Volume 1
Musicians in the Making: Pathways to Creative Performance
Edited by John Rink, Helena Gaunt and Aaron Williamon
Volume 2
Distributed Creativity: Collaboration and Improvisation in
Contemporary Music
Edited by Eric F. Clarke and Mark Doffman
Volume 3
Music and Shape
Edited by Daniel Leech-Wilkinson and Helen M. Prior
Volume 4
Global Perspectives on Orchestras: Collective Creativity and Social Agency
Edited by Tina K. Ramnarine
Volume 5
Music as Creative Practice
Nicholas Cook
STUDIES IN MUSICAL PERFORMANCE AS CREATIVE PRACTICE
Until recently, the notion of musical creativity was tied to composers and the works
they produced, which later generations were taught to revere and to reproduce in
performance. But the last few decades have witnessed a fundamental reassessment
of the assumptions and values underlying musical and musicological thought and
practice, thanks in part to the rise of musical performance studies. The five volumes in
the series Studies in Musical Performance as Creative Practice embrace and expand
the new understanding that has emerged. Internationally prominent researchers,
performers, composers, music teachers and others explore a broad spectrum of
topics including the creativity embodied in and projected through performance,
how performances take shape over time, and how the understanding of musical
performance as a creative practice varies across different global contexts, idioms
and performance conditions. The series celebrates the diversity of musical perfor-
mance studies, which has led to a rich and increasingly important literature while
also providing the potential for further engagement and exploration in the future.
These books have their origins in the work of the AHRC Research Centre
for Musical Performance as Creative Practice (www.cmpcp.ac.uk), which con-
ducted an ambitious research programme from 2009 to 2014 focused on live
musical performance and creative music-making. The Centre’s close inter
actions with musicians across a range of traditions and at varying levels of
expertise ensured the musical vitality and viability of its activities and outputs.
Studies in Musical Performance as Creative Practice was itself broadly con-
ceived, and the five volumes encompass a wealth of highly topical material.
Musicians in the Making explores the creative development of musicians in
formal and informal learning contexts, and it argues that creative learning is
a complex, lifelong process. Distributed Creativity explores the ways in which
collaboration and improvisation enable and constrain creative processes in
contemporary music, focusing on the activities of composers, performers and
improvisers. Music and Shape reveals why a spatial, gestural construct is so
invaluable to work in sound, helping musicians in many genres to rehearse,
teach and think about what they do. Global Perspectives on Orchestras consid-
ers large orchestral ensembles in diverse historical, intercultural and postcolo-
nial contexts; in doing so, it generates enhanced appreciation of their creative,
political and social dimensions. Finally, Music as Creative Practice describes
music as a culture of the imagination and a real-time practice, and it reveals the
critical insights that music affords into contemporary thinking about creativity.
Music and Shape
Edited by
Daniel Leech-Wilkinson
Helen M. Prior
1
1
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.
9 8 7 6 5 4 3 2 1
Printed by Sheridan Books, Inc., United States of America
CONTENTS
List of contributors ix
List of illustrations xvi
About the Companion Website xxiii
Preface xxv
DANIEL LEECH-WILKINSON AND HELEN M. PRIOR
Reflection David Amram 383
Reflection Antony Pitts 386
Notes 389
Index 397
LIST OF CONTRIBUTORS
Mordechai Adler graduated from Tel Aviv University in 2014 with a PhD in
musicology. His dissertation, ‘Cross-modal correspondence and musical repre-
sentation’, combines empirical studies of cross-modal perception with musical
analyses. Adler is currently developing a music education method using cross-
modal correspondences.
David Amram is one of the most prolific and performed composers of his gen-
eration, and has left a unique mark on the world of music. He became the first
composer-in-residence with the New York Philharmonic in 1966 at the request
of Leonard Bernstein. At eighty-six Amram continues to work as a classical
composer, multi-instrumentalist, band leader, lecturer and guest conductor,
constantly composing as he tours the world.
Mark Applebaum is Associate Professor of Composition at Stanford
University. His solo, chamber, choral, orchestral, operatic and electroacoustic
work has been performed widely and includes notable commissions from the
Merce Cunningham Dance Company, the Fromm Foundation and the Vienna
Modern Festival. Many of his pieces challenge the conventional boundaries of
musical ontology. Applebaum is also an accomplished jazz pianist and builds
electroacoustic sound-sculptures out of junk, hardware and found objects.
Max Baillie is a leading instrumentalist of his generation, equally at home on
both violin and viola. As a performer he has appeared on stages from Carnegie
Hall to Glastonbury and from Mali to Moscow in a diverse spectrum of styles
including classical, pop, folk and electronic music, alongside leading artists
from around the world. He plays principal viola in the Aurora Orchestra and is
part of a series of unique creative projects which go beyond the concert stage.
Philip Barnard worked for the Medical Research Council’s Cognition and
Brain Sciences Unit (CBSU) in Cambridge from 1972 to 2011, where he car-
ried out research on how memory, attention, language, body states and emo-
tion work together. He is now retired but remains a visiting researcher with the
CBSU. Since 2003, he has been collaborating with Wayne McGregor | Random
Dance to develop productive synergies between choreographic processes and
our knowledge of cognitive neuroscience.
George Benjamin was born in 1960 and began composing at the age of seven.
After studying with Messiaen he worked with Alexander Goehr at King’s
ix
x List of contributors
College, Cambridge. His Ringed by the Flat Horizon was performed at the BBC
Proms when he was just twenty. Written on Skin has been scheduled by numer-
ous international opera houses since its 2012 premiere in Aix. He regularly
conducts some of the world’s leading orchestras and since 2001 has been the
Henry Purcell Professor of Composition at King’s College London.
