by aural rules of performance as much as Western music is governed by written precepts.

Some key points are exposed by this discussion. The seemingly contradictory status of Gamil’s conceptualisation, also discovered in information received from other

Egyptians who deny that their musical practice is rule-bound, reinforces that musicians from oral traditions, rightly or wrongly, conceptualise a rule formalised in writing as a far more restrictive imposition on musical practice than their own oral-aural conventions of performance.

Furthermore, I suggest that by considering their own parameters of performance enigmatic, an eminence is correspondingly apportioned to enlightened performers, and their performances are valued as such. I propose that this value is accentuated because the oral-aural nature of rule transmission and reception enforces a responsibility on musicians to determine for themselves what the parameters of performance are. I assert that this autodidactic behaviour carries with it a ‘weight of genius’ that is highly

180 esteemed. By maintaining that access to the parameters of performance are only by way of the ‘ear’, this constraint acts as an intellectual gate, granting those who attain the knowledge by way of their own aural endeavours an artistic status that henceforth endows their performances with a corresponding prestige.

Returning to Gamil’s narrative of the improvisatory process, always humble, he is not afraid to admit that he does not always improvise as originally as he might like, repeating himself occasionally during performances. As he says, he is ‘the same person’ whenever he performs, but his aim is always to introduce a level of originality and innovation. He asserted that he does think about what he is playing throughout an improvised performance, particularly regarding the maqām of the composition that is serving as the vehicle of departure. Like Tawadros, Gamil also premeditates where and when he will play a qafla, a short concluding phrase that serves an important cadential function, not only because it aids in the structuring of a taqsīm, but also because it includes the audience in the performance by providing a signpost at which they can interact with the performer. By exclaiming ‘Allah’ at a qafla point, the audience indicates that they appreciate a performer has performed an appropriate cadential phrase, particularly due to the role the qafla may play in identifying a modulation, away from or back to the original maqām tonal centre.

Eliciting an audience’s reaction indicates a successful performance, and Gamil emphasised that he considers any performance a failure that does not evoke this reaction throughout the taqāsīm. Gamil takes time in the course of performing a taqsīm to analyse the effectiveness of a qafla, and he will restate it more prominently if he has not received the desired reaction from the audience that he expects of his playing. Always striving to develop his abilities, he spends part of his regular practice sessions inventing qafla in various maqām contexts, revising those that are part of his existing functional

181 vocabulary, whilst also learning the qafla of other musicians in the aural-imitative manner employed throughout his autodidactic study of Egyptian music.

Mohammed Antar

Nāy player Mohammed Antar told me that he operates under a far more structured understanding of taqsīm than the other musicians I interviewed. Stemming from conflicting opinions regarding details of Arabic music theory, this disparity in opinion is central to my investigation because it highlights what the taqsīm performance practice is in Egyptian music today, and what musicians do in order to develop the ability to improvise in this tradition. Furthermore, the preceding information derived from Dr

Alfred Gamil, whose experiences neatly encapsulate and elaborate upon those obtained from other respondents, provides a useful model of comparison with Antar’s perspective.

Antar is the only musician I discovered throughout my research who affirmed an understanding of the now apparently obsolete theory concerning maqāmāt outlined in

Scott Marcus’ research (1987, 1989), as discussed in Chapter Two’s analysis of Arabic music theory. The theory advocated by Antar is found in the Mashāqah treatise33, and it defines each maqām by describing various pitches, the order that they should be played in, and how they should be approached: a melodic pathway. Opinion holds that this definitional method is now unknown and unused in Egypt, falling out of favour with musicians sometime soon after the treatise was formalised in 1840. It was certainly put aside after the 1932 Arabic Music conference in Cairo, at which time maqāmāt were represented as ascending and descending octave-bound scales using Western notation

(Marcus, 1987, p. 37). Antar, who had an international musical education, suggests that

33 Discussed in Chapter Two.

182 this phenomenon only occurred in Egypt, asserting that musicians throughout much of the Arab world know the maqām sequences.

Antar claims that this methodology is the only correct way to define a maqām theoretically, as opposed to the perceived inferior method that attempts to accomplish the same end by outlining an octave-bound ascending and descending scale. Antar describes the worded definitions found in the Mashāqah treatise as maqām sequences:

‘Something like a piece, a small piece, one sentence, one phrase, and you have to stick to this sequence’. While being interviewed, Antar demonstrated some of these maqām sequences, singing them during our first interview and playing them on a nāy in our subsequent time together. Commonly, before launching into a demonstration of a maqām, Antar would announce, ‘This is maqām (bayyāti)’, and he would then proceed to sing a melodic line. From this phrase Antar suggests that one can ascertain the identifying notes of the maqām and the manner in which these should be combined melodically. Antar also asserts that a taqsīm performed according to traditional style is informed by knowledge of the theory’s prescribed melodic pathways. A performer improvising without this knowledge is simply not performing a traditional taqsīm in

Antar’s opinion, but rather improvising in an idiosyncratic style.

His knowledge of this theory, established when studying Turkish nāy playing, was furthered as a result of his interaction with Lebanese musician and theorist Nidaa Abou

Mrad, who had made a series of recordings with the assistance of Antar’s brother.

Available on compact disc (Abou Mrad 2005), the first volume released presents 47 maqām as model-melodies,34 derived and interpreted from the Mashāqah treatise (from a total of 95), followed by short demonstration taqāsīm. The recording is designed to inform listeners by demonstrating how the maqām descriptions should be realised as

34 Labelled as ‘model-melodies’ in the liner notes of Abou Mrad (2005), audio CD.

183 sound, or as Abou Mrad states: ‘to concretise the listening process of this model repertoire’ (Abou Mrad 2005, audio CD liner notes, p. 5). The recording also demonstrates, albeit briefly, how these melodic scaffolds may be elaborated upon in improvised performance. Antar is a blind musician, so he has learnt these maqām model-melodies aurally, via examples played to him by sighted musicians who have read the Mashāqah treatise and translated the descriptions onto their instruments. In his teaching, he passes this knowledge onto his students35 strictly through call and response imitation.

The Mashāqah treatise describes a model melody for 95 maqāmāt: pitch signposts that are to be passed through in a certain order for varying lengths of time. For example, as described in the recording’s liner notes, the formula for bayyātī al-husaynī is prescribed thus:

Insist on the degree husaynī, slightly touch nīm’ ajam, play husaynī, before insisting on nawā, and jhārkāh before insisting on sīkāh, then nawā, husaynī, followed by a descending course of notes to dūkāh (Abou Mrad 2005, audio CD liner notes, p. 6).

This prescription for bayyātī al-husaynī is realised as a notated score in the liner notes, reproduced in Figure 4. For comparative purposes, Figure 5 illustrates how maqām husaynī36 is represented in modern theory. The Mashāqah model melody presents bayyātī al-husaynī starting on the 5th degree and descending to tonic, while the modernised representation portrays the maqām as an octave-bound scale.

35 His students are foreigners or informal local students, as Antar does not possess the necessary qualifications to teach in an official capacity in an institution, a problem facing many musicians. The irony is noted in the context of one of my respondents telling me that teachers who have these qualifications often lack professional performance experience. 36 Maqām husaynī is also known as bayyātī al-husaynī, given it is part of the bayyātī family.

184 Figure 4: Score realisation of the worded description of bayyātī al-husaynī (Abou Mrad 2005, audio CD liner notes, p. 6)

Figure 5: Maqām husaynī (bayyātī al-husaynī) as represented in modern theory

All 95 worded descriptions were transcribed into score notations before making the recording (Abou Mrad 2005, audio CD liner notes, p. 5), although except for the example, Figure 4, the remaining model-melodies were only published as audio. These performances are complete with undescribed details such as ornamentation, specific quartertone intonation, and vibrato.37 The decision to not publish the model-melodies as notated scores is perhaps indicative of the aural tradition in which they operate and therefore faithful to the traditional mode of transmission. However, as an artifact, concretised as such and fixed in time, the recording portrays these model-melodies that represent the worded maqām sequences as fixed and prescriptive structures, even though the manner of interpretation is inherently subjective given the imprecise nature of the instructions.

Although they are projected as fixed structures on the recording, Antar emphasises that the model-melodies are in fact not fixed. Each melodic phrase exemplified on the recording is simply one manifestation of a specific maqām and hypothetically two melodies that appear to be different can be identified as being representative of the same

37 These other details are not mentioned in the Mashāqah treatise description of how to realise the maqām sequences.

185 maqām as long as they offer an interpretation that addresses the flexible instructions.

The individual pitches that are selected for use in the model-melody, as well as the order and manner in which these are combined, align a given model-melody with a maqām. Whilst this is prescriptive, it is not as prescriptive as a fixed composition. Nor is it as imprecise in describing the melodic movement, and limited in the information and guidance offered, as an octave-bound scale. The demonstration recordings should be viewed in this positive light.

It was extremely satisfying to discover modern-day knowledge of this theory, let alone being able to interview an Egyptian musician who professed a functional improvisation methodology based on Mashāqah’s nineteenth-century treatise. However, over the course of my research time in Cairo, it became obvious that such knowledge, and perhaps more importantly acceptance of the theory was certainly not widespread.

Use of the theory in practice was limited. Marcus’ conclusion (1987, p. 37) – that the theory had been superseded soon after its formalisation as a published work – rang true.

Indeed, Antar’s convictions had led me to expect such a finding given that he, too, declared Mashāqah’s ideas had been lost in Egypt.

While I remained hopeful to find other likeminded musicians who would share

Antar’s belief in this ostensibly past moment of Egyptian music theory and taqsīm performance practice, the ideas advocated by Antar were met with considerable opposition when raised in subsequent interviews. It became clear that the theory is either unknown, or if known, as in the case of Dr Alfred Gamil, it is considered irrelevant to modern-day taqsīm performance practice in Egypt. As an example of the conflict that arose during an interview, Antar even went as far as to say that Egyptian-

Australian ‘ūd player Joseph Tawadros (who was present at these interviews as my

186 translator), whilst being an excellent performer, was not performing traditional taqāsīm when he improvised.

Tawadros, not surprisingly, took exception to this claim on several grounds. He argued that even if melodic sequences were perceivable in the style of taqāsīm performances played by musicians of the past, as suggested by Mashāqah’s descriptions, Tawadros asserted that musicians do not know them or reference them today when improvising. Moreover, he maintained that the Egyptian taqsīm tradition has evolved beyond defining maqāmāt, and therefore rules of taqāsīm, in the manner recorded by Mashāqah. Returning to an adherence of these sequences is an interesting concept, but Tawadros feels that doing so is not representative of the current state of

Egyptian taqsīm performance. Tawadros was certainly motivated to defend his performance style given Antar’s initial claim, though the logic of his rebuttable rings true and was supported by other musicians.

During another interview, I discussed Antar’s ideas with Dr Alfred Gamil. While recognising that approaches such as Antar’s do exist, he felt, like Tawadros, that these sequences did not characterise taqsīm performance practice in Egypt, especially when it came to learning to improvise. For both Gamil and Tawadros, taqsīm performance is held to be a far freer practice than an adherence to prescriptive maqām sequences such as those championed by Antar would permit. During concerts, Tawadros often introduces taqāsīm by giving a brief definition to the audience. The principle elements of his definition are that he will: improvise in a particular mode (described in the way that a Western scale is defined – in an ascending and descending octave-bound fashion); modulate to wherever he feels like modulating to; and return to the initial modal area at the end of his improvisation. While simplified for the purposes of educating his largely non-Arab audiences, the melodic freedom inherent to Tawadros’ definition is clear.

187 Attempting to temper these disparate positions, during another interview with Antar

I asked him to discuss these issues with Tawadros at greater length. Using another musician as an example, one who was also one of my interview subjects, Antar suggested that violinist Esawi Dagher played in a style that reflected the sequential style of past generations. Whilst recognising Dagher would not have ‘learnt’ the sequences as such, he suggested that the sequences were encoded in Dagher’s style, inherited via imitation during in situ training with Sufis.

It was certainly clear from my time with Esawi Dagher that he firmly believes that taqāsīm have no rules; however, he did recommend listening to and copying singers and religious recitations. Indeed, whilst teaching me certain maqāmāt, Dagher made amendments to my imitations of his phrases. I will discuss this lesson in more detail at the end of this chapter, but in summary he taught me to perform sub-phrases, ornaments and quartertones according to the dictates of an unidentified prototype that he has in his mind, absorbed osmotically over the course of his career, and represented intuitively in his performance style. Importantly, my experiences with Esawi Dagher reinforced the accuracy of those sources discussed at length in Chapter Two that assert that each maqām has special melodic features (Marcus 1993, p. 108), and that

a maqām is more of a schema than simply a scale. It is a process of melodic movement, and an operational protocol of improvisation according to the models and forms of melodic and rhythmic organization. It is a complex system that implies the illustration of its structures in modal melodic development … (Ayari and McAdams 2003, p. 164).

Dagher’s pedagogic behaviour was reminiscent of the melodic movement emphasised by the Mashāqah treatise and the research of D’Erlanger, detailed by Marcus (1987,

1989). That is, the historical context summarised in Chapter Two was certainly reflected in my experiences with Dagher.

188 Perhaps Antar’s claim is not as dubious as Tawadros and Gamil maintain, but determining this would require a study beyond the scope of this thesis. A large-scale analysis of taqāsīm performances given by both older and more recent generations of performers would be needed to determine if the Mashāqah maqām conceptualisation is subconsciously present in their playing. It would also reveal whether or not Antar’s inferred conclusion is true: that there is a major difference between the taqāsīm of these older musicians and the younger generation, as represented by Tawadros. In any case, while the conclusions of such a study would be fascinating, this thesis is an examination of how Egyptian musicians learn to improvise. Whilst Antar has studied the Mashāqah maqām sequences, he is alone in the greater context of my research. No other musician I interviewed professed to conceptualise maqāmāt or perform taqāsīm in a manner that implied knowledge of those theories espoused by Antar.

Post-1932, after which time Arabic music theory was formalised in the proceedings of the Arabic Music Conference, there was certainly no mention of these melodic sequences; maqāmāt were defined as octave-bound scales (Marcus 1987, p. 39). It is therefore not surprising that Tawadros and other performers representative of a younger generation profess no knowledge of these melodic sequences. They perform taqāsīm with the concept of melodic freedom clearly foremost in their minds. Of course, depending upon what players they have encountered and mimicked in the course of their careers, their melodic structures and style could echo this past practice of traditional Arabic music potentially formed on the scaffolds of these melodic sequences, but it is clear that they do not learn to improvise with any guiding notion of these rigid melodic pathways, even though they may do so intuitively. This conclusion is significant for the purposes of this thesis. Antar and other advocates of the Mashāqah treatise may indeed be adhering to a more structured, formalised approach to taqsīm

189 performance practice, but there is no doubt from the totality of my research that in

Egyptian music today these maqām sequences are not known, let alone learnt in the course of learning how to improvise. Antar is a minority neo-traditionalist in this regard, pursuing an anachronism.

Ironically though, the freedom that Tawadros and Gamil advocate is that which grants Antar the freedom to learn and to improvise in whichever manner he chooses, even if he is reviving an out-dated practice in the process. Importantly, criticism of

Antar is levelled at his campaign for stylistic correctness, not at the neo-traditionalism of his personal musical style.

Joseph Tawadros

Listening to recordings and imitating the improvisations of past masters is a point of commonality between all of my interview subjects, including Australian-Egyptian ‘ūd player Joseph Tawadros. A detailed investigation of his development as a musician illuminates central themes regarding what Egyptian musicians do to develop the ability to improvise.

As introduced above, Tawadros’ childhood aural environment was a transplanted snapshot of Cairo, as thorough an immersion as possible in Egyptian music and film.

Once he had established a strong affection for traditional music, he began researching theoretical concepts such as maqāmāt online, on websites such as Maqam World. He purchased his first ‘ūd from a family friend, percussionist Tarek Sawires, a seller of imported Egyptian instruments and craft wares at local markets in the inner-west of

Sydney. Tawadros then began experimenting, learning traditional melodies on the ‘ūd, putting into practice rudimentary guitar technique he had taught himself under his older brother’s guidance. At first, Tawadros’ attempts were extremely ad hoc, playing on a

190 three-string ‘ūd that he attempted to re-string with a complete set of strings sent to him from Egypt, only to find that his attempts resulted in several broken strings.

Eventually Tawadros began taking ‘ūd lessons from Sydney-based ‘ūd player and pianist, Mohamed Youssef, also of Egyptian background. Travelling every Sunday by train to Youssef’s home in Lakemba, Sydney, Tawadros quickly developed a passion for ‘ūd playing, developing under Youssef’s patient and encouraging manner. Taking lengthy weekly lessons, frequently lasting for over three hours, Tawadros remembers quickly picking up how to hold the ‘ūd and plectrum by observing Youssef, who did not dictate technique explicitly. However, Tawadros took longer to master the down-up plucking technique used by ‘ūd players. He remembers having difficulty doing this due to the time he had spent playing guitar, causing him to initially play the ‘ūd using a modified guitar technique of down-down strokes. To this day Tawadros feels his technique is distinctive when compared to other ‘ūd players: reluctant to use the plectrum in an upward fashion, he favours using multiple downward strokes. Seeing the positive outcomes resulting from his abnormal technique tendencies, he notes this as a reason for his ability to play in odd-numbered meters, facilitated by multiple downward strokes in combination with the usual down-up combination.38

Even though Tawadros began this more formal period of study, he did so while being motivated by a wave of natural enthusiasm, never feeling like he had to practice, but rather always answering the desire to play his beloved instrument at irregular moments during the day. A time he found particularly pleasant to play was before he went to school in the morning, noting that often he would get so involved in discovering a new technique or phrase that he did not want to leave to catch the school bus. Always motivated by passion-driven inclination rather than forced discipline, Tawadros never

38 For example, playing in three is facilitated by plucking down-up-down/down-up-down, and in five by plucking down-up-down-down-up/down-up-down-down-up.

191 set aside formal practice time as a high school student, and still does not do so today. In fact, during his time at high school he took up the elective study of music so that he could tell his parents he had been studying for school during those times that he was regularly practising at home.

Tawadros’ desire to continually listen to new music was enabled by his lessons with Youssef. He introduced Tawadros to recordings of a range of ‘ūd players with whom he probably would not have come into contact in the more traditional and insular musical environment of Egypt. Listening to players such as the cosmopolitan Tunisian

Anouar Brahem and Palestinian Simon Shaheen expanded Tawadros’ concept of what could be performed on the ‘ūd beyond traditional parameters. This education was accompanied by an introduction into jazz, rock, and music of other traditional cultures.

His lessons with Youssef were always taught aurally and never structured, with much of the lesson time being dedicated to lengthy attempts at improvising, labelled as jamming by Tawadros. Through necessity he gradually developed the ability to quickly and easily memorise the new material that Youssef introduced him to, and correspondingly, he also developed a very refined sense of relative pitch. Today he can learn a new melody, depending on its complexity, after only a few hearings, and his ability to learn aurally far exceeds his ability to read music, even though he is musically literate, graduating from the University of New South Wales with a Bachelor of Music

(Honours) in April 2006.

During his lessons with Youssef, advice and amendments were limited, which allowed him to find his own way and to develop an understanding of style and theoretical concepts such as modal modulation. This was enabled by trial and error experimentation coupled with self-guided correction based on the traditional parameters

192 that he had absorbed in the course of listening to and imitating Youssef’s playing and that of other players on recordings.

Listening back to recordings of his early lessons, Tawadros notes a definite lack of direction in his playing: a particular non-Arabic style characterised his improvisations.

He feels that he did not develop a concept of style and tradition-appropriate modulation practices until his teacher went overseas for a period of time, forcing him to remember what his teacher sounded like. Tawadros believes this reinforced in his mind the sound of traditional ‘ūd playing, an internal realisation that helped to shape his improvised melodies.