Malcolm Bilson has been a key contributor to the restoration of the fortepiano
to the concert stage and to fresh recordings of the ‘mainstream’ repertory. He
has recorded the Mozart piano concertos with John Eliot Gardiner and the
English Baroque Soloists, and the complete Mozart and Schubert solo sona-
tas. Bilson gives concerts, masterclasses and lectures around the world. He is a
member of the American Academy of Arts and Sciences and has an honorary
doctorate from Bard College.
Lucia D’Errico is an artist devoted to experimental music, performing on
plucked string instruments. As a performer and improviser, she collaborates
with contemporary music groups and with theatre, dance and visual art com-
panies. An artistic researcher at Orpheus Institute Ghent, she is part of the
ME21 research project. Her doctoral research (on the docARTES programme)
focuses on recomposing baroque music. She is also active as a freelance graphic
designer.
Scott deLahunta has worked as writer, researcher and organizer on a range of
international projects bringing performing arts with a focus on choreography
into conjunction with other disciplines and practices. He is currently Senior
Research Fellow at Deakin University (Australia) in partnership with Coventry
University (UK), R-Research Director (on sabbatical) at Wayne McGregor |
Random Dance, and Director of Motion Bank/The Forsythe Company.
Simon Desbruslais has an international reputation as a trumpet soloist, spe-
cializing in the performance of baroque and contemporary music. His solo
disc Contemporary British Trumpet Concertos on Signum Classics includes
new works written for him by John McCabe, Deborah Pritchard and Robert
Saxton. He is a lecturer in music at the University of Hull, and has taught at the
universities of Oxford, Bristol, Nottingham and Surrey. He is writing a mono-
graph on the music and music theory of Paul Hindemith, based on his doctoral
dissertation from Christ Church, Oxford.
Zohar Eitan is a professor of music theory and music cognition at the
Buchman-Mehta School of Music, Tel Aviv University. His recent research
was published in Cognition, Journal of Experimental Psychology: Human
Perception and Performance, Experimental Psychology, Music Perception,
Psychology of Music, Musicae Scientiae, Empirical Musicology Review and
Psychomusicology.
List of contributors xi
has been involved in diverse art and educational projects around the world.
Much of his work is inspired by his own experience of synaesthesia.
Daniel Leech-Wilkinson studied at the Royal College of Music, King’s College
London and Clare College, Cambridge, becoming first a medievalist and then,
since c. 2000, specializing in the implications of early recordings, especially
in relation to music psychology and performance creativity. He led a project
on ‘Expressivity in Schubert song performance’ within the AHRC Research
Centre for the History and Analysis of Recorded Music (CHARM), followed
by ‘Shaping music in performance’ as part of the AHRC Research Centre for
Musical Performance as Creative Practice. Books include The Modern Invention
of Medieval Music (2002) and The Changing Sound of Music (2009).
Anna Meredith is a composer and performer of both acoustic and electronic
music. She has been Composer in Residence with the BBC Scottish Symphony
Orchestra, RPS/PRS Composer in the House with Sinfonia ViVA, the classical
music representative for the 2009 South Bank Show Breakthrough Award and
winner of the 2010 Paul Hamlyn Award for Composers. HandsFree (2012), a
PRS/RPS 20x12 Commission for the National Youth Orchestra, was performed
at the BBC Proms, Barbican Centre and Symphony Hall as well as numer-
ous flashmob performances around the UK. Her debut EP Black Prince Fury
was released on Moshi Moshi records to critical acclaim including Drowned in
Sound’s ‘Single of the Year’.
Milton Mermikides is Lecturer in Music and Head of Composition at the
University of Surrey, Professor of Jazz Guitar at the Royal College of Music,
and deputy director of the International Guitar Research Centre. He is a com-
poser, guitarist and sound artist with a keen interest in a range of disciplines
including jazz, popular, electronic and ‘world’ music, improvisation, digital
technologies in analysis and creative practice, music perception, art/science col-
laboration, and data sonification.
Richard G. Mitchell is a film composer. He graduated from Central Saint
Martins in fine art and film, where he composed for students at Saint Martins,
Royal College of Art and National Film School. His best-known works are
A Good Woman (Scarlett Johansson, Helen Hunt), To Kill a King (Tim Roth,
Rupert Everett), and Grand Theft Parsons (Johnny Knoxville), and he has an
Ivor Novello Award for Trial by Fire, a Royal Television Society Award for the
BBC The Tenant of Wildfell Hall, and a Polish Academy Award for Günter
Grass’s The Call of the Toad.
Adam Ockelford is Director of the Applied Music Research Centre at the
University of Roehampton, UK. His research interests are in music psychol-
ogy, education, theory and aesthetics(particularly special educational needs
and the development of exceptional abilities); learning, memory and creativity;
xiv List of contributors
Figures
xvi
List of illustrations xvii
R.19 Bach, B minor Mass, Gloria II, bars 57–61: a) Gesellschaft edition,
followed by b) a notated interpretation 244
R.20 Bach, B minor Mass, Cum sancto spiritu, bars 111–17 245
R.21 J. S. Bach, Complete Trumpet Repertoire, Vol. III with my
annotations 245
R.22 Tchaikovsky, Swan Lake Suite Op. 20a, ‘Intrada’, rehearsal
mark 13 246
R.23 Pritchard, Skyspace (2012), third movement, notated for piccolo
trumpet in A, bars 1–8 246
R.24 Beethoven, Piano Sonata in F minor, Op. 2 No. 1, first movement,
bars 1–9 249
R.25 Three-dimensional mixing metaphor 279
R.26 The Metaphysics of Notation, panel 4 285
R.27 The Metaphysics of Notation, panel 4 close-up: descending
‘shields’ 286
R.28 The Metaphysics of Notation, panel 4 close-up: sinusoidal curve 287
R.29 The Metaphysics of Notation, panel 5 288
R.30 The Metaphysics of Notation, panel 5 close-up: materialization of
rectilinear forms 289
R.31 The Metaphysics of Notation, panel 5 close-up: contrasting materials,
‘heart guitar’ and canonic dots 291
R.32 The Metaphysics of Notation, panels 3, 4, 5, 6 and 7 in stacked
arrangement 293
R.33 The Metaphysics of Notation, close-up: circle and oval pair inverted
across panels 3 and 4 294
R.34 The Metaphysics of Notation, close-up: ‘scroll’ with number five
inverted across panels 3 and 4 295
R.35 The Metaphysics of Notation, close-up: panels 4 and 5 inverted
shields, connection to the ‘heart guitar’ 296
R.36 The Metaphysics of Notation, close-up: panels 5, 6 and 7 dangling
angles, chain of circles, dot clock 297
R.37 The Metaphysics of Notation, panels 9 & 10 in stacked
arrangement 299
R.38 Watercolour paintings: (a) Red and White and (b) Fireworks 304
9.1 RP’s synaesthetic experience to overtone singing by Wolfgang
Saus 315
R.39 Timothy B. Layden, Dark Glistening. 322
10.1 Selected illustrations for the productions of (a) ATOMOS,
(b) ENTITY and (c) UNDANCE for Wayne McGregor | Random
Dance 330
10.2 (a) Still from video annotating form and flow in Forsythe’s
One Flat Thing; (b) Difference forms in movement viewed from
above 331
List of illustrations xxi
Tables
xxiii
PREFACE
Daniel Leech-Wilkinson and Helen M. Prior
‘Music and what?!’ people have tended to ask us. But, as Godøy’s remarks sug-
gest, that puzzlement is not shared by musicians: they always seem to know
what we’re referring to.1 In a sense it’s that discrepancy that inspired this book
and the research project from which it has emerged. For although the connec-
tion between shape and sound may seem mystifying to others, Prior (2012)
finds that professional musicians use ‘shape’ to talk and think about how to
perform notes, phrases, melodic lines, melodic patterns, harmonic features,
harmonic patterns, rhythms, movements, compositions, changes in loudness,
tempo and expression; and this applies in classical music, jazz, folk, pop, rock,
urban, world musics and crossover, and for people who originated from thirty-
one countries, 43 per cent of them fluent in a language other than English.
Moreover, for speakers of languages which do not use a simple equivalent to
‘shape’ in discussion of music, the concept was nevertheless immediately recog-
nized from their own musical discourse. The use of the term is also not merely
a current ‘fashion’: there is evidence to show its use by composers, performers
and critics throughout the twentieth century and to some degree earlier (see
Table 0.1).2 Evidently, shape is a concept that is flexible, ubiquitous and very
useful when thinking and speaking about performance and composition.
With so many and such varied uses, it cannot just be visual or tactile shape
that we are dealing with. Shape must be doing much broader metaphorical
work, transferring into different, less tangible domains including time, quan-
tity, intensity, complexity, speed and emotional response, at least.3 One way of xxv
looking at this is to say that shape means so many things in relation to music
xxvi Preface
(continued)
Preface xxvii
TABLE 0.1 Continued
Shape in Nalen Anthoni (critic) ‘The music breathes a life of its own as he ardently
relation to inflects its phrases to shape the tension of his line.’
musical (Anthoni 2008: 65)
expression Dietrich Fischer-Dieskau ‘Shape the endings of the long phrases in the
(performer) recitative in a way that the conductor can
easily follow you.’ (Dietrich Fischer-Dieskau in
Monsaingeon 1992)
Trevor Harvey (critic) ‘The orchestral playing is not just good, it is really
outstanding: the conductor knows how to give
us flexible and shapely phrases as well as tightly
rhythmic music.’ (Harvey 1954: 59)
Rachel Podger (performer) ‘With Vivaldi there are so many possibilities to
shape the music.’ (Podger 2003: 15)
Stephen Plaistow (critic) ‘Richter doesn’t shape the actual subjects in
the fugues very much, preferring to state them
flatly and to let the counterpoint achieve its own
expressiveness.’ (Plaistow 1965: 114)
Alec Robertson (critic) ‘It is a pity this artist has so little feeling for the
shape of a phrase.’ (Robertson 1947: 165)
Stanley Sadie (critic/musicologist) ‘Another thing Podger is specially good at is the
shaping of those numerous passages of Vivaldian
sequences, which can be drearily predictable,
but aren’t so here because she knows just how to
control the rhythmic tension and time the climax
and resolution with logic and force.’
(Sadie 2003: 51)
Shape in Aaron Cassidy (composer/ ‘… the notion that the primary morphological
relation to musicologist) unit—not only in my music but also in music in
movement general—is not merely the aural gesture, but far
or gesture more importantly, the physical gesture. I would
assert that the shapes and local forms that we
hear and process as listeners are at their core
the byproducts of physical, visceral activities and
energies, and, further, that the physical motion
required to create a particular sound or set of
sounds is the most important component of a
gesture’s morphological identity.’ (Cassidy 2004: 34)
that it in effect means nothing at all. But that kind of throwing up of hands in
despair doesn’t lead to very penetrating scholarship; and in any case, its very
imprecision may prove to be its raison d’être. Better, then, to approach shape as
a concept with some unusual and intriguing properties, and to try to find out
what those might be and what they might suggest about its place in the brain’s
responses to music.