Tawadros’ knowledge of traditional songs grew substantially due to the musically fertile home environment in which he lived, and often he would take songs to Youssef for elaboration. Not only did Tawadros absorb a stream of culture at home, but also in addition to his mother’s love of music, several family members provided motivation for his journey. Tawadros’ maternal grandfather Mansi Habib was a violinist and ‘ūd player, and although he tragically died long before Tawadros was born, his reputation and legacy provided Tawadros with substantial motivation and continues to do so today.

His maternal uncle Yacoub Mansi Habib, a professional trumpet player, also played an important role in Tawadros’ life, initially sending taped lessons containing important pedagogical songs such as dūlāb and other historically important recordings given by

Egyptian ‘ūd players with whom he performed and had met during his professional career. Although Yacoub Habib had retired from playing when I met him in Cairo in

2008, it was clear that in addition to being a significant mentor to Tawadros, he had facilitated numerous introductions for Tawadros with local musicians on his trips to

Cairo, many of whom became important influences in his development as a professional

‘ūd player, and to a lesser extent, as a fine Arabic violinist.

193 One of the most important of these relationships that Tawadros formed as a result of his uncle’s assistance was with the violinists, brothers Esawi and Abdo Dagher, both showing Tawadros heartening generosity, especially in the case of Esawi, with whom

Tawadros has spent many hours sitting, jamming, talking, and absorbing his culture’s musical tradition. Even though he is now familiar with Esawi Dagher’s style, noticing a repeated phraseology from one performance to the next, he visits him whenever he can on his frequent return trips to Cairo, as he feels he absorbs more Egyptian musical culture through the experience. He is driven to capture in his own ‘ūd playing the vocal effects that characterise Dagher’s style, such as slides between notes, dynamic note swells, and vibrato.

In Australia, while continuing his lessons with Youssef, Tawadros says that he was always teaching himself, discovering new traditional melodies and craving to learn new aspects of Arabic music theory. Yearning for new knowledge to fuel his self-ordained quest to become an ‘ambassador for the ‘ūd’, Tawadros became essentially autodidactic, trying to learn everything he could about his instrument and Egyptian music in general.

His first meeting in Cairo with Dr Alfred Gamil is evidence of his success in this pursuit. Impressed by the extent of Tawadros’ knowledge, Gamil told him that even the local students that Gamil taught at The Higher Institute for Arabic Music would not have the detailed knowledge of modal theory that Tawadros demonstrated. Gamil repeated this compliment in my presence. Furthermore, Tawadros’ knowledge of

Egyptian music history, including a familiarity with important musicians of the past, a grasp of influential Egyptian music literature, and an awareness of important recordings, astounded Gamil, inspiring him to give many hours of informal lessons to Tawadros, who spends several nights improvising with Gamil into the early hours of the morning whenever he returns to Cairo.

194 Interacting with a variety of professional musicians, Tawadros also noticed that traditional melodies did not have a form that was as fixed as Western melodies, the custom allowing for multiple melodic variations for any given song. This flexibility speaks to the nature of aural transmission, and the existence of melodic variance points to the proximity of composition to improvisation. In the Egyptian case, this is not only due to the aforementioned emblematic nature of traditional melodies, but also due to the fact that learning multiple renditions of a traditional melody, thereby incorporating a large portion of ornamentation and expressive techniques, is a process that shares elements in common with the process of improvisation.

Reflecting on the role Youssef’s teaching played in his development, Tawadros points out that whilst directly prescribed alterations were rarely made by Youssef in regard to both ‘ūd technique and taqsīm theory, he did receive some advice about the structure and trajectory of his improvisations. Specifically, Tawadros was advised not to be ‘pick happy’, that is, he was encouraged to reduce the speed at which he played so that his improvisations could develop at a slower pace. Initially Tawadros was reluctant to do so, at first disliking players such as Munir Bashir whose technique he thought was simplistic given the lack of speed demonstrated when improvising. However, he now enjoys taqāsīm performed by players exactly because of the manner in which they slowly reveal their improvised phrases. His own playing now reflects this maturity, though he does possess an astounding virtuosity.

5.5 Conclusions

Improvisation in Egyptian music, taqsīm and mawwāl, is an intuitive musical behaviour.

Even though most performers view it as having no rules as such there is certainly a necessity for traditional musical details, tafasil, to be enacted during performance.

195 Tafasil are internalised by modelling musical behaviour on traditional prototypes. In doing so, instrumentalists and vocalists are able to perform in a stylistically appropriate fashion with ihsas; characterising their performance as a taqsīm rather than as an idiosyncratic improvisation.

This is realised in the course of performance by: combining the pitch material of maqāmāt and rendering the melodic characteristics of each maqām correctly, performing the correct intonation of pitches in each maqām, and by incorporating idiom-specific expressive techniques such as vibrato, ornamentation, melismae, glissandi and other details that primarily originated in, and still inhabit, traditional vocal music. Indeed my fieldwork establishes that traditional vocal music, sacred and secular, figures as a prominent archetype for all Arabic music. Finally, performances are structured traditionally by performing cadential figures, qafla, at appropriate points whilst improvising, and performers also undertake modal modulations in a stylistically appropriate manner.

The means through which musicians develop the ability to improvise, thereby realising these genre-specific musical elements, is multifaceted and is enabled by engaging in various learning environments that often overlap and interact. These include: parental apprenticeships, informal apprenticeships with master musicians, vocalists and ensembles, musician networks, autodidactic study, and institutionalised study. Indeed, for many musicians in Egypt, the ability to improvise is generated by interacting in some way in each learning environment, and those born into a musical family have an identifiable and unsurprising advantage. Specifically regarding the majority of musicians whom I interviewed, development was enabled primarily through autodidactic study within an environment stimulated and facilitated by one or more family members, master musicians, and ensembles. While some musicians active in

196 Cairo today have studied formally in an institution such as The Higher Institute for

Arabic Music, the limited role such institutionalised study plays in developing the ability to improvise has been established.

Within these enriching milieus, arduous aural learning represented in the habitual imitative practice of memorising prototypes (compositions and improvisations) enable musicians to develop a high level of aural acuity through this intensive aural activity.

Furthermore, by constantly imitating archetypal examples musicians internalise both the content and style of traditional improvised practices. The contents of taqāsīm are guided by absorbed knowledge of how maqāmāt should be treated melodically, how modulations can be navigated appropriately, and the manner and means through which taqāsīm are structured. Concurrently musicians develop an internalised sense of style that shapes expressive techniques such as manipulation of timbre, vibrato, ornamentation, articulation, and portamenti. The synergy of these activities, where advanced aural acuity combines with an internalised knowledge of content and an intuitive sense of style, enables musicians to improvise traditional taqāsīm or mawwāl in a stylistically appropriate manner, spontaneously crafting improvised melodies by ear that conform to the principles that have been internalised.

5.6 A final thought: can improvisation be taught? Or must it be learnt?

Although Egyptian musicians do not learn to improvise through explicit didactic instruction delivered by a teacher, the question remains: can improvisation be taught?

Or must it be learned?

197 Esawi Dagher

As indicated above, Esawi Dagher suggested that taqāsīm have no rules, and he consequently asserted that he could not teach someone to improvise. Rather than acknowledging the complexity of the situation as represented by Thomas’ analysis presented above – that the complexity of the melodic details (tafasil) inherent to taqsīm prohibits the reduction of performance practice to a set of rules – it was stated simply that it could not be taught. However, in addition to recommending that a musician wishing to learn how to improvise should mimic a vocalist singing religious material, in a subsequent meeting Dagher gave me a lesson where his expressed intention was to pass on the ihsas, or Eastern sound of certain phrases to me. He did not explicitly articulate the musical components separately, nor did he identify them specifically as the tafasil that conjure the appropriate ihsas. However, by playing examples of what I should play he was teaching me to improvise phrases according to a set of unarticulated requirements.

The process we engaged in was one in which I imitated him phrase by phrase until he indicated approval of my replication. Corrections made focussed on specific details such as the contextualisation and intonation of quartertones,39 the position and structure of ornaments, and the use of cadential figures to structure my improvisations. Of particular emphasis was the manner in which my playing accurately replicated the vocal-style phrasing, including techniques such as swells, portamenti between notes, the speed and width of vibrato, and the tapering of notes as performed by Arabic vocalists such as the sheikhs whom Esawi Dagher trained with as violin accompanist.

Reflecting retrospectively upon the emphasis he placed on achieving these vocal effects after returning to Australia, having just read Thomas’ recently published thesis, I

39 Or rather more accurately, three-quarter tones.

198 was reminded of the vocal origin of many of the ihsas, as reinforced by Thomas (2006, p. 188). In my lesson with Dagher, eventually I combined these phrases, or phrases that were close variations of Dagher’s prototypes containing what I perceived to be the essential tafasil, into longer form improvised phrases. Dagher then criticised these longer phrases, focusing on one or another tafasil element that needed further refinement before my playing represented an acceptable design achieving appropriate ihsas. As I mentioned earlier, Dagher’s pedagogical emphasis reinforced the historical perspective outlined in Chapter Two concerning maqāmāt and taqsīm conceptualisation, even though Dagher did not show knowledge of these sources as such.

Furthermore, this experience suggests that whilst he was not abstracting rules and teaching phrases to me through a progressive didactic methodology, the process of imitating model phrases through mimetic call and response teaching, combined with verbal teacher criticism certainly had noticeable results in my own playing to the extent that my responses became more acceptable to Dagher. It seems true that taqāsīm, as governed by the imperatives described as tafasil by Thomas (2006, p. 187), can only be learnt by imitation given their complexity due to their vocal origins. However, what my interaction with Dagher showed is that if a student possesses a sufficient aural ability so as to not frustrate a teacher, a teacher can teach students to perform phrases of taqsīm with appropriate ihsas through teacher-student imitation and importantly, verbal criticism and specific correction. I was fortunate to benefit from Esawi Dagher’s generosity and expertise, and in doing so was able to bring Thomas’ assertion to life:

‘successful students are able to learn the implicit considerations that inform melodic performance through painstaking listening and imitation of masters’ (2006, p. 190).

Dagher’s active role in this process demonstrates that it is possible to teach improvisation to musically advanced students possessing sophisticated aural skills.

199 Indeed, with the aid of master teachers, learning to improvise may be accelerated beyond the gains that are made pursuing this process autodidactically.

Dr Alfred Gamil

Although recognising the inherent difficulty in this stage of teaching, Dr Alfred Gamil also feels that he can teach improvisation, and he details how he has approached teaching a taqsīm method to beginner improvisers. After outlining the notes of a maqām, he asks the student to ‘introduce him to the maqām’. Gamil prescribes a specific method for beginning the first phrase of the taqāsīm, stressing that it should not begin on the tonic, but rather should approach the tonic from elsewhere in the maqām.

Proceeding in an ascending fashion after the tonic of the maqām has been introduced, he stipulates the taqsīm should continue by introducing each note of the maqām, making each a temporary focal centre highlighted by ornamentation and vibrato. He demonstrates this in an ascending fashion, at times returning briefly to the maqām tonic.

The first portion of his pedagogical method is focussed around the first jin, or tetrachord. According to Gamil, the second jin should be the next major focal point of the taqsīm performance, each note exposed in a similar fashion to the notes in the first jin. After both tetrachords have been introduced adequately, a modulation can take place, and Gamil teaches beginning students to make the modulation from the highest note of the first tetrachord, which then becomes the first note of a new maqām’s lower jin, a conjunct modulation by way of a common note. Gamil also demonstrates a number of other modulatory possibilities at this point, outlining how they may take place theoretically.

Reflecting retrospectively on this interview, Gamil’s pedagogical narrative reveals substantial evidence supporting those sources introduced in Chapter Two that advocate

200 a definition of maqāmāt that goes beyond modern scale-based conceptualisations.

Furthermore, Gamil’s careful prescription of melodic movement, via the tetrachords of the maqām, also demonstrates a similarity to the Mashāqah treatise and research by

D’Erlanger, explicated by Marcus (1987, 1989). This suggests that while these sources were not referenced explicitly by Gamil, and while he may deny they impact upon his performance practice, I propose that their style of conceptualisation is apparent in his musical behaviour, and I suggest that this approach has been internalised passively into his methodology by learning the musical tradition aurally, as an autodidact.

Whilst recognising that it is possible to teach a method of improvisation to students,

Gamil stressed that success is dependent on the student’s imaginative talent. He also emphasised the importance of knowing the compositions and improvisations of other musicians:

You can teach improvisation but it is not so easy, and it depends on the imagination of the student … and also it depends on how many songs and improvisations he [has] memorised, he [has] heard. He has to have [memorised] a lot, and [then has to] forget it.

Thus, even though Gamil can lead a student through how a taqsīm should be played, the amount of time spent autodidactically learning the songs and improvisations of the

Egyptian tradition figure heavily in the process that leads to an eventual ability to improvise.

Joseph Tawadros

Despite being largely autodidactic, having developed within a non-prescriptive teacher- student relationship, Tawadros now teaches ‘ūd in a far more prescriptive fashion than that in which he learnt. His methodology is direct and detailed, employing many of the activities that he has developed as part of his autodidactic mission.

201 Included in this pedagogical repertoire is a practice that he continues to make effective use of throughout the course of his career: the composition of study pieces designed to specifically develop certain ‘ūd techniques for improvisation. Unlike

Western instrumental studies that strive solely to develop technique for use in performing compositions, including at times the studies themselves, the overriding success of Tawadros’ etudes is due to the manner in which they facilitate the development of a strongly technical improvisational fluency. Dr Alfred Gamil also composes and practices studies as a part of his autodidactic preparation, a methodology that he professes has been adopted through his interactions with Tawadros.

Few of Tawadros’ studies are notated. Rather, they are memorised through the compositional process and he relays them to his students phrase by phrase aurally. In fact, Tawadros often composes new studies spontaneously during lessons as the needs of a student become apparent. I benefitted from these purpose-designed studies when taking lessons with him on the ‘ūd. Students are taught the next phrase once he is content with their replication of his prototype, and most students record his lessons so as to allow time at home to practice them with more accuracy than their less-developed ability to memorise would otherwise allow.

The studies focus predominantly on picking technique combined with developing finger dexterity in certain modes and throughout the range of the ‘ūd. Significantly, a prominent feature of all of his studies is that they are imbedded with a natural concept of motivic development and melodic form. When learning these studies, the benefit gained is not from memorising phrases to be reused when improvising, but rather, the internalisation of a dexterous technique and a method of melodic variation that facilitates melodic fluency when improvising. Given the fact that taqsīm is a monophonic modal improvised practice, melodic fluency is paramount.

202 Not surprisingly, this pedagogical concept is also directly influenced not only by the manner in which Tawadros developed his own improvisational style, but also by the context in which he views the improvisational process:

Improvisation is like a bunch of studies that you’ve learnt, [but] it’s not note for note, it’s with listening, it’s everything you’ve ever listened to in your head, everything you’ve ever absorbed in your head, and that moment in time is how your mind decides to arrange them, and it depends on what’s going on in the background; that influences your thing. If there’s a riff, if there’s a rhythm, or whatever, a drone, then that’s going to influence the way your mind arranges the music at that point. So, you better have the technique to accommodate that.

The final sentence of this interview excerpt, ‘So, you better have the technique to accommodate that’, is perhaps the most illuminating in that it clearly demonstrates the importance that technique plays in being able to improvise. As expressed in the earlier portion of the excerpt, Tawadros feels that what is improvised is strongly influenced by what has been heard and played before. When musicians improvise, they are traversing musical pathways that have been travelled in someway before, both by themselves and by others. Fingers follow familiar patterns, and musicians often play a type of melodic shape or phrase that has been played in some fashion previously. A musician’s ear is able to guide a melody’s construction based on what has been heard before.

Additionally, absorbed methods of motivic development operating on an intuitive level enable melodic fluency that is guided by the capricious spontaneity of the moment and shaped by the real-time events experienced in the course of performance. At all stages

Tawadros reinforces that this process is enabled by well-established technique, and this is certainly the focus of his pedagogy.

203 Chapter Six The Cross-cultural Comparison

In this chapter I will present the details of a cross-cultural comparison between the learning environments and related methodologies that musicians encounter when learning to perform improvisations in Egyptian music, jazz, and Hindustani classical music. At all times during this analysis, historical perspectives will be contextualised with modern-day trends so that the modes of transmission currently active in each tradition may be represented by this cross-cultural comparison. The comparison will draw out parallels and disparities as discovered in: (i) the environments in which musicians learn to improvise, (ii) what the methodologies are that musicians undertake within these spheres, and (iii) how these methodologies are taught or assimilated into autodidactic practice – the mode of transmission. Throughout this comparative analysis

I will consider best practices, and I will propose potential methodological transfers between traditions. Additionally, as a means of furthering an understanding of the improvisational process from preparation to performance, the impact that teachers have on the development of improvisational ability will be evaluated.

6.1 Comparison of learning environments

6.1.1 Early childhood learning environments

By comparing the information presented in Chapters Three, Four, and Five, it may be concluded that through experience of actual performance practice and recordings, contact with music in early childhood has historically originated within the family environment in all three traditions under examination. Egyptian children often rely on

204 practical coincidence, being born into a family of musicians, to dictate whether they are exposed to music making at a young age.

Alternatively, opposition to music making can sometimes make it difficult for youngsters to pursue music making, though my research demonstrated that this can be overcome by the committed. Furthermore, as detailed in Chapter Five, some Egyptian respondents noted that their interest in music at a young age was inspired by recorded music in the home environment, curiosity being developed outside direct contact with family musicians, though often inspired by their parent’s enthusiasm for listening to music. Historically, once traditional music has been introduced to children, experimental playing will often occur in a nurturing family environment, where preliminary teaching of some kind may take place.

As Chapter Four detailed, a similar trend is noticed in the jazz tradition, evidenced by adult musicians who reflect upon the stimulating sound environment that they encountered at home as young children, often created by parents who were musicians, professional or amateur, or perhaps just music devotees regularly listening to their favourite jazz recordings or radio programs. Inspiration gained from listening to recordings or live performances often translated into early investigative forays with available instruments. Whilst exceptions may exist, it is extremely rare in either tradition for musicians to acknowledge a childhood environment that was devoid of music, and it is most common for musicians to have parents who were involved in music making in some form, even if only as housebound amateurs. If not, social interactions outside of the home are nominated as a source of early musical inspiration

(such as in the church or school in the case of jazz musicians). These often supplement the experiences of those children who fortuitously interacted in a musically rich environment at home.

205 The requirement of a musical home environment during childhood is accentuated incontrovertibly in the traditional North Indian experience. As Chapter Three detailed, children who eventually continue to a musical career have historically, almost always, come from musical families. Early childhood encouragement resulting in experimentation with music has traditionally been enriched and developed through teaching given by family members. Before their teenage years, a child was normally taken into the nurturing musical household of a guru who may have been a relation of the new disciple, though as Chapter Three detailed, the definition of a musical family following a gharāna lineage is no longer the most prevalent pattern of musical development.

6.1.2 Apprenticeships

Although musicians in all three traditions are inspired and educated throughout early childhood by interacting in their family environment, substantial differences emerge after childhood. Traditionally, the experiences of North Indian musicians may be uniquely distinguished by the guru-shisyā relationship, identified by the ritualised structure that is embodied within this model, and derived from an Indian religious consciousness. This formalised system of transmission lying at the core of the

Hindustani music tradition is historically characterised by an intense daily training, which has been documented as beginning with students as young as under ten years of age (Shepherd 1976, p. 52), but also commonly originating during the teenage years

(Shepherd 1976, p. 51). In an arrangement lasting until they become professional musicians in their own right, shisyā are noted as commonly living with their guru, even in cases where they are not immediately related (for example, when the guru is an uncle or a parent’s cousin), or when they are not biologically related to the shisyā at all. This

206 contrasts markedly with the jazz and Egyptian music traditions, which are less characterised by familial specialisation. Although family-based learning has occurred to some extent, these instances have certainly not been as formalised, structured, nor as common as those that have occurred historically in Hindustani music, and at no time have they been posited as an ideal paradigm. Furthermore, the Hindustani appreciation of the guru-shisyā relationship is evidenced in the fact that familial opposition to a musical career was traditionally rare in India. This may not be the case in the growing middle class today.