It was with this aim as an ideal, albeit one we could not hope to realize,
that we planned and carried through a three-year research project (2009–12) on
music and shape, funded by the UK’s Arts and Humanities Research Council
within its Research Centre for Musical Performance as Creative Practice.4 In the
event we managed to continue for a further two years, since there was so much
to do and King’s College London continued to provide support. This book
contains some of the results of that project work (the chapters by Küssner,
xxviii Preface
from a single note to a whole piece of music; it can apply to scores, perfor-
mances and listening experiences, and within those to such varied features as
narrative structure, form, loudness, brightness, tempo, speed, density, register,
intensity, harmonic or interval patterning, pitch direction, sound spectrum,
distance and timbre. As such, it acts as a highly efficient synthesizing tool for
musicians to use in order to negotiate the vast array of musical choices avail-
able to them in performance.
Shape’s flexibility and usefulness are just as clear from the range of other views
that this book offers. In Adam Ockelford’s Chapter 5, shape is seen as a core
property of music that links together its notation, its audible features and our
cognition of musical structure. In Michael Spitzer’s Chapter 4, a sonata by Bach
is compared ‘both to the shape of particular emotional behaviours and to the
expressive shapings of a formal model’ as well as ‘performance styles of “expres-
siveness” ’. For Milton Mermikides and Eugene Feygelson, writing about impro-
visation in Chapter 6, shaping processes are conceived of as strategies through
which material is selected and transformed within musical space. For Philip
Barnard and Scott deLahunta, in Chapter 10, ‘shape’ is a useful concept for
dancers and choreographers not just to describe bodily configuration and move-
ment but also ‘to index the more ineffable meanings and relationships that are
intuited to “make sense” in an artistic context’. For Rolf Inge Gødoy (Chapter 1),
‘shape-cognition in music is opening up new areas of musicological, aesthetic
and affective psychological research, as well as providing practical tools in artis-
tic creation, for example in the domains of sonic design and various kinds of
multimedia art’. For the synaesthetes with whom Jamie Ward works, shapes not
only are a means of conceptualizing complex interactions of musical features and
the feelings they seem to trigger, but are experienced ‘at multiple levels in music:
from single notes through to whole compositions and performances’ (Chapter 9)
as sensations automatically generated by hearing music. Among the various
aspects of shape our practitioners discuss in their Reflections, George Benjamin
mentions shape especially in relation to form, Malcolm Bilson to performance
style, Stephen Hough to composition style, Timothy B. Layden to visual impres-
sions, Lucia D’Errico to bodily sensation, Alex Reuben to body movement,
Alice Eldridge to both gesture and visual representations, Richard G. Mitchell
to emotional change, Evelyn Glennie to dynamics (in the fluid sense), David
Amram to musical character, I-Uen Wang Hwang to rhythm and metre, Max
Baillie to harmony, Simon Desbruslais to timbre, Steven Isserlis to narrative,
Steve Savage to sonic landscape, Antony Pitts to initial inspiration, and Julia
Holter to closure. It goes without saying that all of these factors could be written
about separately and in much greater depth, and indeed they have been. But the
point is not that ‘shape’ could always be replaced by a more precise term—one
which varies according to the context in which shape is being used. Rather, what
we need to ask is why shape is so useful in the sample of contexts discussed here,
and by implication in so many others; and why it is so much more useful than
the more precise term that might pin it down in each case.
xxx Preface
Concepts very like the ‘synthesizing’ notion we discuss here have been
invoked in the past. Mine Doğantan-Dack has summarized this interestingly in
her essay in the Empirical Musicology Review volume mentioned above:
Christian von Ehrenfels, who is best-known today for his article titled ‘Über
Gestaltqualitäten’, i.e. ‘On Gestalt Qualities’, … published in 1890 …
argued that each experience we have of a Gestalt or form in any sensory
modality is cognized as structurally analogous to the experience of a spa-
tial shape. In other words, spatial Gestalten serve in his view as references
for our comprehension of forms in other modalities. An immediate impli-
cation of this idea is that concepts related to the perception of spatial
shapes can be applied to shapes extended in time—for instance, tonal
patterns. Indeed, the idea that there are similarities of form between dif-
ferent fields of experience is one of the most important conclusions of
Ehrenfels’ article. (Doğantan-Dack 2013: 213–14)
Yet neither of these studies was followed up at the time, probably because they
had no points of contact with contemporary musicology, whose concern above
Preface xxxi
all was to present music as a subject for historical and textual study. Closest
in the intervening years, as Doğantan-Dack points out, was Susanne Langer
(1942), whose interest in how music feels brings some of her work into the same
orbit. And indeed, shape’s re-emergence recently can be understood as part of
a growing interest in those musical practices and responses that draw on feeling
more than on thinking; this is a result of the increasing focus of music stud-
ies on emotion, enabled by the development of music psychology and neuro-
science. In this context, Kim, Doğantan-Dack and Leech-Wilkinson (this last
in our volume) have all (independently) pointed to child-psychiatrist Daniel
Stern’s work on vitality affects (2010), which in a sense (though unknown to
Stern) extends Truslit’s work, as a valuable theoretical base for understanding
musical shape. Interrelations with other work are suggested, too, by Godøy in
the continuation of the quotation that begins this Introduction:
We could thus speak of widespread and deep-rooted shape cognition in
music, as well as in human reasoning in general, as suggested by some
directions in the cognitive sciences, foremost by so-called morphodynami-
cal theory and so-called cognitive linguistics. (Godøy 2013: 223)
Much relevant work has been done by researchers studying music and gesture,
outstandingly Godøy himself and Marc Leman (Leman 2007; Godøy and
Leman 2010). Gesture clearly implies shape: it is often considered as includ-
ing performers’ executive and expressive movements—that is, how they move
while they play—but it has also been used extensively to talk about habits in
the forming or performing of short sequences of notes (Gritten and King 2006,
2011). Yet while gesture is closely tied to indicative human movement, shape
seems more abstract and thus more flexible in its application to musical and
other kinds of action.
Another difficulty is hinted at in Leech-Wilkinson’s chapter, where the pos-
sibility is raised that a sense of ‘shape’ arises from a submodal feature common
to all the sense modalities. This extends beyond the cross-domain mapping
that several chapters (Ockelford’s; Eitan, Timmers and Adler’s; and Spitzer’s
in particular) see as crucial to shape’s multiple applications. A submodal role
for shape might explain why our understanding of what shape refers to in
music is at once so multifaceted and so hazy, and why it may always remain
so. For submodal features, as pointed out by Marks (1978), are necessarily
beyond conscious perception: they are components in our sensory experience
but not accessible to consciousness directly through the senses. Alex Reuben’s
impressionistic Reflection on his work as a filmmaker may well be pointing
towards this aspect of shape: in using shape to link feelings in different senses
and art forms, he is not being merely touchy-feely but may be drawing on the
submodal qualities of shape (operating in the recently discovered domain of
multisensory perception) as an aspect of the dynamics of all sensory experi
ence. What previously seemed fanciful now is beginning to seem simply
xxxii Preface
correct: feelings aroused by one sense can be linked by the brain to feelings
aroused by others, so that input in one mode can be read in terms of the
impressions arising from others—and not just for synaesthetes, in fact partic-
ularly not for synaesthetes since for them the effect is fixed whereas for others
it varies with context. Synaesthetes, nevertheless, offer particularly interesting
insights into musical shape. For, as Jamie Ward has shown, their experiences,
though remarkably varied, still make better sense to nonsynaesthetes than
artificial alternatives.
It looks, then, as if the kind of work that ‘shape’ does for musicians draws
on some quite fundamental aspects of perception, while at the same time offer-
ing us a host of ways of thinking about the experience and practice of music
on many other levels. The chapters and Reflections are interleaved and ordered
so as to emphasize interconnections. While they are grouped thematically into
five sections—shapes mapped, composed, performed, seen and felt—there is
also a gradual shift of theme so that the borders between sections are fuzzy. To
read from cover to cover, then, should be to take a journey through views of
music and shape. Most contributions speak of multiple facets of this complex
relationship, however: other orderings are possible, and dipping in and out will
often make further connections apparent.
References
Anderson, W. R., 1934: record review of HMV DB 2142, The Gramophone 11 (no. 130): 396.
Anonymous, 2008: interview with Anthony Marwood, Gramophone 85 (no. 1029): 15.
Anthoni, N., 2008: review of Arkiv Production 477 7371–2, Gramophone 86 (no. 1035): 65.
Barenboim, D., 2005: ‘Barenboim on Beethoven: masterclasses’ (EMI DVD 68993).
Berliner, P. F., 1994: Thinking in Jazz: The Infinite Art of Improvisation (Chicago and
London: University of Chicago Press).
Britten, B. and D. Mitchell, 1969: ‘Benjamin Britten in conversation with Donald Mitchell’,
CD booklet accompanying BBC Legends: Britten Mozart Requiem (BBCL 4119–2).
Cameron, L., 2010: ‘What is metaphor and why does it matter?’, in L. Cameron and
R. Maslen, eds., Metaphor Analysis: Research Practice in Applied Linguistics, Social
Sciences and the Humanities (London: Equinox), pp. 3–25.
Cassidy, A., 2004: ‘Performative physicality and choreography as morphological deter-
minants’, in C.- S. Mahnkopf, F. Cox and W. Schurig, eds., Musical Morphology
(Hofheim: Wolke Verlag), pp. 34–51.
Doğantan-Dack, M., 2013: ‘Tonality: the shape of affect’, Empirical Musicology Review 8/
3–4: 208–18.
Ehrenfels, C. von, 1890: ‘Über “Gestaltqualitäten” ’, Vierteljahrsschrift für wissenschaftliche
Philosophie 14: 242–92.
Eigeldinger, J.-J., 1997: ‘Chopin and “La note bleue”: an interpretation of the Prelude Op.
45’, Music & Letters 78/2: 233–53.
Preface xxxiii
Gibbs, R. W., 2008: ‘Metaphor and thought: the state of the art’, in R. W. Gibbs, ed., The
Cambridge Handbook of Metaphor and Thought (Cambridge: Cambridge University
Press), pp. 3–13.
Godøy, R. I., 2013: ‘Shape cognition and temporal, instrumental and cognitive constraints
on tonality. Public peer review of “Tonality: the shape of affect” by Mine Doğantan-
Dack’, Empirical Musicology Review 8/3–4: 223–6.
Godøy, R. I. and M. Leman, 2010: Musical Gestures: Sound, Movement, and Meaning
(New York: Routledge).