Changes in Indian society discussed in Chapter Three have resulted in it becoming rare for musicians to take on full-time shisyā. This situation has also undergone changes from the disciple’s perspective: today shisyā often balance other responsibilities with time spent with their guru (Ruckert 2004, p. 36). Moreover, education is now offered in private music colleges, taken up by children whose parents are attracted by the social and educational benefits that this experience affords their children (Ruckert 2004, p. 37) rather than the serious possibility of pursuing a musical career. As such, and in some cases given the inadequacy of training in some institutions, it is extremely rare for graduates from these colleges to undertake a career as a professional musician (Meer

1980, p. 147).

Both Alter’s and Scott’s research elaborated on pedagogy delivered in other colleges in which teaching based on an adapted guru-shisyā model is employed.

However, children – that is, students at the age at which youngsters would begin serious study with a guru – were not noted in the student body. Instead students were primarily young adult Indians wishing to pursue musical careers. Mature-aged Western musicians desiring to learn Indian music in a near-traditional environment also made up the intake of students in the comparable institutions situated in the West. Some also attend those

207 colleges in India. As reinforced in Chapter Three, although the guru-shisyā learning environment has become less common in its original form, it remains the underlying pedagogical goal for emulation and accepted best practice in the Hindustani tradition.

The guru-shisyā model’s philosophical conceptualisation, and consequential influence over modernised adaptations of the model is that which distinguishes Hindustani music from jazz and Egyptian music in this regard.

At this point I will emphasise an exception to this finding. I refer to my exposition in Chapter Five that detailed a virtually extinct mode of transmission existing in the

Arabic music tradition, in decline since World War One. In this practice, singers and sometimes instrumentalists worked closely with religious singers, learning repertoire and absorbing aspects of style and performance conventions through imitative performance practice. As such, historically a practice existed in the Egyptian tradition that was comparable to the guru-shisyā tradition of North Indian music, although this was never the totality of an Egyptian musician’s training, or an aspect of every Egyptian musician’s training. Nevertheless, although this similarity existed historically, few of those musicians I interviewed were able to take part in this learning environment, as it was virtually extinct. Most significantly, unlike the North Indian experience, this religious apprenticeship system has not been translated to an institutional model, which has followed a secularised Western-styled format instead. Furthermore, while some of the Egyptian musicians I interviewed followed a family tradition, often of their father or older brother, there were numerous exceptions to this pattern, and the relationship they engaged in with family members was not formalised in terms of philosophy or religion beyond a strong belief in musical inheritability.

Moreover, the educational experiences encountered by Egyptian musician respondents who learned their craft from an older family member differed on a case-by-

208 case basis concerning the intensity, methodology, content, and regularity of instruction.

Whilst musicians such as Ahmad El Haggar learned his tradition from his grandfather and father in a manner that resembles the Hindustani model, his instruction was given at irregular intervals, on a weekly not daily basis, and it ceased well before he began his tertiary studies. Furthermore, he did not live with his grandfather, his primary teacher of traditional music, but rather visited him at weekly intervals. Of those musicians I interviewed, Abdo and Esawi Dagher’s periods of time working with sheikhs bear the closest resemblance to the historical Hindustani model, especially in the case of Abdo, who lived with Sufi orders after he left his father’s home as a young teenager.

Nevertheless, Abdo had to move to Cairo to turn earn a living, and the career that he then pursued bore no direct resemblance to the tradition lived by the Sufi orders, even though his musical style is identified as being indebted to the religious music performed in these orders (Dagher 2000). The Dagher brothers, who are elders of their musical tradition, have lived experiences that are representative of an informal apprenticeship- like learning environment, but they are certainly the exception: none of the current generation of musicians I interviewed had apprenticed with sheikhs or Sufi orders.

Learning the improvised jazz craft from a parent or older sibling has taken place historically, though as Chapter Four detailed, comparable to the Egyptian tradition and distinct from the North Indian model, it has not occurred as part of a formalised apprenticeship system embodied within musical tradition. In a similar fashion to the divergence of experience within the Egyptian music tradition, the nature of these relationships varies from musician to musician in the jazz community. While some may benefit from familial experience, the time span that a novice may spend under the guidance of a family member can vary drastically from family to family. So to can the regularity of instruction, the content, and mode of transmission experienced as they

209 learn. Furthermore, nothing comparable to the common Hindustani practice of a shisyā living with his or her guru was evidenced in research conducted into the jazz tradition outside of parental connections.

Slawek (2009, pp. 201-203) presents an exception to this finding, asserting that

American guitar teacher Joe Srgo was perceived by his students (who included his nephews) in a manner similar to a shisyā's perception of their guru. Srgo was held in the highest esteem by his students, and his stable life, influenced by his religious beliefs

(2009, pp. 204, 215), allowed Srgo to provide a consistent education to his students, much like the manner in which Hindustani gurus care for the musical education of their shisyā. Slawek also invokes the 'socio-musical unit of gharana in Hindustani music'

(2009. p. 203) to describe Srgo's teaching practice, drawing another link between the seemingly disparate traditions. In including Slawek's study (2009) I emphasise that his findings, when viewed within the broader spectrum of jazz literature are an uncommon, albeit fascinating exception to the rule.

6.1.3 Informal apprenticeship-like relationships with master musicians and in ensembles

My research established that it is more common for musicians in the Egyptian tradition to undertake more informal periods of learning under the guidance of a master musician who often is not a relation, and normally under the guidance of more than one experienced musician. Periods of learning varied from musician to musician, and at times substantial inspiration and growth was experienced after only a short interaction.

Lengthy tenures with one master musician other than a family member were less common. As Chapter Four emphasised, jazz musicians have also historically interacted with esteemed professionals, though apprenticeship-like experiences tend to have occurred as musicians entered the professional arena on the bandstand, rather than as

210 young students, as is common in the Egyptian model. Before engaging in apprenticeship-like experiences in the professional arena where more formal and often more advanced training takes place, jazz musicians tend to interact in preparatory simulations called jam sessions.

6.1.4 In situ training: the jam session

Before jazz musicians are granted professional performance opportunities, they interact in a variety of informal learning environments in which they refine knowledge of repertoire and develop their improvisational ability. The discussion of jam sessions presented in Chapter Four established that these opportunities allow musicians to develop and prove their improvisational ability in a model that replicates professional performance, but in a context where assessment by peers is often immediate and explicit. This model reinforces the traditional autodidactic emphasis in jazz, where the responsibility is placed on students to discover both what they need to know and how to teach it to themselves, rather than being taught the content by a master or guru.

Furthermore, a strongly competitive element was exposed throughout this discussion.

Although not named as such (except by a younger generation of musicians), jam sessions are also part of the music scene in Cairo. Whilst sessions in the jazz tradition often took place in music venues, Chapter Five demonstrated that the Egyptian equivalent is often found in the home of esteemed masters where younger musicians gather to perform informally and are edified by interacting with the master, other professionals, and other likeminded students. In this environment, analogous to jam sessions in the jazz tradition, students learn traditional repertoire, techniques of improvisation, and stylistic aspects of their tradition all by way of osmotic and mimetic learning.

211 It is important to note that in both traditions, jam sessions are not exclusively pedagogic, and in some instances are not pedagogic at all. Furthermore, while jam sessions exist as learning environments in jazz and Egyptian music, the fact that the sessions are often conducted in the presence of an experienced Egyptian master instrumentalist distinguishes those activities conducted in the Egyptian tradition from those experienced by aspiring jazz musicians within their peer network. Indeed, a jam session in the Egyptian tradition may supplement a young musician’s learning experiences that have already been instigated with the master musician.

The unabashed competitiveness common to the jazz experience was not evident in the Egyptian jam sessions that I witnessed. From my observations I would consider that this is the result of the Egyptian experiences being mediated by a senior master musician, in conjunction with multiple generations of the music community often being in attendance. Behaviour demonstrated whilst sitting with a master musician in this situation is defined by the cultural tendency of expressing respect for one’s elder, a tendency which deems drawing attention to oneself inappropriate. Improvisations that I witnessed undertaken by young musicians in front of master musicians such as Abdo

Dagher were modestly performed and extended by seeking his approval to continue.

As a point of distinction, research indicates that jam sessions are not a prominent aspect of the Hindustani tradition. While an equivalent can be observed in the practice of students playing for each other in the presence of their guru, the interactive and competitive facets emphasised in discussions of jazz jam sessions are by nature not common to these Indian experiences. Given that the individual soloist is the focus of performances, and although interaction occurs between accompanying musicians and soloists, the lauded interactive learning that takes place between sparring soloists in the jazz jam session is not possible in the Hindustani model, as the discussion of the

212 performance context in Chapter Two outlined. In this way, similarities are evident with the cultural dynamics of the Egyptian experience. Additionally, the structure of

Hindustani performances, in which soloists play for extended periods of time, does not accommodate more than one soloist within the ensemble hierarchy. Even though adaptations of learning exist that allow for a succession of students to each elaborate for a short time in a rāga, the point of focus is the pedagogic interaction with the teacher, not necessarily a mutually encouraging or competitive interaction between students

(Personal communication, Dr John Napier, 10 March 2012). Table 2 summarises these findings.

Table 2: The use of jam sessions as a learning environment

Interaction in jam Master teacher- Peer-mediated sessions mediated

Jazz Hindustani n/a n/a Classical Music Egyptian Music

Following informal learning experiences in jam sessions, jazz musicians often progress to on-the-job training (see Table 3 for a summary of the following analysis). This type of apprenticeship-like learning experience is a prevalent trajectory for jazz musicians who, after being noticed by an established senior musician, are recruited into the ranks of a working band and given the opportunity to prove their mettle in actual performance situations. The experiences of musicians debuting in Miles Davis’ quintet, such as John

Coltrane, Tony Williams, and Wayne Shorter attest to this trend. Art Blakey’s Jazz

Messengers has been another important vehicle for on-the-job ensemble training led by an experienced professional. Though these are famous instances, this pattern is repeated across the profession, at all levels of expertise and prestige. Similarly, as I described in

213 Chapter Five, training within the ranks of ensembles is a common learning environment for Egyptian instrumentalists, who work as section players in traditional orchestras and tahkt ensembles, learning alongside more experienced players and soloists. Young musicians are educated in traditional repertoire and refine their improvisation skills in rehearsals and finally in performances. In contrast to the jazz and Egyptian experience,

Chapter Three emphasised that given the nature of Hindustani music performance practice, which focuses on the individual relationship between teacher and student, and is principally a soloistic presentation, apprenticeships within ensembles are not a common pathway traversed by Indian musicians. Musicians are historically not accepted as solo performers until they have received their guru’s blessing, often garnered after shisyā prove themselves by accompanying their own guru, and performing as a soloist supporting their guru.

Table 3: The nature of historical apprenticeships

Egyptian Hindustani Jazz Music Classical Music

Formal

Informal

Derived from religious belief

Regular, long-term, with one master-teacher

Irregular, with multiple master- teachers

Live in master-teacher’s home

Ensemble-based informal apprenticeships

214 6.1.5 Institutional training in modern-day contexts

Jazz

As Chapter Four clarified, institutional training is now a fundamental component in the education of almost all jazz musicians. Tertiary students study jazz degree programs in college, conservatorium, and university settings, and they are taught using curricula and examined according to criteria that have been adapted from and built upon a foundation derived from Western classical music pedagogy. The benefits and shortcomings of this change in how jazz musicians learn their craft have been discussed at length in Chapter

Four. As a point of summary, I would assert that whilst the structured teaching of jazz may deliver students a well-rounded education in historical and theoretical issues, and even though the communal education environment may provide students ample opportunities to network, successful students realise that they must engage in arduous autodidactic study and individual practice, including many hours spent transcribing recordings of past masters, before they are able to join the professional ranks of jazz musicians.

Indeed, it may be argued that many of the skills demonstrated by students in performances are developed outside the institutional context during individual practice sessions. Guidelines may be prescribed by individual instrumental teachers in private lessons and in group improvisation classes, but the individual effort and responsibility for teaching one’s self is still a vital aspect of the process, and the most important learning environment is still a musician’s practice room. Institutions merely supplement necessary autodidacticism, especially concerning the development of improvisational ability, for which the standard required that enables students to perform professionally is extremely high.

215 Egyptian music

This trend is magnified in the Egyptian music institutional experience. Chapter Five has detailed the ways in which the curricula of Egyptian tertiary institutions fail to represent traditional music, diverting many instrumentalists onto a pathway of Westernisation, while their education in traditional performance practices is left to activities undertaken outside of the institution. One reason suggested by Dr Alfred Gamil for why tertiary institutions seemingly neglect traditional song is that Arabic song is not suited to organisation in a progressive syllabus, in as much as it is not possible to arrange the repertoire in a progressive manner over a long-term learning curve.

Moreover, specifically regarding learning to improvise, students receive negligible instruction at institutions such as The Higher Institute for Arabic Music. Instead, teachers grade them predominantly on their ability to perform Western art music and compose using Western modes of composition, rather than adapting the institutional learning environment to enable the instruction of traditional music by incorporating pedagogical models derived from traditional imitative methodology. To learn traditional music and improvisation, students still rely on traditional modes of transmission centred on autodidactic mimetic and osmotic learning models.

Hindustani classical music

The adaptation of institutional structures to educate musicians in Hindustani music has been more successful in comparison to the Egyptian model. Reacting to the gradual demise of the traditional guru-shisyā relationship (and responding also to the desire of

Western musicians to be educated in their traditional music), Hindustani musicians have succeeded in a number of cases by adapting selected traditional methods of instruction derived from the guru-shisyā parampara to a modern college setting. In making this

216 assertion, I emphasise that the numerous music colleges that provide a general education, as opposed to those that aim to produce professional performers, should be set to one side. Their activities are not related to training North Indian musicians to improvise, and their graduates rarely, if ever, pursue a musical career.

Models representing a more successful adaptation were detailed in the analysis of research undertaken by Alter, Scott, and Booth in Chapter Three. While the individual attention afforded by the traditional guru-shisyā relationship cannot be fully recreated in these modern contexts, important methods of instruction based on imitating the teacher have remained largely intact. Students learn in the traditional oral-aural form, where the musical tradition is transmitted in a holistic experiential manner rather than in a Western style of instruction abstracted from the performance context and taught theoretically. By basing modern adaptations on a traditional imitative pedagogy, the Hindustani model of institutionalisation has succeeded where the Egyptian attempt, and to a lesser extent the jazz attempt, have failed. See Table 4 for a summary of these findings.

I propose that an adaptation comparable to the Hindustani model would be an excellent basis on which to design a mode of instruction that better suits the teaching of traditional Egyptian music and improvisation, if one is deemed necessary. Indeed, while musicians still develop the ability to improvise in a traditional style outside institutions, given that the current model seemingly distracts students by emphasising Western music, an alternative would serve future generations of Egyptian musicians well, especially if Egyptians do seek to maintain the distinct identity of traditional classical music, a challenge clearly identified by the participants at the Arabic music conference meeting in 1930 in Cairo (Thomas 2006, p. 9).

Comparing the institutionalisation of these three traditions requires a complex evaluation. On the one hand, jazz has shared a close history with Western art music,

217 effectively co-existing within the same broad society for most of jazz’s history. At times each has influenced the development of the other, and jazz inherited musical structures and harmonic concepts from its Art music counterpart. Indeed, Chapter Four detailed the benefit that jazz musicians have garnered by studying Western music theory. As such, transplanting jazz education into the college or university environment is arguably a more appropriate decision, whereas imposing this structure on non-Western traditions requires a seemingly more complicated transition.

Nonetheless, as Chapter Four concluded, jazz musicians must still heed the responsibility to identify and learn the many skills necessary for performance, particularly, though not limited to, the practice of improvisation. Chapter Five demonstrated that this autodidactic behaviour also remains a key factor in the development of improvisational ability in Egyptian music. In contrast, Chapter Three demonstrated that Hindustani musicians traditionally learn many performance skills necessary for improvisation by interacting intensely with their guru’s teachings.

Consequently, given that guru didacticism has always underpinned how Hindustani musicians have traditionally learnt, it is not surprising that the institutionalised learning environments that maintain an adaption of this traditional model demonstrate that when appropriately conceived, institutionalisation is more suited to the Hindustani tradition.

Interestingly, students in modern jazz educational contexts do learn directly from teachers who, in response to the structure of the institution, teach students in areas that were previously developed autodidactically. Even so, students discover that extensive individual practice is still required to develop skills for improvisation, and they undertake important activities such as memorisation and transcription of solos that were undertaken by previous generations of ‘non-schooled’ musicians. Although teachers may introduce ideas and aid in the development of performance skills through advice

218 and critical feedback, substantial autodidacticism and individual practice are still necessary.

Table 4: How institutionalisation has been implemented

Egyptian Music Jazz Music Hindustani Classical Music

Attempted

Implemented in partially a suitable selectively manner

Improvisation skills taught partially successfully

Autodidacticism remains essential n/a for developing it never was improvisational ability Western-style Partially. Some argue institutionalisation is that autodidactic unsuited. When conceived learning is necessary Institutional Adaptation may be appropriately, to develop a unique learning suited possible, similar to the according to adapted voice, and therefore to the tradition Hindustani example traditional modes of the tradition is (i.e. according to transmission. unsuited to total traditional modes of institutionalisation. transmission).

If students were able to interact with their teacher on a daily basis, their teacher may be able to have more of a role in the development of improvisational ability. However, when lessons are less frequent, possibly on a weekly basis, extensive autodidactic learning will remain necessary to conquer the challenges inherent in improvisation. It is quite possible that autodidactic behaviour would remain necessary in any case, given the complexities of the tasks at hand. These include the traditional emphasis on creating a unique improvisational voice, which itself has the unintended advantage of enforcing

219 autodidacticism. Indeed, working closely with a teacher to the extent of replacing autodidactic behaviour with a teacher’s guidance would most likely reduce the individual character of the student’s improvisational style in a way that may be unsuited to the jazz tradition. The conventional learning focus would be reversed. Consequently, traditional autodidacticism will remain essential until such time that institutions develop a pedagogical approach that is more effective in developing a unique improvisational ability. It therefore seems that deemphasising the teaching of improvisation is appropriate, with the greatest value of the jazz institution being the instruction of theoretical concepts and the facilitation of professional networking and ensemble learning, in the absence of traditional networking opportunities.

6.2 Comparison of practices and related modes of transmission

Within the learning environments compared above, musicians undertake a series of practices that facilitate the development of improvisational ability. These have been individually identified in each musical tradition in Chapters Three, Four, and Five. An analysis will now be conducted that comparatively evaluates these practices, while also comparing the various modes of transmission at work in each learning environment.

Tables 5 and 6 summarise this comparison.

6.2.1 Melodic prototypes

The overwhelming point of commonality in the development of improvisational ability in Hindustani music, jazz, and Egyptian music is the internalisation by memorisation of genre-specific melodic prototypes. As Chapters Three, Four, and Five detailed, the musical information contained within these models has been compared to genetic information: a musical equivalent to the physical similarities between parent and child,

220 or perhaps even more accurately, the shared physical characteristics between siblings.

The composed and improvised music of a particular tradition is similarly identified by its shared musical characteristics. These character-giving musical genetics must be realised by musicians in performance for their improvisations to be acceptable representations of genre-specific parameters and rules of performance. Thus, in each tradition musicians memorise to the point of internalisation various melodic prototypes, enabling them to intuitively create their own models based on these internalised prototypes.