Gritten, A. and E. King, eds., 2006: Music and Gesture (Aldershot: Ashgate).
Gritten, A. and E. King, eds., 2011: New Perspectives on Music and Gesture (Aldershot:
Ashgate).
Harvey, T., 1954: record review of Decca LW 5114, The Gramophone 32 (no. 374): 59.
Kim, J. H., 2013: ‘Shaping and co-shaping forms of vitality in music: beyond cognitiv-
ist and emotivist approaches to musical expressiveness’, Empirical Musicology Review
8/3–4: 162–73.
Langer, S., 1942: Philosophy in a New Key (Cambridge, MA: Harvard University Press).
Leech-Wilkinson, D. and H. M. Prior, 2014: ‘Heuristics for expressive perfor-
mance’, in D. Fabian, R. Timmers and E. Schubert, eds., Expressiveness in Music
Performance: Empirical Approaches across Styles and Cultures (Oxford: Oxford
University Press), pp. 34–57.
Leman, M., 2007: Embodied Music Cognition and Mediation Technology (Cambridge,
MA: MIT Press).
Mahnkopf, C.-S., F. Cox and W. Schurig, 2004: Musical Morphology (Hofheim: Wolke
Verlag).
Marks, L. E., 1978: The Unity of the Senses: Interrelations among the Modalities
(New York: Academic Press).
Monsaingeon, B., 1992: The Mastersinger—Lesson III (EMI DVB 3101949).
Plaistow, S., 1965: review of Deutsche Grammophon (S)LPM18950, The Gramophone 43
(no. 507): 114.
Podger, R., 2003: ‘A question to … Rachel Podger, Baroque violinist’, Gramophone 80
(no. 966): 15.
Prior, H. M., 2012: ‘Shaping music in performance: report for questionnaire participants
(revised August 2012)’, http://www.cmpcp.ac.uk/wp-content/uploads/2015/09/Prior_
Report.pdf (accessed 9 April 2017).
Quinn, M., 1999: review of Droffig National Trust NTCC014, Gramophone 76 (no. 914): 72.
Robertson, A., 1947: record review of Decca M 602, The Gramophone 24 (no. 287): 165.
Rothfarb, L., 2001: ‘Energetics’, in T. Christensen, ed., The Cambridge History of Western
Music Theory (Cambridge: Cambridge University Press), pp. 927–55.
Sadie, S., 2003: review of Channel Classics CCS15958, Gramophone 80 (no. 966): 51.
Stern, D., 2010: Forms of Vitality: Exploring Dynamic Experience in Psychology, the Arts,
Psychotherapy, and Development (Oxford: Oxford University Press).
Truslit, A., 1938: Gestaltung und Bewegung in der Musik (Berlin: Chr. Friedrich Vieweg).
PART 1
Shapes mapped
Reflection
Evelyn Glennie, percussionist
The shape of music is constantly fluid because nothing resonates the same
twice. Every sound and shape is born and reborn. When music is printed on the
page it takes shape in my imagination with the eye leading the way.
As a performer, the environment is my instrument and percussion instru-
ments are my tools to deliver the sound. I can provide all the musical ingredi-
ents for the environment I am immersed in. The acoustic will mould the sound
meal which is thus delivered to the audience. The members of the audience will
have differing perspectives on the sound and shape according to where they are
situated and their emotional state at the time.
Listening is ever-present, recognizing that the body is a huge ear that allows
us to experience the sensation of the sound journey, reached far beyond the
capacity of the ear alone. That in turn creates the fluid shapes in music.
3
1
Shape representations
Needless to say, music and shape is a very extensive topic, with ramifications to
most areas of music and music-related research. Yet out of all this material, it
could be useful to take a brief look at some aspects of western music notation
and more recent instances of shape representations in musical research, to bet-
ter situate our motor theory perspective on shape in music.
6 Music and Shape
FIGURE 1.1 A pianola representation of the first eight bars of J. S. Bach’s Fugue in C major, Well-
Tempered Clavier Book I. This representation highlights the gradually expanding pitch space,
fanning out to several octaves from the initial middle C. The shape of this pitch space expansion is
one of the main architectural elements here (as well as in the rest of J. S. Bach’s works and much other
music for that matter); however, the timescale of this kind of shape is rather slow, i.e. is on what we
call the macro timescale (see p. 14 below).
For one thing, western common practice notation, as well as recent exten-
sions such as MIDI pianola representation (Figure 1.1), can partly be regarded
as a kind of choreographic script, a system for denoting sound-producing body
motion to be realized by performers. Trained score readers may readily see corre-
spondences between the graphical shapes in the score, the required motion shapes
of the performers and the emergent sonic shapes, in particular as pitch contours
and rhythmical-textural shapes. In other cases, there may be less c lear relation-
ships between visible shapes in the score and subjectively perceived sonic shapes;
e.g. a tamtam strike may be indicated in the score as a single onset point in time,
perhaps with some dynamic marking and indication of the type of mallet to be
used, yet the result in performance is a protracted and extremely complex sound.
Evidently, timbral features are in general not well represented in western
notation because of its focus on pitch and duration. And as we know, this focus
has tended to leave expressive features of pitch, dynamics and timing outside
the mainstream conceptual apparatus, relegating these to the domain of perfor-
mance practice, a focus that has led to problems when attempting to represent
music of other cultures by western music notation transcriptions. But within
this pitch-and duration-focused western musical culture, we have also seen
some further abstractions from perceptual features, such as at times disregard-
ing octave placement, equating for instance an octave-compressed chord with
a widely spaced chord.1 Similar distortions of perceptually salient pitch shape,
and also of rhythm–shape relationships, are found in twentieth-century serial
and integral serial music, as well as in so-called pitch-class set theory, effectively
resulting in what could be called a ‘spatiotemporal collapse’ of salient percep-
tual features (see Godøy 1997 for a discussion of this).
Key-postures, trajectories and sonic shapes 7
conceptualizing features that can be ordered into more abstract symbolic sys-
tems such as those of pitch and duration, it has not been well suited to con-
ceptualizing more continuous, composite and multidimensional features. In
assessing the work of Schaeffer and followers, we find the idea of using various
shape images as a nonsymbolic means for feature representation to have been
an attractive solution, something that we now see has an affinity with body
motion (Godøy 2006).
Shape ontologies
Shape cognition
as well as its challenge, in our context: if we do not somehow have such over-
views of lived experience and are just submerged in a continuous stream of
sensations we will not be able to make sense of the world in general or of music
in particular, as was pointed out by Edmund Husserl more than a century ago
(Husserl 1991). To Husserl, it was obvious that we need to interrupt the con-
tinuous stream of sensations from time to time, and make overview images of
whatever is being perceived, by a series of intermittent ‘now-points’ (Godøy
2010b). Shape cognition could then be defined as our capacity to capture and
handle the ephemeral and temporally distributed features of music, as well as
other lived experience. And with presently available methods and technologies
for recording and processing both sound and body motion, we have the pos-
sibility of ‘freezing’ transient sound and motion and examining them at leisure
as shapes.
Historically, one of the first and most extensive projects on shape cognition
originated in music with gestalt theory in the last decades of the nineteenth cen-
tury (Smith 1988), with, among other things, a focus on how shapes emerge and
are conserved across different instances, such as melodies across various instru-
mental or vocal guises. Gestalt theory was later extended to other domains,
and is now often primarily associated with the visual. The remarkable insights
of early gestalt theory concerning coherence criteria in shape cognition still
have validity today, both in auditory perception (Bregman 1990) and in human
motor control (Klapp and Jagacinski 2011).
But one of the most extensive recent research efforts on shape cognition is
no doubt that of so-called morphodynamical theory (Thom 1983; Petitot 1985,
1990; Godøy 1997). The gist of morphodynamical theory is that human per-
ception, understanding and reasoning are based on ordering sensory input as
shapes, or in the words of René Thom, the leading figure of this theory, ‘the
first objective is to characterise a phenomenon as shape, as a ‘spatial’ shape. To
understand means first of all to geometrise’ (1983: 6).2
Of interest here is the morphodynamical distinction between the ‘control
space’ and the ‘morphology space’, meaning a distinction between the input
and the perceived results of any generative model (Petitot 1990), be that in
physics, biology, behavioural sciences or other domains such as musical sound.
It is always the perceived shapes—the features of the morphology space—that
are of interest for us here in musical contexts, and the distinction between con-
trol and morphology spaces helps us to determine what are ontologically com-
parable features and avoid various mapping mismatches or category mistakes
as mentioned above.
The distinction between control and morphology spaces is particularly use-
ful for exploring categorical thresholds between shapes. This means making
systematic explorations of perceived shapes by generating incrementally differ-
ent variants through what is often called analysis-by-synthesis (Risset 1991).
A simple but important example of this is the distinguishing of ‘percussive’
Key-postures, trajectories and sonic shapes 11
and ‘bowed’ sounds by the steepness of the attack segment at the beginning
of the sounds: with a very short attack we get the subjective sensation of a
percussive sound, and when gradually increasing the duration of the attack,
we sooner or later get a ‘bowed’ sound sensation. In other words: we explore
the thresholds between these two sound categories (features in the morphology
space) by incrementally varying the duration of the attack segment (a value in
the control space).
The analysis- by-synthesis approach enables exploration of perceptually
salient features by comparing incrementally different variants along several
feature axes, for example combining the incremental attack dimension (‘sharp-
ness’) with feature dimensions for spectral centroid (a measure of ‘brightness’
in timbre perception) in a two-dimensional analysis-by-synthesis exploration.
The analysis-by-synthesis approach is actually what people practise in music
production contexts, tweaking the buttons for equalizing, reverberation or
other kinds of effects processing in the mixing studio, or adjusting drum mem-
branes, instrument and microphone placement, and so on in the recording
room, until the ‘right’ sound is found. When individual musicians or conduc-
tors repeatedly try out versions of singular sounds or phrases until they find the
sonic expression they are searching for, they too are practising an analysis-by-
synthesis approach.
In summary, the analysis-by-synthesis approach is holistic in the sense of
allowing us to evaluate perceptual features of a whole chunk of sound, and in a
way it also bridges the symbolic-to-subsymbolic divide, which in the terminol-
ogy of Schaeffer is called the abstract–concrete divide: singular values along an
axis (or a scale) are abstract, whereas sonic objects with multiple features that
are holistically perceived as shapes are concrete (Schaeffer 1966; Chion 1983).
Related to analysis-by-synthesis is the idea of blending two shapes, an idea
that has become popularly known as the ‘morphing’ of visual images (such as
human faces) or of sounds, the latter case also being known as cross-synthesis.
There are various signal processing models for this, but it is also possible to
generate a series of incremental variants, say between sound A and sound B,
and explore the categorical threshold between the two.