Composed melodic prototypes

Jazz musicians begin absorbing stylistic traits by learning the melodies of repertoire known as ‘standards’: those songs that are used as the harmonic framework for jazz improvisation formalised in score publications such as The Real Book. As Chapter Four summarised, these songs were originally derived from a variety of popular music sources, and over time other original compositions not derived from popular sources have also found their way into this perpetually growing collection. At this point, it should be emphasised that often knowledge of these songs originates in childhood and in such cases internalisation is achieved unintentionally through passive listening: aural osmosis in a musically rich childhood environment. This is later reinforced intentionally by actively internalising prototypes. By absorbing these compositions musicians gain a functional understanding of the interplay between melody and harmony, and as the number of examples known by a musician grows, commonalities are noticed between how harmony and melody interact and an aural-awareness of the ‘sound of jazz’ is developed by way of this dynamic interplay.

221 The prototypical value of the melody as a possible realisation of the underlying harmony is further emphasised by the practice of using the melody as a springboard for improvisations. Beginning improvisers often incrementally elaborate upon the original melody by developing personalised renditions of the original version, extending to using the melody as a fading shadow to which they occasionally refer in their improvisation, to eventually creating an improvisation that bears no obvious relationship to the original melody, but has grown out of this process of gradual experimentation (Berliner 1994, p. 70).

As Chapter Three demonstrated, by learning compositions North Indian musicians gain a similar benefit to that which jazz musicians receive. A guru teaches compositions

(bandish, gat, chiz) to shisyā using model phrases, whereby an example is sung or played by a guru and is copied until successfully replicated by shisyā. In doing so students develop a functional understanding of the stylistic and modal characteristics of the rāga chosen for performance. Chapter Three detailed the extent to which gurus in the North Indian tradition teach students using model phrases when teaching the arohavaroh and related phrases, imparting compositions, introducing new rāgas, and when instructing students in methods of improvisation. As research presented in

Chapter Three illustrated, compositions are blueprints that represent how some features of a rāga should be presented in performance: they offer a melodic prototype. Rather than teachers explicitly articulating rules of performance, learning compositions allow shisyā to absorb these distinguishing attributes osmotically, and a musician’s knowledge of one rāga expands in proportion to the number of compositions that are known. Indeed, as Chapter Four illustrated, jazz musicians undergo a similar transition of knowledge: as they increase the number of compositions known in certain chord

222 progressions, they develop a growing understanding of how the harmony functions and the variety of ways a melody may be designed over such progressions.

Significant similarities are found when comparing the role of pre-composed melodies in jazz and Hindustani music with the melodies of Egyptian music, centred again in the interaction between informal interactions during childhood in which musical characteristics are internalised unintentionally, and the intentional memorisation and autodidactic learning of prototypical melodies in later stages of development. Although a repertoire is not officially identified as ‘standard’, there are certainly songs derived from popular music, and made famous by Egyptian singers such as Umm Kulthūm and Mohamed Abdul Wehab, that are known throughout the Arabic world, as the research presented in Chapter Five has articulated. I discovered that these melodies are not only well known by all the musicians I interviewed, but also by people from all stations in life. Comparably, the popular songs that became the standards learned by jazz musicians were also known by the broader North American society, whereas bandish that are still learned by Indian musicians are not necessarily a well- known artifact, and in many cases they have never been publically disseminated.

Esawi Dagher asserted and demonstrated the extent to which the melodies of Umm

Kulthūm’s songs contain characteristics that, once internalised, allow musicians to replicate aspects of style and maqām practices, that is, how the traditional modes should be played to bring out their unique character. My description of Dagher’s display in

Chapter Five, weaving between vocal and instrumental phrases from an Umm Kulthūm song demonstrated this fact. This phenomenon compares directly to the benefit

Hindustani musicians receive when they learn compositions, in as much as the process schools them in a plan of how a rāga may be realised in performance. Reflecting upon this, I am reminded of Lee Konitz’s attitude toward the melodies of jazz standards, as

223 discussed in Chapter Four. His lyrical style advocates developing improvisations from melodies, insofar as the line is blurred between what is melody and what is improvisation. Common ground is shared by these traditions, as represented in this philosophy of improvisation.

Furthermore, as my interaction with violinist Esawi Dagher evidenced, Dagher clearly identified the extent to which traditional melodies embody an Arabic style, described by him as an ‘Eastern flavour’. Indeed as Chapter Five disclosed, numerous musicians identified knowledge of this repertoire as essential to their functional understanding of the stylistic parameters of Egyptian music. In this way, the function of these songs in the education of musicians in Egypt is certainly similar to the role played by jazz standards in jazz culture. This fact was more specifically articulated by musicians such as Dr Alfred Gamil and Ahmad El Haggar, who identified their knowledge of this traditional repertoire as fundamental to their understanding of theoretical Arabic modal concepts codified in the maqām system, which they assert was learnt unintentionally by becoming knowledgeable in this repertoire. I assert that this process should be labelled aural osmosis.

A parallel is apparent in the jazz tradition, where by learning standard repertoire, musicians learn harmonic structures. However, it is rare for musicians to learn this unintentionally, given the multilayered complexity inherent to jazz harmony. A more considered and concentrated approach is necessary. Notably, both jazz and Egyptian musicians, whose knowledge of traditional repertoire, at least initially, is often developed through informal listening sessions as part of daily routines, differ from their

Hindustani counterparts, who are taught compositions by their guru after deciding to pursue the journey of being a shisyā. This is particularly the case today, whereas in the

224 past living with a guru and overhearing performances and lessons would have allowed for some of this musical information to be absorbed.

In addition to popular songs, Chapter Five demonstrated that Egyptian musicians also learn other traditional classical compositions: the short dūlāb, the longer samai and longa, and other traditional compositions, all embodying essential stylistic attributes, and all exemplars of melodic potential. Given that improvisations are commonly performed as instrumental introductions and interludes between vocal compositions, or between sections of the one composition, knowledge of these vocal compositions is particularly helpful in stylistically refining an instrumentalist’s improvisations. As in the jazz tradition, these pre-composed melodies not only train musicians aurally in an appropriate style, but furthermore, as in the case of jazz melodies and Hindustani compositions, they are used during performance as a catalyst for improvisation, or as a recurrent ‘theme’ (or starting point).

As I wrote in Chapter Five, during my interaction with Dr Alfred Gamil he spoke of how dawālīb (plural of dūlāb) are often performed before improvisations to prepare the musician to improvise in a manner that realises the modal characteristics of the maqām (mode) that is chosen for improvisation, while also introducing the modal context to the audience. A close similarity therefore exists between the Egyptian dūlāb and Hindustani compositions (bandish, gat, chiz). Performing a melody, before improvising in a way that realises its underlying harmonic progression, is also a fundamental aspect of traditional jazz performance practice, as Chapter Four emphasised. There are certainly significant correlations between all three traditions in this regard.

225 Influence of religious melodic prototypes

Egyptian Esawi Dagher also suggested that the adhān recitation, the call to prayer that is undertaken at regular intervals of the day and amplified throughout Islamic cities, is another important form that embodies the essence of Arabic style, absorbed passively simply by living in an Islamic society. As I articulated in Chapter Five, Dagher asserted that I only needed listen to a sheikh recite these prototypes in order to learn how to improvise in an Egyptian style.

As Chapter Four detailed, traditional North American religious contexts such as the church have contributed to the musical education of children by providing early performance opportunities, as well as direct assistance by providing instruments and lessons to children. Indeed, one of the first opportunities to sing was given through family religious observance. There is also an established proximity of style between the preaching style of African-American ministers of religion, African-American religious music, and certain aspects of the style of gospel, soul, and jazz musicians.

More specifically, it has been suggested that saxophonist John Coltrane improvised with the chanted intonation of a Baptist preacher (Porter 1998, p. 246), and a direct link has been drawn by Phyl Garland between the techniques employed by singers such as

Aretha Franklin and Ray Charles, and the Baptist preachers of his youth (Garland 1969, p. 24). Williams-Jones also emphasises this relationship by identifying the similarity between a preacher’s rhetorical delivery of the gospel message, and the style of the gospel singer: ‘the singer perhaps being considered the lyrical extension of the rhythmically rhetorical style of the preacher’ (Williams-Jones 1975, p. 381). Although there are clear stylistic associations, identified by methods of emphasis and articulation, between religious practice and soul, gospel, and jazz music, a comparable similarity of melodic content that is identified as existing between Egyptian Islamic vocalisation and

226 improvisation in Egyptian music was not noted by jazz musicians when reflecting upon their improvisatory melodic style.

As Chapter Four described, responses collected by Berliner (1994) regarding the influence of the church were limited to evidence of early performance opportunities and material support. Contrasting of course is the situation of Hindustani musicians, whose musical and religious identities are tied inextricably together. Chapter Three detailed the extent to which the musical journey is a spiritual path empowered by discipline, commitment, and submission to one’s guru. The extreme proximity of religion to

Hindustani music influences not only the way in which musicians philosophically conduct themselves, but from a theoretical and practical perspective, the influence also derives and expands from the tradition’s Vedic origin.

There is an important distinction to be made in the North Indian case: the practice of the Hindustani melodist should not be viewed as necessarily imitating a more religious vocal practice. Rather, in many cases, the genres in which they sing are considered to have developed historically from explicitly religious genres, and are now maintained to be religious in their own right. This contrasts to the influence of religious prototypes on Egyptian music and jazz, in which religion has influenced, to various extents, secular musical practices. For Hindustani musicians, their life as a musician is a religious practice. As Ustad H. Sayeeduddin Dagar explained with reference to his singing during a 2005 dhrupad performance in France, ‘This is Vedic way, when (we does [do]) chanting’ (H.S. Dagar, 2005). Identifying a Hindu source is particularly notable given that he is a Muslim, and he is representative of the wonderful synthesis of religions that has occurred in Hindustani classical music.

As Wade (2000) further explains, genres of Hindustani vocal music, such as dhrupad and khyāl, feature compositions whose lyrical content can be either devotional

227 or secular, or even both. What distinguishes genres is to be found more in the treatment of the subject matter, rather than the content as such. Reflecting upon the devotional aspect of compositions, Wade also suggests that the lyrical content can synthesise

Hindu and Islamic spiritual interpretations (2000, p. 173). Her analysis suggests that the distinction between the sacred and secular is not as clearly demarcated as in Western musical tradition, and as such, these forms lend themselves to a variety of expressive themes.

Improvised melodic prototypes

While it has been established that in all the traditions under examination, musicians imbibe pre-composed melodic prototypes as stylistic and theoretical blueprints, improvised prototypes can play an even more direct role in the development of improvisational ability. Chapter Four detailed the degree to which jazz musicians, both novice and experienced, deduce by ear what other musicians have improvised and then practice this material until it is memorised and internalised. Music is often also transcribing in notated form after it has been memorised to facilitate further analysis.

Egyptian musicians undertake an almost identical practice whereby taqāsīm recorded by famous improvisers are learnt by ear (Marcus 2002, p. 43), though in the Egyptian tradition these aural replications are rarely if ever transcribed in notated form. In

Hindustani music, whilst recordings are replicated by ear during individual practice time, this habit is derived from the underlying pedagogical custom of students imitating their teacher during lessons to the point of replication through antiphonal model phrase instruction as detailed in Chapter Three. Of course, in the context of changing pedagogical dynamics, where daily lessons are difficult to coordinate, students are more likely to make use of recorded information for these imitative purposes.

228 6.2.2 The role of discrete musical phrases in the development of technique related to improvisational fluency

Musicians in all three traditions practise discrete units of music tailored specifically to develop improvisational ability. These musical exercises often simulate actual situations

– for example, challenges experienced during performance – and they allow musicians to develop competence in preparation for actual performance circumstances. Commonly derived from the aforementioned melodic prototypes, and also expressly composed, the abbreviated musical phrases often allow for an ease of cyclical repeatability that enables musicians to master improvisational challenges in shorter form. When combined with other discrete phrases in practice, this facilitates the development of a more fully functional improvisational technique. Learning lengthy melodic prototypes allows musicians to internalise essential stylistic elements in complete form, and becoming proficient in these abstracted discrete musical phrases that are employed as exercises allows musicians to develop the technique that catalyses stylistically appropriate performance.

Chapter Three detailed the extent to which North Indian musicians repetitively practice tāns, tihāī phrases, and qaida. Taught in an imitative call-and-response fashion using model phrase instruction, these phrases serve the dual purpose of improving technical dexterity while also allowing musicians to internalise how these phrases are constructed conceptually. This enables them to improvise their own phrases based on these internalised concepts, using a specifically appropriate technique that has been developed by practising these abstracted prototypical examples. Furthermore, by practicing cadence-like mukhrā phrases cyclically, musicians develop competence in improvising in the space between the end of one mukhrā and the point at which the next mukhrā phrase might be introduced to the rhythmic cycle. Abstracting this performance

229 situation and repetitively practicing simulations enhances their ability to meet this challenge during improvised performances.

Moreover, Slawek notes a striking similarity between a permutation method in

Indian music known as 'merukhand' and a method developed by guitar teacher Joe Srgo, both which are designed as a means of 'exhausting all [melodic] possibilities within given parameters' (2009, p. 209), thereby developing a superior technique for

'elaborating and extending improvisations' (2009, p. 209). Furthermore, as discussed at length in Chapter Four, in the fashion of memorising the solos of other musicians and making transcriptions of these improvisations, jazz musicians have commonly taken smaller segments (known as licks) from complete solos. Abstracted from their original context, they are practised repetitively in a manner comparable to Hindustani musicians practicing tāns and tihāī phrases. Musicians are then able to repeat these phrases to the point of technical perfection. In so doing they develop not only the requisite technique, but also internalise stylistic architecture such as genre-specific mode-usage, phrase structure, details of how rhythm is employed, and importantly how these aspects function as a musical whole.

Additionally, these phrases are often transposed to multiple keys, if not into all the remaining eleven key centres. This practice aids in the development of aural ability, whilst also revealing ever-changing technical challenges as phrases are transposed into new key centres. Once these technical challenges are resolved through repetitive practice, a musician’s technique is enhanced in these new modal contexts. This multi- modal process reinforces how these phrases are constructed, which in turn develops an intuitive ability to generate novel phrases based on these principles when improvising.

Transcription and the practice of phrase abstraction have inspired the publication of improvisation anthologies for at least 80 years (Berliner 1994, p. 97). Although they

230 provide easy access to transcriptions, use of these publications may reduce aural skill development. In the absence of traditional notation in Hindustani or Egyptian music, musicians develop excellent pitch recognition skills. Perhaps jazz musicians might be better served if they restricted their use of these publications, so that they too may develop aural skills that empower improvisation. However, given that these publications give musicians easy access to solos and their extracted licks, short cutting the arduous time-consuming task of deducing the solos aurally, they do hold some value when used in a balanced manner.

Those Egyptian musicians I interviewed did discuss the use of abstracted discrete musical phrases, and my observations detailed in Chapter Five provide some evidence for this practice existing in the Egyptian context. A rehearsal session that I attended at the home of Egyptian violinist Abdo Dagher bore some strong similarities to the didactic imitative methodology employed by Hindustani teachers today, drawn from the guru-shisyā tradition. At the beginning of this rehearsal musicians aurally copied a number of discrete phrases played by Dagher. These were seemingly designed, as deduced from their repetition, to elucidate various aspects of the mode that formed the base for the composition that they would later rehearse. These phrases were played and copied cyclically until the replications met with Dagher’s approval, indicated by the introduction of a new phrase.

The phrases that he didactically exhibited to his ensemble detailed various expressive techniques: manners of ornamentation, portamenti and vibrato, as well as such subtleties as the precise tuning of various quartertone pitches. Dagher reinforced these elements throughout his repetitions so that those present could adequately imbibe these components aurally and replicate them to his liking. This is certainly comparable

231 to the model phrase instruction of discrete phrases that Alter describes as active in

Hindustani pedagogy, as analysed in Chapter Three.

Abdo Dagher’s brother Esawi also provided evidence for this method of skill development when demonstrating imitations of extracted phrases sung by vocalists.

Additionally, when teaching me, he used this as a method to communicate content and correct style, requiring me to repetitively practice certain key phrases in order to develop the correct pitch of quartertones, speed and width of vibrato, and phrasing.

Moreover, as discussed in Chapter Five, Mohammed Antar advocated a near-extinct system of practising ‘maqām melodies’: short phrases that encapsulate the essential sound elements of each maqām. By way of this repetitive method, Antar suggested that stylistically appropriate technique was internalised and embodied in a musician’s taqāsīm performances, although this specific methodology is not widely practised in

Egypt today. Table 5 summarises the comparison that has been undertaken above.

232 Table 5: The use of melodic prototypes to develop improvisational ability

Hindustani Jazz Egyptian Music Classical Music

Prototypes written sources memorised aurally also used

Composed prototypes

Improvised prototypes

Discrete phrases model phrases, used in practice and tāns, tihāī, mukhrā licks, patterns maqām melodies, performance qafla

Sacred prototypes style only; not a for style and widespread influence content

Secular prototypes n/a, the distinction for style and between sacred and content secular is not as clear

Prototypes teach style of genre

maqām usage, Prototypes rāga blueprint, model for interplay modal modulation, elucidate theory of tāla, tihāī, mukhrā, between harmony cadential figures genre tāns and melody (qafla) Use of composed prototypes as springboard for improvisation though inherent Development of system of phrase collections recombination undertaken emphasised

233 6.2.3 The role played by the intentional formation of a phrase lexicon and by the premeditation of cadential figures and other structural devices

In addition to the use of discrete phrases to develop fluent improvisational technique and to reinforce aspects of style, a comparative examination of jazz and Hindustani music demonstrates that preplanning musical material to use in improvisations is a more common procedure than is suggested by popular conceptions, which assert that such processes are incompatible with improvisation. Egyptian musicians I interviewed stated that they engage in the intentional formation of phrase collections and premeditation of cadential figures.

Dr Alfred Gamil described his personal approach to improvisation as one based on developing a storehouse of phrases. These phrases are often derived from other songs or improvisations, and they are modified to some extent through the process of repetitively practising them, a process whereby he adopts them into his personal phraseology. When improvising, the phrases serve the purpose of providing signposts within the overall structure of his performance, and they act as bridges between other novel materials.

Furthermore, premeditating the structure and placement of cadential figures known as qafla is a similar process that Gmail engages in, one that Joseph Tawadros also practices, as Chapter Five showed. Gamil’s premeditation of techniques of modal modulation has also been highlighted in Chapter Five and represents an additional practice-time activity in which solutions to improvised challenges are developed for use in performances.

Furthermore, Tawadros echoes Gamil regarding the practice of developing a musical vocabulary, but rather than identifying single phrases per se, he views the studies that he composes and practices until completely internalised as the source of material, which he says his ‘mind decides to arrange’ in response to external stimulation

234 in the moment of improvisation. As detailed in Chapter Five, Tawadros also notices that other older players, such as Esawi Dagher, similarly have a stock of phrases that they play when improvising, a spontaneous recurrence of habitual musical material learnt through repetitious rehearsal over long periods of time. Nettl and Riddle noticed a similar occurrence of motivic material throughout the taqāsīm of Ali Jihad Racy, which they speculate is a function of the tendency in Arabic music to ‘explore and develop successive pitch levels’ (1998, p. 388).

Chapter Four demonstrated that the development of a musical vocabulary for use in the jazz improvisational process has been identified by sources such as Berliner (1994) and Knox (2004). Berliner’s description of this process, ‘this involves acquiring a complex vocabulary of conventional phrases and phrase components, which improvisers draw upon in formulating the melody of a jazz solo’ (1994, p. 95), is echoed by Knox who also suggests ‘the primary element of the masterfully improvised jazz solo is the reorganisation of a vocabulary obtained from hours of transcription, transcription study, and transcription practicing’ (2004, p. 42). In my discussion of this issue in Chapter Four, I have argued that what should be emphasised is the

‘formulating’ component of Berliner’s description. That is, by developing a ‘vocabulary of phrases’, musicians develop an intuitive knowledge and corresponding ability to generate similar phrases when improvising. While memorised phrases are used as bridges and safety nets in solos, the reiteration of phrases is not the aim of developing this phrase vocabulary.

Sources examined in Chapter Three did not suggest that North Indian musicians develop a vocabulary of phrases as such. Rather, in discussing the memorisation of phrases such as tāns, research presented in Chapter Three explained that these smaller internalised phrases inspire inventiveness because they are learnt as part of a system of

235 recombination that is built upon ‘a language of formulas’ (Alter 1997, p. 81). This language of formulas, Alter asserts, ‘can be seen in the tan exercises’ (1997, p. 81), and it enables musicians to intuitively develop their own phrases in a variety of contexts when improvising. Furthermore, in later stages of the pedagogical process, by imitating a teacher’s phrases, a student’s creativity and intuition is developed because the phrases often become too difficult or complex for exact repetition. As such, a functional method of creative variation is developed alongside the already internalised language of formulas, allowing students to balance originality with adherence to traditional musical parameters. The process of variation discovered by learning the tan exercises and other discrete phrases is what is important to the improvisational process, not the memorisation and performance of learnt phrases per se.

6.2.4 The pedagogical interaction of voice and instrument: vocal imitation by instrumentalists, and vocalisation as pedagogy

In order to learn music and develop improvisational ability, Egyptian musicians

(vocalists and instrumentalists) engage in the imitation of vocalists. As detailed in

Chapter Five, vocal sounds dominate Egyptian culture, enveloping cities every morning at the sound of the call to prayer, which is repeated at intervals throughout the day, and reinforced by the mass popularity of recording artists such as Umm Kulthūm. Chapter

Five emphasised the extent to which vocalists learn their craft by frequently mimicking live and recorded singers to the point of memorisation. This influence is also witnessed in the style and technique of instrumental music. Egyptian instrumentalists that I interviewed, such as Esawi Dagher, explicitly advocated imitating vocalists as the primary method through which a musician could learn to improvise. Indeed, the vocal inflections and expressive techniques prevalent in Esawi Dagher’s playing, a style reflected in the playing of all Egyptian musicians I interviewed (especially string and

236 nāy players) attest to the practice of imitating vocal content and style as one means through which instrumentalists learn to improvise.

Conversely, the opposite trend is at work in the jazz tradition, where it is rare for instrumentalists to mimic vocal music when learning to improvise, and consequently when performing, although isolated examples exist of instrumentalists creating vocal effects in their improvisations (such as Louis Armstrong, who unsurprisingly was also a vocalist). Alternatively, scat singing is illustrative of a practice whereby vocalists imitate instrumental improvisations, commonly singing nonsense syllables to create instrument-like timbres and effects. Furthermore, North American vocalist Eddie

Jefferson developed a practice known as vocalese, adding lyrics to pre-existing instrumental compositions and improvisations, leading to the incorporation of at times virtuosic instrumental technique into his vocal performances. Mark Murphy, vocal group Manhattan Transfer, and more recently Kurt Elling are well known for continuing

Jefferson’s musical legacy of vocalists imitating instrumental improvisations. Elling’s renditions of saxophonist John Coltrane’s composition Resolution from the album ‘A

Love Supreme’ (Coltrane 1995), demonstrates an impressive attention to detail in his performance of lyrics on Coltrane’s composed melody and improvised solo (Elling

2003).

Like Egyptian singers, jazz singers do learn by imitating other jazz vocalists, although when learning to improvise, vocalists tend to derive inspiration and edification from improvisations by instrumentalists. While vocal music is an important antecedent in the jazz tradition (standards often originating as vocal songs), jazz instrumentalists do not learn to improvise by imitating vocalists as a common activity, unlike Egyptian musicians who undertake this activity in combination with others.

237 Like Egyptian musicians, Hindustani instrumentalists imitate vocalists when learning traditional music. This practice is carried into the performance arena, and it is observed in the convention of an accompanying melodic instrument echoing segments of a vocal soloist’s improvised phrases. Indeed, by accompanying a singer, instrumentalists such as sarangī or harmonium players can eventually become soloists themselves. This is an indication of the pedagogical benefit that can be gained – that is, acquiring the ability to improvise – courtesy of the imitative accompaniment of vocal soloists. This practice, like many elements of Hindustani music pedagogy, is simulated extensively when learning, as Chapter Three elucidated. Indeed, imitating the voice has been highlighted as an origin of all North Indian instrumental music (Stevens 1975, p.

37). Subsequent observations of Hindustani pedagogy have noticed imitation of the voice – that is, copying vocal renditions of compositions and improvisations – as a prominent methodology in instrumental and vocal music alike. Furthermore, it is quite common for an instrumental soloist to sing at some point in their performance and in doing so demonstrate a well-developed vocal ability. This often occurs when a non-

Indian audience is being instructed.

In addition to the discussing the practice of instrumentalists imitating vocalists singing traditional compositions and improvisations, Chapter Three outlines the way in which vocalisation as used in the pedagogy of North Indian music uniquely distinguishes the manner in which North Indian vocalists and instrumentalists develop improvisational ability. Sārgām, a syllabic mnemonic device, is used in the absence of written notation, and as such, has the additional benefit of delivering Indian musicians an empowered aural acuity.

Although Western solfège used at The Higher Institute for Arabic Music detailed in

Chapter Five is used as a means of developing aural skills, it is not used in the absence

238 of written notation, and Chapter Four did not identify this as a common practice in the jazz tradition or in modernised educational contexts. As such, the benefits derived in the

North Indian experience, particularly with regard to the development of pitch perception, indicate that this could be a potentially useful methodological transfer to modern institutions that aim to teach Egyptian music and jazz. While Egyptian and jazz musicians develop enhanced aural skills by deducing musical material by ear, this practice takes place privately. If institutions included an oral-aural form of pitch labelling in their pedagogy, the efficiency and effectiveness of aural-acuity-reliant imitative activities conducted during individual practice may be enriched. The pitch perception benefits of oral-aural pitch labelling systems are suggested by Miyazaki and

Ogawa (2006), who found that children who were taught imitatively while singing solfège syllables and playing a keyboard experienced remarkable improvement in absolute pitch testing over a four-year period.

Berliner’s account of jazz musicians singing the improvisations of other musicians before translating the sung prototypes onto instruments (1994, pp. 95–98) is similar to the Hindustani practice of instrumentalists learning by vocalising musical material.

However, the practice in the jazz tradition is neither as formalised nor as common as the

Hindustani exemplar, and perhaps most importantly, pitches are not assigned to specific syllables in the jazz practice. As such, the full aural benefit is not experienced.

Regarding imitation, instrumentalists from all traditions share in the practice of imitating other instrumentalists. As discussed at length in Chapters Three, Four, and

Five, instrumentalists mimic other instrumentalists when learning compositions and when learning to improvise. In the jazz tradition, this is encapsulated in the aforementioned practice of transcription and is additionally undertaken in live contexts.

239 As already discussed, Egyptian musicians carry out a similar practice in their traditional methods of developing improvisational ability. In addition to vocalising musical material, Hindustani teachers also train students by requiring them to imitate phrases played on instruments. However, what differentiates North Indian methods of transmission from those employed in the other traditions is the extent to which instrumental imitation is used in conjunction with vocalisation and vocal imitation.

I have established that the musicians studied in this dissertation practise numerous activities in common, the most significant being the aural imitation and internalisation of melodic prototypes and formulaic principles. Another vital point of commonality shared by all traditions is the necessity for musicians to engage in long-term, arduous, solitary practice. Slawek has also emphasised the universal importance of practice in the development of improvisational ability. In the conclusion to his comparative study of

Hindustani sitār and jazz guitar he asserts that when learning both instruments he

'learned that much depended on rigorous practice that explored repetitive exercises'

(2009, p. 216). Be it the riaz of Hindustani musicians, the woodshedding of jazz musicians, or the sequestered practice sessions undertaken by their Egyptian counterparts, this cross-cultural comparison suggests that improvised musical performance and the associated internalisation by way of aural saturation and repetitive imitation would not be possible without this important pursuit.

240 Table 6: Modes of transmission active in each tradition

Hindustani Classical Jazz Egyptian Music Music

Aural osmosis

Autodidactic Didactic teaching is a learning used to more prominent develop aspect, although some improvisational autodidactic learning ability can not be excluded Imitation of teacher Imitation of teacher Imitation of historically limited, historically limited, teacher as nor is it common in nor is it common in primary means of modern pedagogical modern pedagogical transmission contexts contexts

Imitation of live or from relevant model phrase live or from recordings, model instrument or instruction recordings phrase instruction in voice ensemble rehearsals

Instrumentalists extensive extensive imitate vocalists

Vocalists imitate scat-singing, instrumentalists vocalese

limited to informal limited to the use of extensive, singing of solfège in tertiary Vocalisation as formalised – sārgām, improvisations before training as a means of pedagogy bol transferring to teaching written instrument notation

Substantial practice required riaz woodshedding beyond transmission

241 6.3 Conclusion: how improvisational ability is developed most effectively

From childhood to musical maturity and beyond, musicians in the three cultures examined are immersed intensely in the prototypes of their musical tradition. During childhood, the informal osmotic interactions within the family environment or greater social network instigate the passive aural internalisation of genre-specific musical components and an awareness of how they interrelate to generate the characteristics of a musical tradition. The intensity of these early interactions inspires interest in music making, which leads to experimental playing and a nurtured introduction to singing or the learning of an instrument by a family member or close associate of the family. In subsequent developmental stages, the didactic guru-shisyā parampara and modern adaptations of this spiritual-pedagogical relationship and the autodidactic development of Egyptian and jazz musicians demonstrate that advanced improvisational ability may be developed either didactically, autodidactically, or by engaging in both means of teaching.

Historically, early cultivation of musical skill has been facilitated by an oral-aural mode of transmission and reception, in which the intentional imitation of composed and improvised prototypes features as the primary means through which young musicians assimilate the intricacies of their musical culture – that is, theory, style, techniques, and content. This shared imitative behaviour is reinforced and refined by extensive individual practice in all traditions examined by this thesis.

The multifaceted learning environments inhabited by musicians as they learn to improvise support autodidactic learning strategies and didactic teaching methodologies.

No matter the mode of transmission, jazz, North Indian, and Egyptian musicians develop improvisational ability principally by imitating prototypes to the point of

242 internalisation. A long-term process that begins osmotically in childhood becomes an intentional practice that is sustained throughout a musician’s career.

I assert that it is significant that imitation has remained fundamental to developing improvisational ability even though a transition towards institutionalisation has been experienced in each tradition, and irrespective of the fact that the musicians of each tradition interact in a partially oral-aural context today when compared to the wholly oral-aural context of past learning environments. It is the oral-aural nature of learning, to which imitation is central that continues to primarily facilitate the development of improvisational ability.

I highlight the significance of the aural and mimetic nature of this key developmental practice, and I assert that this is what enables musicians to assimilate stylistic parameters whilst concurrently developing the ability to play by ear anything that is reasonably perceptible. As a result, during improvised performance, musicians are able to spontaneously produce that which they conceive internally, and given the extent to which they have absorbed detailed stylistic models, internally generated phrases are imbued by genre-specific features that appropriately characterise improvisations in the image of internalised prototypes.

The cross-cultural comparison that I have undertaken throughout this thesis has not only established what musicians do to develop the ability to improvise, but by identifying the means through which this skill is acquired, I have provided significant insights concerning the development of expertise as it relates to improvisation. In the final chapter of this thesis I show to what extent these findings conform to other research concerning the development of expertise in the Western classical tradition, while also contextualising the entire thesis within a developmental psychobiological narrative.

243 Chapter Seven Correlation and Conclusions

This thesis has endeavoured to demonstrate how musicians of three genres develop improvisational ability by considering the learning environments and specific practices that characterise the process of acquiring the capacity to improvise. By way of cross- cultural comparison, this thesis has indicated that whilst there are a variety of learning environments that situate the developmental process, the incidence of imitative behaviour is overwhelmingly fundamental to the development of improvisational ability. The principal practice that the musicians of all three cultures engage in involves learning prototypes aurally. By doing so they (i) develop enhanced aural acuity, (ii) internalise the style and content of phrases that are stylistically archetypal, and (iii) develop fluent technique that is tradition-specific by repetitively practising these prototypes.

As I suggested in the conclusion of Chapter Six, highly developed aural perception ultimately enables musicians to copy with ease, as though automatic, a process colloquially known as ‘playing by ear’. Combining this skill with fluent technique and an intuitive knowledge of style and content gleaned from internalised prototypes, musicians are able to produce spontaneously self-generated music that conforms to stylistic tradition. This process is supplemented by theoretical knowledge that is intuitively derived from internalised prototypes, which are learnt autodidactically or taught. These findings describe not only the developmental process, but in doing so, also provide reasoned insights into the nature of improvisation within a tradition.

In this final chapter I will firstly demonstrate the extent to which the findings of my cross-cultural comparison conform to the most recently developed theories of expertise

244 acquisition. I will then establish that the overwhelming incidence of imitative behaviour as practised in all three traditions has a developmental origin in early childhood development. I will argue that the interactive imitation that is the basis of development from early childhood facilitates a mode of improvisation in communication and language, and as such it is logical and inevitable that imitation is the basis of developing the ability to improvise in music, given that it is an ingrained and fundamental human behaviour associated with other improvised practices. Finally I will assert that any definition of improvisation within a tradition should be informed by these findings, and this thesis will close by reflecting upon the discussion introduced in Chapter One concerning a definition of improvisation.

7.1 The acquisition of musical expertise: the pre-existing framework of expertise development

Musicians who develop mastery over the art of improvisation in the manner exhibited by this thesis cultivate abilities that catalyse expert performance. Although the acquisition of musical expertise has been the subject of much research undertaken over the last two decades, these investigations have primarily focussed on the acquisition of expert performance in Western art music. Studies in other fields of musical performance, such as non-Western improvised traditions, have been limited. Thus, I will now consider the extent to which the findings of the cross-cultural comparison conform to theories of expertise acquisition.

Ericsson, Krampe, and Tesch-Römer (1993) conducted a now widely cited study on the acquisition of expert performance. Situating their research in the realm of West

German classical violinists, Ericsson et al. aimed to redirect popular theories that viewed inherited capabilities as the principal determinant of one’s attainable performance level (1993, p. 364). The framework they established to explain the

245 acquisition of expertise centres around the role of deliberate practice. Specifically, in addition to proposing the fallibility of heritable talent theories, the writers argue ‘that the maximal level of performance for individuals in a given domain is not attained automatically as [a] function of extended experience’ (1993, p. 366), but rather, that ‘the level of performance can be increased even by highly experienced individuals as a result of deliberate efforts to improve’ (1993, p. 366).

Ericsson et al. define the focus of their research, deliberate practice, as ‘activities that have been invented with the primary purpose of attaining and improving skills’ as distinct to other activities encountered in the course of normal social interaction (1993, p. 367). Lehmann and Gruber (2006) develop an even more specific definition for deliberate practice: ‘a set of structured activities that experts in the domain consider important for improving performance’ (p. 459). Contextualising their investigation within pre-existing theories of learning and improvement, Ericsson et al. (1993) cite

Bower and Hilgard (1981) and Gagné (1970) who characterise circumstances for

‘optimal learning and improvement’ as existing where a subject is driven by motivation to undertake the required effort, in a context where prescribed tasks take into account pre-existing knowledge, in which immediate feedback regarding performance is received, and where repetition of the same or similar tasks is undertaken (1993, p. 367).

Accordingly, a framework that explains ‘expert performance in terms of acquired characteristics resulting from extended deliberate practice and that limits the role of innate (inherited) characteristics to general activity and emotionality’ (Ericsson et al.

1993, p. 363) is proposed with the intention of explaining ‘how differential levels of performance are attained as a function of deliberate practice’ (Ericsson et al. 1993, p.

368). More specifically, the theoretical framework of their research – derived by reviewing literature that has investigated the acquisition of expertise in an array of fields

246 as diverse as sport and chess – is that the amount of deliberate practice undertaken by an individual is monotonically related to their acquired performance, it extends for at least ten years, and it is optimised within several constraints, including time, energy, access to teachers, training materials, and adequate resources. Furthermore, summoning requisite motivation to undertake effortful long-term practice, whilst balancing practice time with rest to allow for recovery and renewal, are also identified as conditions which limit the maximisation of deliberate practice (Ericsson et al. 1993, p. 369).

From this basis the framework proposed by Ericsson et al. (1993) to explain the acquisition of expertise is structured around an investigation of how these constraints are navigated and resolved by individuals who eventually attain expert-level performance. Initially relying on Bloom’s (1985) three-stage portrayal of the preparatory journey that successful individuals negotiate, Ericsson et al. summarise the first stage as being characterised by playful experimentation and recognition of talent, ending with the commencement of lessons and deliberate practice within a structure assisted by parental encouragement.

The second phase is distinguished by an ‘extended period of preparation’ and ends when individuals makes a ‘commitment to pursue activities in the domain on a full-time basis’. The third stage of development witnesses the commitment made at the end of the second stage being realised through long-term deliberate efforts to improve, and ends when individuals are able to pursue a professional career in the domain, or alternatively, when they cease to be involved on a full-time basis (1993, p. 369). Ericsson et al. conclude that this framework needs to be extended to ‘accommodate eminent performance’, by including a fourth phase in which ‘the individuals go beyond the knowledge of their teachers to make a unique innovative contribution to their domain’

(1993, p. 369). This stage is signified by the achievement of eminent performance, as

247 opposed to expert performance where individuals achieve mastery over existing knowledge and current performance standards (Ericsson et al. 1993, p. 392).

Within this framework, individuals who successfully advance from one phase to the next do so because they are able to overcome constraints to deliberate practice. Ericsson et al. suggest the first constraint to overcome is that of resources. Specifically related to musical endeavours, individuals must have access to musical instruments, lessons with teachers or masters, learning resources, and physical practice space. The role played by parents in providing solutions to these problems is asserted to be vital (Ericsson et al.

1993, p. 370) and is motivated by a belief that their child is talented, even though at times no evidence may actually exist to justify this premise (Bloom 1985, referenced by

Ericsson et al. 1993, p. 370).

The second constraint is that of effort. Fundamentally, individuals must summon the effort required to undertake the arduous amounts of practice required of them over the decade or longer period of commitment. Ericsson et al. assert that in order to maximise expertise, deliberate practice, effortful as it is, must also be maximised (1993, p. 370). While studies highlighted by Ericsson et al. identify notional amounts of practice as effective daily maximums, the general conclusion offered is that short-term practice needs to be balanced with periods of recovery, and over the long term, should be increased slowly to allow for individuals to adapt to the increasing demands. As such, the effort constraint placed on deliberate practice is theoretically negotiated (1993, p. 371).

Closely linked to effort is motivation. At the crux of the theoretical framework proposed by Ericsson et al. is that

deliberate practice is not inherently enjoyable and that individuals are motivated to engage in it by its instrumental value in improving performance. Hence, interested individuals need to be engaging in the activity and motivated to improve performance before they begin deliberate practice (1993, p. 371).

248 Reflecting back to the first developmental phase explicated by Bloom (1985), playful experimentation motivates the commencement of deliberate practice, at which time parents offer the structure required to maintain deliberate practice while also continuing to emphasise the motivating connection between practice and improvement (Ericsson et al. 1993, p. 372).

Over time, often with the aid of teachers, individuals are able to learn ways to assess their own improvement, concomitantly observing the effectiveness of their practice, thereby allowing them to maintain motivation in the midst of arduous effort.

Within this feedback loop, the role of activities such as public performances and competitions provide important short-term goals that also motivate specific areas of improvement (Ericsson et al. 1993, p. 372). Even more specifically, Renwick and

McPherson (2002) illustrated that when children were motivated by a piece of music that inspired them, they practised in a far more complex manner, employing various advanced techniques of mental representation, than when they were uninterested.

Motivation is paramount for successful skill acquisition, and even more essential to the arduous acquisition of expertise.

Ericsson et al. assert that by later stages of expertise acquisition, motivation to undertake deliberate practice becomes so intrinsically connected to an individual’s desire to attain expert level performance that motivation to practice becomes difficult to assess due to the pronounced extent that the pursuit of expertise has become intertwined with their life (1993, p. 372). Finally, Ericsson et al. make the important assertion that without the goal of improving performance, ‘Many individuals who have practiced for a long period of time give up their aspirations to compete and excel in an activity’ and accordingly cease practicing due to waning motivation (1993, p. 372).

249 After establishing this general framework, as derived from literature investigating expertise acquisition in a variety of fields, Ericsson et al. (1993) then determined the extent to which this theoretical framework could explain the development of musical expertise as observed in West German classical violinists. At the outset they predicted that their study would reveal that the developmental progression of classical musicians to expertise would conform to the generalised framework.

Specifically, they anticipated that the level at which musicians currently performed would be directly related to amounts of past practice. Furthermore, they predicted that musical expertise cannot be attained with less than ten years of deliberate practice, and that this practice begins at low levels and slowly increases over the decade or longer period of development (1993, p. 372). Furthermore, they predicted that the highest amount of weekly deliberate practice would be related to both the highest level of improvement and to the highest level of performance, and such practice would be arduous, undertaken over the long term at a constant rate, balanced with rest for recovery, and would not be perceived as being inherently enjoyable (1993, p. 372).

In summary, the study by Ericsson et al. (1993) provided evidence for these predictions, whilst finding no support for the hypothesis that innate musical talent is that which determines an individual’s level of performance. Moreover, the writers suggest that in addition to deliberate practice, what is essential is that an individual’s parents view them as talented. This determines whether or not crucial support is directed towards an individual when attempting to overcome initial constraints and undertake progress towards expertise (1993, p. 399). Ironically, while disproving the notion of innate talent as that which determines an individual’s eventual rise to expertise, the study does demonstrate that the myth of inherited giftedness is that which promotes early steps towards expertise acquisition, as belief in this myth seems to be a prime

250 motivation to parental support. Without this belief, many parents may not be prepared to invest of their time and resources.

Significantly, asserting the direct relationship between deliberate practice and expertise acquisition, Ericsson et al. acknowledge that

It is quite plausible however, that heritable individual difference might influence processes related to motivation and the original enjoyment of the activities in the domain and, even more important[ly], affect the inevitable differences in the capacity to engage in hard work (deliberate practice) (1993, p. 399).

Thus, while specific musical talent has been dismissed by Ericsson et al. (1993), heritable differences have been hypothesised that relate both to an ability to be motivated and a capacity to undertake arduous effort. This caveat leads the researchers to assert that their findings do not suggest that ‘everyone can easily attain high levels of skill’ (1993, p. 400), concluding that the high level of skill that experts demonstrate should not be understood as a capacity to excel only in a specific paradigm, but rather, experts should be understood as having a capacity for maintaining high levels of deliberate practice. They are experts at improving performance (1993, p. 400).

These findings of Ericsson et al. (1993) have been clarified in research by Sloboda et al. (1996), whose study also emphasised the importance of relevant practice and the role of parents in the acquisition of musical skill acquisition. Further developing the research by Ericsson et al. (1993), Lehmann and Ericsson (1997) and more recently

McPherson (1997 and 2005) have focussed their attention on the mental representations and cognitive strategies that characterise skill development in expert musicians, applying them to discussions concerning the education of amateur musicians and music students.

Extending from the conclusions of Ericsson et al. (1993), Lehmann and Ericsson

(1997) confirm that the physiological, cognitive, and psychomotor characteristics of

251 experts, considered for many years to be inherited characteristics, have been shown to be the result of individual adaptations brought about by long-term deliberate practice

(1997, p. 45). Indeed, Lehmann and Gruber (2006) later stated that ‘We can safely assume that music training and practice leads to substantial functional and structural changes in a person’s brain and consequently alters processing capabilities’ (2006, p.

465).

From this evidence base, the emphasis of Lehmann and Ericsson’s research (1997) focuses on the mental representations and strategies that individuals summon to enable growth towards expertise during periods of deliberate practice. Employing a triangular model of mental representation, Lehmann and Ericsson configure a three-way relationship between (i) the ability to mentally represent the desired performance goal,

(ii) the ability to implement this represented goal, and (iii) the ability to monitor current performance, assessing the extent to which the target has been reached (1997, pp. 50–

51).

Within this cycle, maintaining the level of concentration required to be able to undertake these mental representations is an essential aspect of deliberate practice. In early stages of development, when a child’s ability to stay focussed is more limited, the supervision of an adult, who is not necessarily a skilled musician, is identified by

Lehmann and Ericsson as an important aspect of the social infrastructure that helps children maintain concentration during periods of deliberate practice (1997, p. 51). This is also emphasised by Lehmann and Gruber (2006, p. 461).

Moreover, Lehmann and Ericsson emphasise that by teaching individuals in a manner that includes them in the process of goal setting and performance monitoring, an effective feedback loop involving self-regulation can be internally instigated (1997, p. 52). As such, Lehmann and Ericsson affirm that this interplay between internalised

252 mental representations allows individuals to maintain motivation because self- monitoring enables them to notice the positive effect that short-term effort has on long- term performance goals (1997, pp. 54–55). Accordingly, the writers assert that understanding the mental representations of experts is all-important to understanding the acquisition of expertise, given the essential nature of motivation in maintaining the arduous sessions of deliberate practice (1997, pp. 54–55).

The narrative presented above details a framework that explains expertise acquisition both in general terms and specifically concerning musical endeavour in

Western classical music. At the outset of this chapter, I outlined the limitation that whilst this research provides conclusive statements regarding the acquisition of expertise in music – specified in terms of the relationship between deliberate practice, feedback, motivation, and performance – what remains unknown is the extent to which this model of acquisition applies to non-Western improvised traditions, such as those cross-culturally compared by this thesis.

7.2 The development of expertise in the jazz tradition, Hindustani classical music, and Egyptian music

As an introduction to a forthcoming discussion that will determine the extent to which those traditions addressed by this thesis conform to the model proposed by Ericsson et al. (1993), I will first examine research by Gruber, Degner and Lehmann (2004) that addresses a similar question with regard to expertise learning in jazz guitar music. At the outset of their study, Gruber et al. (2004) slightly reconfigured the context in which deliberate practice and expertise acquisition had been understood prior to their research, stressing that ‘Acquisition of expertise can be viewed as a process of enculturation and of becoming a full participant in a community of expert practice’ (2004, p. 223). Within this process of enculturation, they assert expert teachers play a vital role in sharing ‘a

253 body of organised experience’ with newcomers as they are able to predict future skill demands (2004, p. 224), thereby supporting deliberate practice.

Through their analysis of the learning processes employed by jazz guitarists, and comparing this data with the model constructed by Ericsson et al. (1993), Gruber et al. determined that deliberate practice played a similarly central function in the development of expertise in jazz. However, they assert that the definition of deliberate practice needs to be expanded to incorporate both individual and group practice sessions, as well as practice that was self-directed, rather than teacher-led (2004, p.

224). Significantly, they also discovered that jazz musicians enjoyed their practice to the extent that they could not stop (2004, p. 232). Thus, the insistence that deliberate practice is perceived as anything but enjoyable (Ericsson et al. 1993) is actually contradicted by the expanded context created by Gruber et al. (2004).

Waler (1992) asserts that the ideology that depicts practice as a rigorous, arduous responsibility in which ‘[t]he extreme result of this set of ideological values is the complete withdrawal of the musician from his or her public’ (1992, p. 298) is fundamental to the developmental narrative of classical music. Reporting on virtuosic heavy-metal guitarists, Walser correspondingly identifies the model of practice that they started to adopt in the 1980s, ‘stressing rationalisation and technical rigour’ (1992 p.

293), as being derived directly from a classical model. Walser goes further, affirming that virtuoso Swedish guitarist Yngwie Malmsteen’s statements accentuating the personal sacrifices that he has made in aid of his musical career, are influenced directly by a ‘model of music making promulgated in classical music’ (1992, p. 298). It is clear that this ideology of practice, interpreted as being inherently not enjoyable by Ericsson et al. (1993), is derived from a narrow definition of enjoyable, possibly influenced by the Protestant work ethic that pervades Western attitudes towards personal sacrifice.

254 Expanding upon Gruber et al. (2004), who alternatively assert that jazz musicians find practice enjoyable, I will address exactly how the notion of practice as being inherently unenjoyable is incompatible within the Hindustani classical and Egyptian musical context. This explanation will be presented below.

Before doing so, returning to a jazz focus, Gruber et al. assert that ‘The important role of teachers for achievement of expertise (setting goals, error correction, and instruction) in the domain of classical music is far less important in the domain of jazz’

(2004, p. 231). Correspondingly, Gruber et al. (2004) view the kind of autodidactic behaviour that I emphasised in Chapter Four whereby learners copy recordings, as imitating a teacher’s behaviour. However, as I outlined in Chapter Four, Berliner (1994) did not emphasise that autodidactic learning was a replacement for teaching, but rather stressed that the tradition has always placed the responsibility of learning squarely on the novice musician in a traditionally non-didactic learning environment.

From either perspective, when replicating recordings, individuals look for differences between their own playing and that of the recorded expert, and make necessary modifications (Gruber et al. 2004, p. 226). As such, recordings and transcribed solos, labelled as ‘virtual musicians’ by Gruber et al., allow learners to make use of a self-regulated ‘corrective feedback’ framework without the presence of a teacher (2004, p. 227). The feedback loop mediated by a teacher in other traditions is self-mediated with the assistance of a recording. Importantly, Gruber et al. (2004 p.

232) assert that recordings not only serve as the framework through which learners set their own goals, but given that they are chosen for analysis by the learner on the basis of interest, they serve the dual function of supporting motivation to learn, in a similar manner as discovered by Renwick and McPherson’s research alluded to above (2002).

255 7.2.1 Starting age and the role of teachers

Extending beyond the specific jazz guitar focus of research conducted by Gruber et al.

(2004), I will now examine the extent to which the data presented in my analysis of skill acquisition in the jazz, North Indian, and Egyptian traditions, conforms to the findings of Ericsson et al. (1993). Before doing so, I will raise the qualification that my research differs in both methodology and intention to Ericsson et al. (1993). That is, the very specific focus that Ericsson et al. (1993) have taken in their study of West German violinists is contrasted by my broad cross-cultural comparison that focuses specifically on the acquisition of improvisational ability. Even so, useful conclusions can be drawn by contextualising my findings within the framework established by Ericsson et al.

(1993).

The first consideration that I will highlight in this evaluation pertains to the issue of starting age of deliberate practice. In Gruber et al. (2004) it is emphasised that jazz guitarists generally started deliberate practice at 13, later than those classical musicians featured in Ericsson et al. (1993) who were found to begin at eight years of age on average. My findings outlined in Chapter Four suggest that although the experiences of jazz musicians are first motivated during early childhood by inspiration encountered in the family environment and greater social network, thus conforming to the aforementioned continuum starting point proposed by Bloom (1986), deliberate practice does appear to start later.

The homogenised pedagogical structure underpinning the educational experiences of those classical violinists studied by Ericsson et al. (1993) may account for why these musicians demonstrate a tendency to start by eight years of age, which coincided with the age of starting lessons (Ericsson et al. 1993, p. 374). This phenomenon may also be encouraged by the mystique of the child prodigy that has inspired the education of

256 classical students since the era of Mozart (Kivy 1967). The contrasting lack of pedagogical conformity characterising the multifaceted experiences of jazz musicians featured in Berliner’s detailed survey (1994) suggests that the age that jazz musicians begin the essentially autodidactic process that distinguishes the manner in which they acquire expertise is substantially variable.

Secondly, learning a classical Western instrument inevitably requires the guidance of a teacher who encourages deliberate practice to begin (Ericsson et al. 1993), whereas the tendency for jazz musicians is to progress without regular teaching, combining sporadic periods of lessons with self-motivated study in which the role of a teacher is superseded by self-regulated imitation of recordings (Berliner 1994, Gruber et al. 2004).

This is not to mention the often-overlooked fact that unlike Western classical music where the educational focus starts with learning an instrument, the focus in jazz, as well as in Hindustani classical and Egyptian music, starts with learning a repertoire, which leads to musicians commonly learning multiple instruments and also singing.

Consequently, granting importance to a specific starting age that a musician begins practising the instrument on which they later specialise may not be as relevant in these traditions as it is in the Western classical tradition.

As such, a regular pattern concerning the starting age of deliberate practice cannot be derived from the jazz tradition, though musicians generally begin pursuing deliberate practice of jazz performance later than their classical counterparts given the level of maturity required to undertake self-motivated study (Gruber et al. 2004 p. 231).

Whilst I have asserted above that deriving an average starting age may not be as relevant for Egyptian music, I will now investigate the possibility further. The trend in the Egyptian experience, as portrayed in Chapter Five of this thesis, is for early childhood interest inspired by the family environment to lead to experimentation, often

257 followed by some kind of instruction being given by a family member. My research suggests that the musicians I interviewed generally started pursuing music making, on average, by the age of eight.

Specifically, those musicians I interviewed, such as Esawi Dagher, began playing the violin at the age of ten, starting his musical pursuit playing the ‘ūd before making the common transition to violin. Esawi Dagher asserted that his older brother Abdo

Dagher played as a professional by the age of seven and had become a great player by twelve years of age. However, a published biography modifies the younger brother’s claim, asserting that Abdo Dagher played the ‘ūd from the age of seven, and was inspired to play the violin from the age of ten. Officially, his professional career started at 13 (Dagher 2000). In any case, assuming these details of child prodigiousness are not exaggerated, supporting oneself by playing music from the age of 13 is exceptional, and the elder brother’s status as the nationally respected leader of Egyptian violin indicates his accepted uniqueness. Indeed, referring to the discussion of ‘eminent’ performance in

Ericsson et al. (1993), Abdo Dagher has certainly made a unique contribution to the field of Egyptian violin playing, going beyond existing standards of performance. As such, in the framework established by Ericsson et al. (1993) he would be considered to have gone beyond expert performance, attaining eminent performance, evidenced by his exceptionally early progression to expertise.

Other respondents also started at an early age. Nāy player Mohammed Abdula

Fouda began imitating his father at the age of seven, whereas Ahmed El Haggar was taught by imitating his grandfather and father from the age of three. Also an early beginner, Mohammed Antar began piano lessons aged four before progressing to Arabic performance on the nāy. Starting later, Dr Alfred Gamil, while not coming from a musical family, expressed an intense interest in music before studying formally for

258 seven years from the age of 14. Similarly, Mahmoud Bedir began ‘cello aged 14, taking lessons for six years, after which time he continued in an autodidactic fashion learning to perform in a traditional Arabic style in his twenties. Thus, while a large range of starting ages can be derived from my research, and while musicians often learn more than one instrument, on average my respondents started practising in their domain at 8.4 years of age, broadly conforming to the findings of Ericsson et al. (1993).

However, mirroring the experiences witnessed in the jazz tradition, the role of regular didactic teaching in Egyptian music is limited, and the practice of educating oneself from recordings or in live settings, observing and imitating the performances of experts, is the most prevalent form of skill acquisition. Even in situations where family members teach, as in the case of Ahmed El Haggar and Mohammed Abdula Fouda, imitation is still the prominent mode of transmission, contrasting with the more prescriptive mode of teaching witnessed by Ericsson et al. (1993).

Chapter Four has detailed the extent to which the North Indian classical tradition is characterised by a formalised pedagogical system underpinned philosophically by the traditional guru-shisyā parampara. Acknowledging that guru-mediated teaching experiences of the past are nowadays less common, Chapter Four asserted that effective modernised pedagogical adaptations are not only still distinguished by respectful interpersonal interactions between student and teacher, but the imitative mode of transmission whereby students mimic their teacher is still recognised and practised as the most effective pedagogical medium.

In the historical guru-shisyā parampara, children would be accepted by a guru at around eight years of age after showing interest and normally receiving preliminary instruction in the home environment from a family member (Shepherd 1976, p. 52).

This traditional starting age conforms to the trend discovered in classical violinists by

259 Ericsson et al. (1993). However, even in the traditional scenario that is documented by

Shepherd, shisyā who were not from musical families sought a guru in their teens or even later in life (1976, p. 51). Thus, while children have traditionally begun musical pursuits at an age corresponding to the framework of expertise acquisition proposed by

Ericsson et al. (1993), variations have always existed.

However, what is undeniably emphasised in both traditional and modernised settings is the dominant role of teachers in the pedagogical process. Representative of the ultimate prototype and essential to the organised dissemination of stylistic information, the guru as expert teacher guides students through the process of skill acquisition, setting goals, giving feedback, and designing the details of deliberate practice. In this way, the teachers of North Indian classical music, like the teachers of

German classical violin students studied by Ericsson et al. (1993), support students in acquiring the skills and knowledge that enable them to become part of a community of expert practice.

7.2.2 The role of deliberate practice

As determined by the cross-cultural comparison presented in Chapter Six, an aspect of the skill acquisition process that musicians in every tradition analysed by this dissertation share is the role that individual practice plays in the development of improvisational ability. The term ‘deliberate practice’, first coined by Ericsson et al.

(1993) and introduced at the beginning of this chapter, can equally be used to describe the activities and methods that are repetitively practised by those musicians investigated by this thesis.

Unique to the experiences of jazz and Egyptian musicians is the autodidactic nature of this practice. Ericsson et al. (1993) describe practice in classical pedagogy as being

260 initially expertly managed by master teachers, whereas the practices of jazz and

Egyptian musicians, as described in Chapters Four and Five, in spite of the best efforts made to formalise them, are overwhelmingly and characteristically self-designed phenomena. Musicians choose the content of their practice time from traditional methods (such as learning compositions or transcribing recorded improvisations), design exercises individualised to their current educational needs, and simulate performances by pre-meditating a vocabulary of strategies that eventually provide technical capacity and inspiration during actual improvised performance. In contrast, and as described in Chapter Three, the riaz (practice) undertaken by North Indian musicians, like the classical violinists of Ericsson et al. (1993), is guided by their teacher’s example and recommendations.

Returning to my earlier discussion regarding the perception of practice, in which I highlighted the fact that the findings of Gruber et al. (2004) were contrary to Ericsson et al. (1993), reporting that jazz musicians found practice enjoyable, due in part to their practice being a self-designed autodidactic activity, I now turn the focus of this discussion towards the perception of riaz in Hindustani classical music.

As introduced in Chapter Three, Shankar discusses practice in Hindustani classical music with reference to the concept of sadhana, which he describes as ‘practice and discipline, eventually leading to self-realization. It means practicing with a fanatic zeal and ardent dedication to the guru and the music’ (1968, p. 22). In a sense, the ability to practice in a zealous manner is facilitated in part by the ability to be lost in the most repetitive of exercises (Pandit Ashok Roy, personal correspondence with Dr John

Napier, 13 March 2012). The correlation of musical training to spiritual practice consequently deems any negative perceptions inappropriate in the Hindustani context.

261 As such, the assertion made by Ericsson et al. (1993), that deliberate practice is not perceived as inherently enjoyable, does not hold within the Hindustani context.

Modes of learning in the Egyptian tradition, whilst being demanding and involving rigorous individual practice, are also supplemented with remarkably social modes of learning. Those rehearsals that I observed or took part in, as detailed earlier in this dissertation, were formed by a group of friends and acquaintances gathering in the leader’s home for refreshments and social discussion. This relaxed manner did not detract from the quality of musical rehearsal, and it provided a welcoming forum in which to work on skills developed in individual practice sessions. Moreover, in my

Cairo-based research, respondents did not emphasise negative attitudes towards individual practice. Quite the opposite: Alfred Gamil declared an enjoyed addiction to practice, and Joseph Tawadros told me he would often practice rather than do anything else. The autodidactic nature of self-designed practice sessions, as in the jazz tradition, results in an additional level of enjoyment; musicians own the activity. Thus, as in the

Hindustani and jazz contexts, the supposition made by Ericsson et al. (1993), that practice is not perceived as being enjoyable, does not hold for the Egyptian framework.

7.2.3 The ‘ten-year’ rule

Central to the framework proposed by Ericsson et al. is that expert performance is ‘the product of a decade or more of maximal efforts to improve performance in a domain through an optimal distribution of deliberate practice’ (1993, p. 400). Ericsson et al. correlate expertise with ‘international-level performance’ (1993, p. 366), and recognise it as being reached when individuals attain a level of ‘adult elite performance’ (1993, p.

373). Their research concludes that elite performance takes at least ten years of deliberate effort to improve, where deliberate practice is optimally distributed

262 throughout (Ericsson et al. 1993, p. 400). While I have suggested that teachers play a varied role in the attainment of expertise in those traditions featured in this dissertation, deliberate practice is the unifying aspect in my cross-cultural research, conforming to the conclusions of Ericsson et al. (1993). Still to be determined is the period of time through which deliberate practice is distributed in the traditions central to this dissertation. Although my research does not include specific numerical analysis regarding the age that musicians attain international-level elite adult performance, it is possible to develop an understanding of the length of time musicians engage in efforts to improve through deliberate practice towards elite adult performance, as long as it is recognised that this assessment is by its nature, not as specific as the data offered by

Ericsson et al. (1993).

Beginning with my review of North Indian music pedagogy in Chapter Three, it is clear that children have traditionally pursued deliberate practice under the care of a guru from eight years of age. Given that their progression to solo performer would rarely occur before adulthood, it is a fair assessment to suggest that they would have engaged in deliberate practice for at least a decade, thus conforming to the timeframe established by Ericsson et al. (1993). In fact, I emphasise that in the Hindustani context, training would almost certainly last for more than a decade. Shankar suggests that only the

‘elementary training, for a talented and persevering student, should last not less then

[than] five years’ (1968, p. 13), indicating that it would be highly probable that more than a ten-year period would be required for expertise to be reached. Napier recalls that his teacher completed five years studying voice and five years on tablā before even beginning his training on the sarod, adding that his teacher would have been horrified at the suggestion that he had acquired enough training by this stage in his development

(personal correspondence, 12 March 2012).

263 As Chapter Three established, the student body populating effective modern adaptations of this traditional Hindustani pedagogical arrangement are young adults.

The musical background of these students could of course include preparatory years of music making. However, given the negative reviews of these general music institutions highlighted in Chapter Three, it is likely that this preparatory study would not be characterised by the same intensity of later study, nor would it be as rigorous as the traditional guru-shisyā parampara. Thus, in modern adaptations of the North Indian pedagogical model, the length of time that musicians must undertake deliberate practice in order to achieve expertise is unclear. However, given the complex and technically demanding nature of this tradition, combined with the fact that shisyā historically engaged in at least ten years of deliberate practice under the guidance of a guru, it is highly unlikely, based on the conclusions regarding elite performance in Ericsson et al.

(1993, p. 400), that expertise could be developed in North Indian classical music with less than a decade of distributed deliberate practice.

Egyptian musicians featured in my research, who began musical pursuits on average at 8.4 years of age, undertake substantially self-regulated deliberate practice for at least a decade. Recognition of expertise may be given to young musicians such as

Abdo Dagher, though the cultural tendency that I witnessed in the rehearsals I attended in Cairo, detailed in Chapter Five and Six, is for younger musicians to adopt an inferior relational status with older master musicians until such time that their ability is recognised as adults. As a result, the time span between the average starting age of 8.4 years to adulthood conforms to the ten-year rule highlighted by Ericsson et al. (1993).

As established earlier in this Chapter, Gruber et al. discovered that jazz guitarists started practising in their domain later than their classical counterparts, on average at 13 years of age (2004, p. 231) rather than at eight (Ericsson et al. 1993). While musicians

264 may begin playing an instrument in early childhood, as commonly exposed in Berliner’s research (1994), the progression to practising jazz, by necessity, comes only after a general facility has been developed on an instrument, and a solid foundation of musicianship has been established. While seeming to indicate that their developmental pathway does not conform to the ten-year rule advocated by Ericsson et al. (1993),

Gruber et al. (2004, p. 232) remind readers that early careers are less frequent in the jazz tradition, thereby permitting the essential decade or more of deliberate practice to take place, partially during adulthood.

7.2.4 Motivation

As outlined earlier in this chapter, Ericsson et al. (1993) emphasised that remaining committed to deliberate practice through the decade or longer period required to attain expert-level performance requires sustained motivation, both from individuals and from those people supporting their pursuits. The writers identified the important role of initial interest and parental support in triggering original motivation, asserted the role that teachers and parents play in helping individuals notice the positive effects of their practice on improving individual performance, and highlighted the importance of public performances and competitions in maintaining motivation to practice (Ericsson et al.

1993, pp. 317–372). Assuming that motivation plays an equally important role in the traditions featured in this dissertation, I will now assess how this feedback loop manifests itself, instigating and maintaining motivation in these traditions.

Childhood interactions in the jazz tradition, discussed at length in Chapter Four, were also stimulated by parental support in the home environment. The church and school often provided the first performance opportunities, reinforcing early motivation to remain involved in music making and the related practice. Once musicians actively

265 pursue the deliberate practice of jazz, the jam session has traditionally motivated individuals to maximise deliberate practice through the peer-mediated atmosphere characterising these at times intensely competitive sessions. This competitiveness has often been fuelled by the threat of sanction, resulting in exclusion from the current or future jam sessions, as well as related socio-musical embarrassment.

In modern tertiary education settings, the interactive nature of jam sessions has been replaced in part by group improvisation classes. Although these differ in that a teacher mediates them, the comparative assessment that takes place between individuals provides a similar element of competition, albeit in a simulated performance context.

Memorising and transcribing recordings has been highlighted in this chapter as a jazz activity that replaces the role played by teachers in classical music. Comparing one’s imitation of a recorded master’s solo to the original recording provides feedback and inspires motivation when musicians attain a successful likeliness to their self-chosen prototype.

Gruber et al. (2004) suggest that because individuals tend to make the decision to specifically pursue jazz later in life than classical musicians, whose studies tend to be instigated by parents, jazz students tend to make a deliberate decision based on preference. This decision extends to the instrument they will study. While the parents of classical musicians often choose the musical instrument that their child studies, jazz musicians were found to be old enough to make a ‘deliberate decision caused by interest’ (Gruber et al. 2004, p. 231).

The writers suggest that fascination with the idiom and one’s instrument ‘seems to be a more important factor than formal instruction in the beginning of the career’ and in fact discovered that experts all identified fascination as the reason they started playing jazz (Gruber et al. 2004, p. 232). The motivation inspired by this fascination results in

266 intense periods of practice. Correspondingly, given that it is common for musicians to transfer from playing other styles of music to learning jazz, the instrumental dexterity this experience provides aids in the quickening of their progress and maintenance of their motivation. Thus, fascination aided by pre-existing instrumental technique is a central aspects of motivation in the jazz tradition, and it relies on the maturity that is afforded by the later age that musicians tend to actively purse jazz, closely linked to the autodidactic nature of their progression to expertise.

As Chapter Three has emphasised, the philosophy and spirituality embodied in the guru-shisyā parampara underpins the pedagogy of North Indian classical music. The relational dynamics that this relationship accentuates still shape interactions between students and teachers in effective modern adaptations of the traditional mode of transmission. In traditional settings, motivation to engage in deliberate practice was often initially derived from experiences in the family home during early childhood.

Early teaching from a family member normally followed, during which time support and encouragement fostered further motivation. By the time a guru would take over the training of the young disciple, the obedience and reverential behaviour dictated by the guru-shisyā parampara, and fostered by in situ training, provided shisyā with an effective structure through which to maintain their motivation to improve, at all times sustained by an underlying spiritual commitment.

Egyptian musicians detailed in Chapter Five, similarly initially inspired and encouraged by familial musical interactions, are sustained in their efforts by a community of practice that is experienced in group rehearsals mediated by master musicians such as Abdo Dagher, or by virtual masters on recordings. Imitating recordings of famous improvisations, as in the jazz tradition, also sustain a musician’s motivation to persevere. As in the jazz tradition, attaining a successful imitation of a

267 chosen master’s playing motivates the next stage of learning. Egyptian-Australian ‘ūd virtuoso Joseph Tawadros, whose progression to professional practice was outlined in

Chapter Five, was motivated almost solely within a simulated world in which he learnt by ear and imitated traditional recordings. He prospered in a virtual world constructed upon a foundation made by his mother’s collected recordings, and supported over distance by his Egyptian musician relatives. A strong desire to represent his Egyptian heritage in Australia as a self-appointed ‘ambassador of the ‘ūd’ fostered his motivation.

Operating ‘on location’, this familial-cultural pride as the centrepiece of personal motivation was also witnessed in my interactions with Egyptian musicians in Cairo, of whom the vast majority were continuing a long family tradition of musicianship. Be it the Dagher brothers, the El Haggar family of vocalists, or nāy virtuoso Mohammed

Fouda, the family links outlined in Chapter Five foster beliefs in inherited talent, and thereby motivate musicians to remain committed to years of deliberate practice.

Similarly, Ericsson et al (1993) assert that while there was no evidence for an inherited capacity for music making, the belief in this innate talent is vital in the initial stages of music making to sustain motivation. So too are the familial-cultural bonds that join and sustain generations of Egyptian musicians.

7.3 The developmental origin of imitative behaviour

As this dissertation has progressed, the incidence of imitative behaviour, as practised in the learning environments of those traditions that I have compared cross-culturally, has emerged repeatedly as a learning tool of great significance, and as such, has been emphasised as a learning phenomenon of principal importance. In the ensuing paragraphs I will demonstrate that the prevalence of this imitative methodology, while having a significant purpose in facilitating improvisation, also has a profound origin in

268 human behaviour that begins at birth. Firstly, this developmental provenance provides a behavioural explanation for why imitation is the central methodology discovered by this thesis. Secondly, details of neonatal and infant development exposed subsequently will suggest that the established correlation between imitative behaviour and improvisation is an interdependent relationship that also begins soon after a child’s first breath.

This developmental narrative is carefully constructed in Communicative Musicality

(Malloch and Trevarthen 2009), a collaboration that draws upon research primarily in psychology, psychobiology, cognitive science, musicology, and behavioural sciences.

The collected work emphasises that an underlying musicality governs early childhood communication between parent and infant, a process in which mimetic behaviour is at the core.

Bannan and Woodward (2009, p. 468–469) remind readers that auditory consciousness begins in utero where sounds of the mother’s internal organs combine with external stimuli to provide a diverse array of auditory experiences, the influence of which are evident after birth when infants recognise sounds that were originally experienced prenatally. As babies develop, Malloch and Trevarthen assert that a reciprocal and imitative ‘communicative musicality’ is intuitively enacted between mother (parent) and child from these neonatal foundations, formulated in terms of

‘pulse, quality and narrative’ (Malloch 1999, in Malloch and Trevarthen 2009, p. 4):

‘Pulse’ is the regular succession of discrete behavioural events through time, vocal and gestural, the production and perception of these behaviours being the process through which two or more people may coordinate their communications, spend time together, and by which we may anticipate what might happen and when it might happen. ‘Quality’ refers to the modulated contours of expression moving through time … timbre, pitch, volume – or attributes of direction and intensity of the moving body … Pulse and quality combine to form ‘narratives’ of expression and intention. These ‘musical’ narratives allow adult and infant, and adult and adult, to share a sense of sympathy and situated meaning in a shared sense of passing time (2009, p. 4).

269 Continuing from this interaction that is musical, of particular interest is the manner in which mothers communicate with their babies. Research

has revealed a distinctive melodious way of expression – described as ‘motherese’ (or more impersonally, ‘infant directed speech’) – of which the expressive prosodic elements have been confirmed in many different languages (Mazokopaki and Kugiumutzakis 2009 p. 188).

Gratier and Apter-Danon elaborate, stressing that infants and mothers develop a collection of ‘motifs based on repetition and variation of expressive units that carry meaning’ (2009, p. 301). Malloch and Trevarthen assert that this musical interaction is a shared intuition and it makes possible the imitative interaction that occurs between mother and infant, allowing ‘us to share time meaningfully together, in its emotional richness and its structural holding, and for us to participate with anticipation and recollection of pleasure in the “imitative arts”’ (2009, p. 5).

The manner in which a repertoire of motifs is developed between mother and child centres on a cyclical imitative exchange in which the spontaneous sounds that experimentally emanate from an infant are perceived, imitated, and perhaps varied by their mother, after which the infant attempts to imitate the adult response. The cycle continues as such (Gratier and Apter-Danon 2009, p. 301; Bannan and Woodward 2009, p. 480). Throughout this mimetic behaviour (Custodero 2009, p. 514), infants are motivated by the attention and apparent importance that the adult interactions confer, and through this cycle children are initiated into their culture, and begin to learn ‘proto- language’ and activities we may call ‘proto-music’ (Bannan and Woodward 2009, p.

480).

Mazokopaki and Kugiumutzakis assert that this imitative behaviour is part of a larger developmental chronicle:

Imitation, rhythms, rhythmic imitations, melodorhythmic structures and lyrical music … are, all together, basic talents for the growth of music, language, ‘mind reading’, empathy, sympathy, and intersubjectivity in a child, as they

270 certainly were in the early evolutionary stages of the process of hominization called by Merlin Donald (1991) ‘mimetic culture’ (2009, p. 203).

Reviewing over 24 years of research, Bannan and Woodward agree, asserting, ‘The musical/poetic features of infantile vocal sharing evidently prepare the way for learning of language as well as other cultural skills’ (2009, p. 469). Reciprocal imitative behaviour has a fundamental impact on our ability to develop the principal skills that enable social interaction, and as such, it is a vital habitual behaviour instituted from birth.

Whilst having a theoretical evolutionary origin, and a continuing developmental purpose along the human continuum, the detailed evidence of imitation in childhood development and beyond, musical as it is, provides a meta-context for the findings of this dissertation. Elaborating, I draw focus from Bannan and Woodward (2009) who write concerning the interaction between the subjects of this discussion – musicality, improvisation, and imitation:

Infants, like jazz musicians, may create their spontaneous music within any ‘grammatical’ structure of the musical culture that they have absorbed. They create clusters of notes that might be influenced by learnt music, such as the repetition of a phrase with seven regular beats followed by a pause, as in Twinkle, Twinkle Little Star (Young 2003). Deferred imitations of melodies are a natural outcome of children’s impressive capacity for aural memory and invention, and of their intuitive sense of rhythm, phrasing, and expressive narration … Repetition soon evolves into extension and transformation, by controlling a sequence or varying certain elements of a motif, such as the pulse, dynamics, melodic direction or rhythmic content (2009, p. 474).

While not accounting for the complexities inherent to improvisation in those musical traditions that I have compared, nor considering the corresponding arduous long-term practice that musicians undertake, this description of infant behaviour could equally contribute to an explanation of how musicians develop the capacity to improvise. For both infants and musicians, repetitive imitation enables the internalisation of cultural

(genre) specific ‘grammatical’ parameters, while at the same time developing key aural

271 facilities. The cycle, mimetic as it is, involves the processes of extension and variation, which ultimately leads to the creation of novelty, though of course for musicians it is a far more conscious, intentional, and developed process.

Returning briefly to the initial focus of this chapter, these assertions also provide a context for my examination of expertise acquisition in the cultures investigated by this thesis. In light of the psychobiological developmental data presented above, I propose that musicians reach expertise as improvisers because their musical education begins at a time when the imitative behaviour that is so important to the development of their improvisational ability, is an intuitive mechanism already operating within their overarching developmental trajectory.40

In conclusion, I assert that the findings of my cross-cultural comparison, viewed in this meta-context of human development, demonstrate that improvisation, in the context of early childhood and in the specified practice of musicians, revolves around the mastery of (and possibly transcendence of) the fundamental human capacity to imitate.

Furthermore, given that the process through which infants develop essential social skills can be viewed as musical and improvisational, the findings of my cross-cultural comparison are reinforced. Not only do the imitative practices and modes of transmission that I have identified as central to the acquirement of improvisational ability have a clear developmental origin, the act of improvisation itself has a clear origin and purpose in early childhood development. Its reappearance on the concert stage in those musical cultures I have analysed should therefore not be surprising, but a

40 In the Western classical tradition, the emphasis on the acquisition of technique, almost as an end unto itself, and on the learning of content in the form of notated repertoire, have de-emphasised the critical imitative process that is central to learning Western classical music style. However, in those traditions investigated in this dissertation, the imitative mode of transmission is crucial for learning both style and content. An exception to the trend experienced in the Western classical tradition is of course the Suzuki method; and it is noted that prior to the 19th-century Western musicians also learnt chiefly through imitation and apprenticeship.

272 reminder of the interconnectedness of the human experience that stems from a common biological origin at birth.

273 Bibliography

Aiello, R. 1994, ‘Music and language: parallels and contrasts’, in Aiello, R. and Sloboda, J. A. (eds), Musical Perceptions, Oxford University Press, New York, pp. 40– 63.

_____ 2002, ‘Learning jazz, teaching jazz’, in Cooke, M. and Horn, D. (eds), The Cambridge Companion to Jazz, Cambridge University Press, Cambridge, pp. 255–69.

Alter, A.B. 1989, ‘Hindustani Vocal Music: A Case Study of Oral Transmission’, Masters Thesis, The University of Adelaide.

_____1997, ‘Key processes in the oral transmission of Hindustani vocal music’, Musicology Australia, vol. 20, pp. 61–83.

_____2000, ‘Institutional music education: northern area’, in Arnold, A. (ed.), The Garland Encyclopedia of World Music, V: The Indian subcontinent, Garland Publishers, New York, pp. 442–48.

Armstrong, G. and Bates, C. 2001, ‘Selves and others: reflections on sport in South Asia’, Contemporary South Asia, vol. 10, no. 2, pp. 191–205.

Ayari, M. and McAdams, S. 2003, ‘Aural analysis of Arabic improvised instrumental music (taqsim)’, Music Perception, vol. 21, no. 2, pp. 159–216.

Bailey, D. 1992, Improvisation: Its Nature and Practice In Music, British Library National Sound Archive, London.

Baker, D. 1989, David Baker’s Jazz Pedagogy: A Comprehensive Method of Jazz Education for Teacher and Student, Alfred Pub. Co, Van Nuys, CA.

Banerjee, J. 1986, ‘The methodology of teaching Indian classical music’, Journal of the Sangeet Natak Akademi, vol. 79, pp. 11–48.

Bannan, N. and Woodward, S. 2009, ‘Spontaneity in the musicality and music learning of children’, in Malloch, S. and Trevarthen, C. (eds), Communicate Musicality: Exploring the Basis of Human Companionship, Oxford University Press, Oxford, pp. 465–94.

Benson, B.E. 2003, The Improvisation of Musical Dialogue: A Phenomenology of Music, Cambridge University Press, Cambridge.

Berliner, P.F. 1994, Thinking in Jazz: The Infinite Art of Improvisation, The University of Chicago Press, Chicago.

Bhatkhande, V.N. 1981 [1932], Bhātkhande Sangīt-Sāstra: Hindustānī Sangīta- Paddhati, 4 vols., (Hindi trans. Sudama Prasad Dube, et al.), Sangeet Karyalaya, Hathras.

274 Bloom, B.S. (ed.) 1985, Developing Talent In Young People, Ballantine Books, New York.

Booth, G.D. 1982, ‘The Teaching of North Indian Music: Three Case Studies’, Masters Thesis, Kent State University.

_____1983, ‘The teaching of North Indian music’, Contributions to Music Education, vol. 10, pp. 1–7.

_____1986, ‘The Oral Tradition in Transition: Implications for Music Education from a Study of North Indian Tabla Transmission’, Doctoral Thesis, Kent State University.

_____1987, ‘The North Indian oral tradition: lessons for music education’, International Journal of Music Education, vol. 9, no. 1, pp. 7–9.

_____1996, ‘The transmission of a classical improvisatory performance practice’, International Journal of Music Education, vol. 26, pp. 14–26.

Bower, G.H., and Hilgard, E.R. 1981, Theories of Learning (5th ed.), Prentice Hall, Englewood Cliffs, NJ.

Carmi-Cohen, D. 1964, ‘An investigation into the tonal structure of the maqamat’, Journal of the International Folk Music Council, vol. 16, pp. 102–6.

Coker, J. 1986, Improvising Jazz, Simon and Schuster, New York.

Courtney, D. 1995, ‘The cyclic form in North Indian tabla’, Percussive Notes, vol. 33, no. 6, pp. 32–45.

Culicover, P.W. 2005, ‘Linguistics, cognitive science, and all that jazz’, The Linguistic Review, vol. 22, pp. 227–48.

Custodero, L.A. 2009, ‘Intimacy and reciprocity in improvisatory musical performance: pedagogical lessons from adult artists and young children’, in Malloch, S. and Trevarthen, C. (eds), Communicate Musicality: Exploring the basis of human companionship, Oxford University Press, Oxford, pp. 513–29.

Dagher, A. 2000, Fan el-musiqä el-misriyya el-hadithah el-’arabiyyah es-sufiyyah (The New Egyptian-Arabic Sufic Art Music), International Opus, Richmond, Virginia.

Dean, R.T. 1989, Creative Improvisation: Jazz, Contemporary Music, and Beyond, Open University Press, Milton Keynes England; Philadelphia.

Deliége, I. and Wiggins, G.A. (eds), 2006, Musical Creativity: Multidisciplinary Research in Theory and Practice, Psychology Press, New York.

DeVeaux, S. and Giddins, G. 2009, Jazz, W.W. Norton and Company, Inc., New York.

El-Shawan Castelo-Branco, S. 2002, ‘Institutionalization of learning in Egypt’, in Danielson, V., Marcus, S., and Reynolds, D. (eds), The Garland

275 Encyclopedia of World Music, VI: The Middle East, Garland Publishers, New York, pp. 321–24.

Ericsson, K.A., Krampe, R.T., and Tesch-Römer, C. 1993, ‘The role of deliberate practice in the acquisition of expert performance’, Psychological Review, vol. 100, no. 3, pp. 363–406.

Farrell, G. 1997, ‘Thinking, saying, playing: children learning the tabla’, Bulletin of the Council for Research in Music Education, vol. 133, pp. 14–19.

Feldman, W. 1993, ‘Ottoman sources on the development of taksîm’, Yearbook for Traditional Music, vol. 25, pp. 1–28.

Ferand, E.T. 1938, Die Improvisation in der Musik, Rhein-Verlag, Zurich.

Friberg, A. and Sundstrom, A. 2002, ‘Swing ratios and ensemble timing in jazz performance: Evidence for a common rhythmic pattern’, Music Perception, vol. 19, no. 3, pp. 333–49.

Gagné, R.M. 1970, The Conditions of Learning (2nd ed.), Rinehart & Winston, New York.

Garland, P. 1969, The Sound of Soul: The Story of Black Music, Henry Regnery Company, Chicago.

Gratier, M. and Apter-Danon, G. 2009, ‘The improvised musicality of belonging: Repetition and variation in mother-infant vocal interaction’, in Malloch, S. and Trevarthen, C. (eds), Communicate Musicality: Exploring the basis of human companionship, Oxford University Press, Oxford, pp. 301–27.

Gruber, H., Degner, S., and Lehmann, A.C. 2004, ‘Why do some commit themselves to deliberate practice for many years – and so many do not? Understanding the development of professionalism in music’, in Radovan, M. and Dordević, N. (eds), Current Issues in Adult Learning and Motivation: 7th Adult Education Colloquium, Slovenian Institute for Adult Education, Ljubljana, pp. 222–35.

Khan, A.A., Ruckert, G. 1998, The Classical Music of North India: The Music of the Baba Allauddin Gharana as taught by Ali Akbar Khan at the Ali Akbar College of Music (Vol. One: The First Years Study), Munshiram Manoharlal Publishers Pvt. Ltd., New Delhi.

Keightley, K, 2001, ‘You keep coming back like a song: adult audiences, taste panics, and the idea of the standard’, Journal of Popular Music Studies, vol. 13, pp. 7–40.

Kippen, J. 1988, The Tabla of Lucknow: A Cultural Analysis of a Musical Tradition, Cambridge University Press, Cambridge.

_____2000, ‘Hindustani tala’, in Arnold, A. (ed.), The Garland Encyclopedia of World Music, V: The Indian subcontinent, Garland Publishers, New York, pp. 110–37.

276 Kivy, P. 1967, ‘Child Mozart as an aesthetic symbol’, Journal of the History of Ideas, vol. 28, no. 2, pp. 249-58.

Knox, C.W. 2004, ‘Jazz improvisation through reorganization: a return to a traditional learning process’, Jazz Educators Journal, vol. 37, no. 7, pp. 42–4.

Lehmann, A.C. and Ericsson, K.A. 1997, ‘Research on expert performance and deliberate practice: implications for the education of amateur musicians and music students’, Psychomusicology, vol. 16, pp. 40–58.

Lehmann, A.C. and Gruber, H. 2006, ‘Music’, in Ericsson, K.A., Charness, N., Feltovitch, P.J., and Hoffman, R.R. (eds), The Cambridge Handbook of Expertise and Expert Performance, Cambridge University Press, Cambridge, pp. 457–70.

Lipiczky, T. 1985, ‘Tihai formulas and the fusion of ‘composition’ and ‘improvisation’ in North Indian music’, The Musical Quarterly, vol. 71, no. 2, pp. 157–71.

MacKenzie, I. 2000, ‘Improvisation, creativity, and formulaic language’, The Journal of Aesthetics and Art Criticism, vol. 58, vol. 2, pp. 173–79.

Malloch, S. and Trevarthen, C. 2009, ‘Musicality: communicating the vitality and interests of life’, in Malloch, S. and Trevarthen, C. (eds), Communicate Musicality: Exploring the Basis of Human Companionship, Oxford University Press, Oxford, pp. 1– 11.

Marcus, S.L. 1987, ‘The periodization of modern Arab music theory: continuity and change in the definition of the maqamat’, Pacific Review of Ethnomusicology, vol. 5, pp. 35–49.

_____1989, ‘Arab Music Theory in the Modern Period’, Doctoral Thesis, University of California, Los Angeles.

_____1992, ‘Modulation in Arab music: documenting oral concepts, performance rules and strategies’, Ethnomusicology, vol. 36, no. 2, pp. 171–95.

_____1993, ‘Solo instrumental improvisation (taqāsīm) in Arab music’, Middle East Studies Association Bulletin, vol. 27, pp. 108–11.

_____2002, ‘The eastern Arab system of melodic modes in theory and practice: a case study of maqam bayyati’, in Danielson, V., Marcus, S., and Reynolds, D. (eds), The Garland Encyclopedia of World Music, VI: The Middle East, Garland Publishers, New York, pp. 33–44.

Marquis, A.G. 1998, ‘Jazz goes to college: has academic status served the art?’, Popular Music and Society, vol. 22, no. 2, pp. 117–24.

Mazokopaki, K. and Kugiumutzakis, G. 2009, ‘Infant rhythms: expressions of musical companionship’, in Malloch, S. and Trevarthen, C. (eds), Communicate Musicality: Exploring the Basis of Human Companionship, Oxford University Press, Oxford, pp. 185–208.

277

McDaniel, W. 1993, ‘The status of jazz education in the 1990s: an historical commentary’, International Jazz Archives Journal, vol. 1, no. 1, pp. 144–39.

McNeil, A. 1993–1994, ‘Why Hindustani musicians are good cooks: analogies between music and food in north India’, Asian Music, vol. 25, nos. 1 & 2, pp. 69–80.

McPherson, G.E. 1995, ‘Five aspects of musical performance and their correlates’, Bulletin of the Council for Research in Music Education, vol. 127, pp. 115–21.

_____1997, ‘Strategies and skill acquisition in musical performance’, Bulletin of the Council for Research in Music Education, vol. 133, pp. 64–71.

_____2005, ‘From child to musician: skill development during the beginning stages of learning an instrument’, Psychology of Music, vol. 33, no. 1, pp. 5–35.

Meer, W.v.d. 1980, Hindustani Music in the 20th Century, Martinus Nijhoff, The Hague.

Miyazaki, K. and Ogawa, Y. 2006, ‘Learning absolute pitch by children: a cross-section study’, Music Perception, vol. 24, no. 1, pp. 63–78.

Monson, I.T. 1996, Saying Something: Jazz Improvisation and Interaction, University of Chicago Press, Chicago.

Napier, J. 2004, ‘Entering a palace through the servant’s entrance: “a mid-career” approach to the learning of North Indian classical music’, Preparing Musicians: Making New Sound Worlds. Proceedings of the Seminar of the Commission for the Education of the Professional Musician, pp. 165–74.

_____2006, ‘Novelty that must be subtle: continuity, innovation, and ‘improvisation’ in North Indian music’, Critical Studies in Improvisation, vol. 1, no. 3, pp. 1–17.

Nettl, B. 1967, ‘Studies in Blackfoot Indian musical culture, part II’, Ethnomusicology, vol. 11, no. 3, pp. 293–309.

_____1974, ‘Thoughts on improvisation: a comparative approach’, The Musical Quarterly, vol. 60, no. 1, pp. 1–19.

Nettl, B. and Riddle, R. 1973, ‘A study of sixteen performances by Jihad Racy’, Yearbook of the International Council for Traditional Music, vol. 5, pp. 11–50.

_____1998, ‘Taqsim nahawand revisited: the musicianship of Jihad Racy’, in Nettl, B. and Russell, M. (eds), In the Course of Performance: Studies in the World of Musical Improvisation, The University of Chicago Press, Chicago, pp. 369–93.

Nettl, B. and Russell, M. (eds) 1998, In the Course of Performance: Studies in the World of Musical Improvisation, The University of Chicago Press, Chicago.

278 Neuman, D.M. 1980, The Life of Music in North India, Wayne State University Press, Detroit.

Owens, T. 1974, Charlie Parker: Techniques of Improvisation, Doctoral thesis, University of California, Los Angeles.

_____2002 ‘Analysing jazz’, in Cooke, M. and Horn, D. (eds), The Cambridge Companion to Jazz, Cambridge University Press, Cambridge, pp. 286–300.

Porter, L. 1998, John Coltrane: His Life and Music, The University of Michigan Press, Ann Arbor.

Prögler, J.A. 1995, ‘Searching for swing: participatory discrepancies in the jazz rhythm section’, Ethnomusicology, vol. 38, no. 1, pp. 21-54.

Prouty, K.E. 2004, ‘Canons in harmony, or canons in conflict: a cultural perspective on the curriculum and pedagogy of jazz improvisation’, Research and Issues in Music Education, vol. 2, no. 1, pp. 1–18.

_____2005, ‘The history of jazz education: a critical reassessment’, Journal of Historical Research in Music Education, vol. 26, no. 2, pp. 79–100.

_____2006, ‘Orality, literacy, and mediating music experience: rethinking oral tradition in the learning of jazz improvisation’, Popular Music and Society, vol. 29, no. 3, pp. 317–34.

Racy, A.J. 2000, ‘The many faces of improvisation: the Arab taqasim as a musical symbol’, Ethnomusicology, vol. 44, no. 2, pp. 302–20.

_____2003, Making Music in the Arab World: The Culture and Artistry of Tarab, Cambridge University Press, Cambridge.

Ranade, A.D. 1999, ‘Some thoughts on Riaz’, Sangeet Natak, Nos. 131–132, pp. 15–20.

Rao, J., Meer, W.v.d., and Harvey, J. 1999, The Raga Guide, Bor, J. (ed.), Nimbus, London.

Renwick, J.M. and McPherson, G.E. 2002, ‘Interest and choice: student-selected repertoire and its effect on practising behaviour’, British Journal of Music Education, vol. 19, no. 2, pp. 173–88.

Ruckert, G. 2004, Music in North India: experiencing music, expressing culture, Oxford University Press, New York.

Ruckert, G. and Widdess, R. 2000, ‘Hindustani raga’, in Arnold, A. (ed.), The Garland Encyclopedia of World Music, V: The Indian subcontinent, Garland Publishers, New York, pp. 64–88.

Sanyal, R. and Widdess, R. 2004, Dhrupad: tradition and performance in Indian music, Ashgate Publishing Company, Burlington VT.

279

Schippers, H. 2007, ‘The guru recontextualised? Perspectives on learning North Indian classical music in shifting environments for professional training’, Asian Music, vol. 38, no. 1, pp. 123–38.

Scott, S.A. 1997 ‘Power and delight: vocal training in North Indian classical music’, Doctoral thesis, Wesleyan University.

Shankar, R. 1969, My Music, My Life, Jonathan Cape Ltd., London.

Shepherd, F.A. 1976, ‘Tabla and the Benares Gharana’, Doctoral Dissertation, Wesleyan University,

Shiloah, A. 1981, ‘The Arabic concept of mode’, Journal of the American Musicological Society, vol. 34, no. 1, pp. 19–42.

Slawek, S. 2000a, ‘Hindustani instrumental music’, in Arnold, A. (ed.), The Garland Encyclopedia of World Music, V: The Indian subcontinent, Garland Publishers, New York, pp. 188–208.

_____2000b, ‘The classical master-disciple tradition’, in Arnold, A. (ed.), The Garland Encyclopedia of World Music, V: The Indian subcontinent, Garland Publishers, New York, pp. 457–67.

_____2009, ‘Hindustani sitar and jazz guitar music: A foray into comparative improvology’, in Solis, G., and Nettl, B., Musical Improvisation: Art, Education, and Society, Board of Trustees of the University of Illinois, Urbana and Chicago.

Sloboda, J.A., Davidson, J.W., Howe, M.J.A., and Moore, D.G. 1996, ‘The role of practice in the development of performing musicians’, British Journal of Psychology, vol. 87, no. 2, pp. 287–309.

Smith, G.E. 1991, ‘In quest of a new perspective on improvised jazz: a view from the Balkans’, The World of Music, vol. 33, no. 3, pp. 29–52.

Solis, G. 2004, ‘“A unique chunk of jazz reality”: authorship, musical work concepts, and Thelonious Monk’s live recordings from the Five Spot, 1958’, Ethnomusicology, vol. 48, no. 3, pp. 315-47.

Solis, G. and Nettl, B. 2009, Musical Improvisation: Art, Education, and Society, Board of Trustees of the University of Illinois, Urbana and Chicago.

Steele, J. 2004, ‘Improvisation fluency: music education as analogous to second language development’, Canadian Music Educator, vol. 45, no. 3, pp. 24–7.

Stevens, M.R. 1975, ‘The training of Indian musicians’, Music Educators Journal, vol. 61, no. 8, pp. 33–9.

Sudnow, D. 1978, Ways of the Hand, Routledge and Kegan Paul Ltd., London.

280 _____ 2001, Ways of the Hand: A Rewritten Account, MIT Press, Cambridge MA.

Szwed, J. 2003, So What: The life of Miles Davis, Arrow Books, London.

Tanner, P. 1971, ‘Jazz goes to college, part II’, Music Educators Journal, vol. 57, no. 8, pp. 49–93.

Thomas, A.E. 2006, ‘Developing Arab music: Institutions, Individuals, and Discourses of Progress in Cairo 1939–2005’, Doctoral Thesis, Brown University.

Tirro, F. 1974, ‘Constructive elements in jazz improvisation’, American Musicological Society Journal, vol. 27, no. 2, pp. 285–305.

Touma, H. 1971, ‘The maqam phenomenon: an improvisation technique in the music of the Middle East’, Ethnomusicology, vol. 15, no. 1, pp. 38–48.

_____1996, The Music of the Arabs, trans. L. Schwartz, Amadeus Press, Portland OR.

Wade, B.C. 1984, Khyal, Cambridge University Press, Cambridge.

_____2000, ‘Hindustani vocal music’, in Arnold, A. (ed.), The Garland Encyclopedia of World Music, V: The Indian subcontinent, Garland Publishers, New York, pp. 162– 87.

Walser, R. 1992, ‘Eruptions: heavy metal appropriations of classical virtuosity’, Popular Music, vol. 11, no. 3, pp. 263–308.

Williams-Jones, P. 1975, ‘Afro-American gospel music: A crystallization of the black aesthetic’, Ethnomusicology, vol. 19, no. 3, pp. 373–85.

281 Internet bibliography

Dagar, H.S., 2005, ‘Râg Bhopali Part 3’, accessed 28 March 2012, www.youtube.com/watch?v=9_30Tcytj5M&feature=related (at 5’06’’).

El-Shawan Castelo-Branco, S. 2007–2012, ‘Egypt, classical and popular traditions; general background’, Grove Music Online, Oxford Music Online, accessed 10 January 2012, www.oxfordmusiconline.com/subscriber/article/grove/music/08621.

‘Improvisation’, Oxford English Dictionary Online, 2012. Oxford University Press, accessed 20 February 2012, www.oed.com.wwwproxy0.library.unsw.edu.au/view/Entry/92872?redirectedFrom=im provisation#eid.

Maqam World, 2001–2005 (last updated 2007), accessed 10 January 2012, www.maqamworld.com/maqamat.html.

Nettl, B., 2007–2012, ‘Improvisation’, Grove Music Online, Oxford Music Online, accessed 13 January 2012, www.oxfordmusiconline.com/subscriber/article/grove/music/13738.

Parncutt, R. and Levitin, D.J. 2007, ‘Absolute pitch’, Grove Music Online. Oxford Music Online, accessed 13 January 2012, http://www.oxfordmusiconline.com/subscriber/article/grove/music/00070

Qureshi, R. 2007–2012, ‘India; the region: cultural context and musical categories’, Grove Music Online. Oxford Music Online, accessed 10 January 2012, www.oxfordmusiconline.com/subscriber/article/grove/music/43272pg1.

Weisberg, R.W., Brinkman, A.R., Dick, A.S., Fleck, J.I., Niederberger, B., Folio, C.J. and Barrett, F. 2012, ‘Towards a cognitive analysis of creativity: improvisation in jazz’, Uni Graz, accessed 13 January 2012, www.uni- graz.at/richard.parncutt/cim04/CIM_paper_pdf_alt/Microsoft%20Word%20- %20Weisberg_Brinkman_Folio_Dick_Fleck_Niederberg_Barrett_proc.pdf.

282 Discography

Abou Mrad, N., 2005, ‘Music from Lebanon and the Levant of the Arab Renaissance: The legacy of Mīhā’īl Maŝŝāqa (1800–1888)’, CD, INCOGNITO, UPAC 1002.

Coltrane, J. 1995, ‘A Love Supreme’, IMPULSE, IMP 11552.

Elling, K. 2003, ‘Man in the Air’, BLUE NOTE, 7243 5 80834 2 2.

Hawkins, C. 1998, ‘Ultimate Coleman Hawkins’, CD, POLYGRAM RECORDS, VERVE 557538-2.

Omran, M., 2011, ‘Spiritual Voices (1) – Islamic Literature – Azan (Adhān): A Selection Of Rare Records In Egypt, Iran And Turkey’, MP3, ITUNES, MP3 Download.

Tawadros, J., 2010, ‘The Hour of Separation’, CD, ENJA, ENJ-9555.

283 Appendix 1: Excerpt from Alter (1989, pp. 179-186)

284 285 286 287 288 289 290 291