The inherent challenge with such variant shape methods is that interesting
shapes are multidimensional: they usually cannot be characterized by only
one value axis so choices have to be made as to what aspect(s) are focused on.
The same goes for similarity ratings of differing shapes: Which part of the
shapes are we comparing, or are we going for a more global or cumulative
similarity judgement? In their pioneering work on categorization, Eleanor
Rosch and colleagues suggested that categories may be strongly linked with
motor schemas (Rosch et al. 1976): thus the category ‘chair’ may be difficult
to define from construction features alone (there are too many variants of
design, for example from rococo to modern) but is easier to categorize as
something to sit on.
12 Music and Shape
In sum, we can see that shape cognition in music, as well as in general, has to
do with features and categorical thresholds, and that the shape of body motion
can be an important part of understanding categories and shapes in music. It
follows that motor theory should be a part of shape cognition in music.
Motor theory
One leading idea in several domains of the cognitive sciences during the last
three decades has been to regard human cognition as rooted in bodily experi-
ence, as what has been broadly called embodied cognition. An essential fea-
ture of embodied cognition is that perception, thinking and understanding are
all related to mental simulation of body motion, meaning that we mentally
imitate the actions that we believe are the cause of what we perceive or that
actively trace one or more features of what we perceive. As suggested by Alain
Berthoz, with reference to Cézanne, seeing is a matter not just of passively tak-
ing in visual information with our eyes, but also of mentally tracing the outline
of what we are seeing, as if we are ‘touching’ whatever we see with our gaze
(Berthoz 1997).
In the case of spoken language, this means mentally simulating articula-
tory motions of the vocal apparatus when we listen to speech, and in the case
of music, mentally simulating the sound-producing body motions we believe
musicians are making: hearing ferocious drumming, we might imagine ener-
getic hand motions, while hearing soft string music, we might imagine slow and
protracted bowing motion. Such triggering of sound-producing images in lis-
tening means that we associate the shape of the sound-producing body motion
with the shape of the sound that we hear.
This theory of associations of sounds that we hear (or merely imagine) with
some kind of sound-producing body motion is known as the motor theory
of perception, sometimes referred to in the plural—‘motor theories’—because
several versions have been proposed. Originating in the 1960s in linguistics
(Liberman and Mattingly 1985), this can now be regarded as a more general
theory, also including other areas of human cognition (Galantucci, Fowler
and Turvey 2006). The gist of motor theory is that perception is production in
reverse, meaning that when we listen, we project motor images onto what we
are hearing and use these motor images as mental schemas to make sense out
of what we are hearing.
Motor theory concerns learning and expertise: if we are familiar with a lan-
guage or type of music, we probably know in more detail what body motion
goes into producing the sound; yet we may also have sketchier or vaguer motor
images of sounds we are not so familiar with. Although I myself speak nei-
ther Korean nor Polish, I believe I can distinguish these two languages by what
I perceive as their respective required phonological gestures. Having some
Key-postures, trajectories and sonic shapes 13
Shape timescales
The basic tenet of this chapter is that most features of music, ranging from
low-level acoustic and body-motion features to high-level affective and aes-
thetic features, are time-dependent, yet can also be thought of as shapes. Shape
images are in a sense ‘outside time’, to use the expression of Xenakis (1992):
they are ‘snapshots’ of what has unfolded or is about to unfold in time. This
raises issues of continuity versus discontinuity in musical experience (and other
time-related experiences for that matter), issues much focused on by philoso-
phers and psychologists in the nineteenth and twentieth centuries, in particu-
lar by Husserl as mentioned above (see Godøy 2008, 2010b, 2011, 2013). One
approach to this enigma of the temporal versus the atemporal may be to look
at constraints at work in our perception of sound and motion, in particular to
try to single out qualitative differences at the various timescales involved here.
As we know, human hearing is situated in the region of approximately 20
to 20,000 Hz (for healthy young people), with a threshold at around 20 Hz
for fused versus distinct features. This means that the timescale above 20 Hz
14 Music and Shape
Sonic features
These three main types have clear correlates in body motion: the sustained
sonic objects imply a continuous transfer of energy from the body, hence a
continuous effort such as bowing or blowing; the impulsive implies an abrupt,
discontinuous type of body motion, so-called ballistic motion, as in hitting
or kicking; and the iterative implies a rapid back-and-forth or shaking body
motion.
Furthermore, there are categorical thresholds in this typology, and we can
explore these thresholds by producing incremental variants as presented ear-
lier. If a sustained sound is shortened below a certain duration threshold, it
will be perceived as an impulsive sound, and conversely, if an impulsive sound
is extended beyond a certain duration threshold, it will be perceived as a sus-
tained sound. Likewise, if an iterative sound is slowed down to a certain rate,
it will turn into a series of distinct impulsive sounds, and conversely, if a series
of distinct impulsive sounds is accelerated beyond a certain rate, it will change
into an iterative sound. As we shall see later, these category changes are related
16 Music and Shape
5000
Frequency (Hz)
0
0 4.026
Time (s)
5000
Frequency (Hz)
0
0 4.026
Time (s)
FIGURE 1.2 The spectrogram of a sustained deep C double bass tone (top) and the spectrogram of
the same tone passed through a time-varying wahwah filter (bottom). The double bass tone has a
distinct burring sound, what could be referred to as a grain morphology feature in the terminology of
Schaeffer (1966), and the wahwah filtered version of this double bass tone has additionally a slower
open-close-open-close etc., a gait (or allure) morphology feature in the terminology of Schaeffer
(1966). At two different timescales, both grain and gait are clearly body-motion shape-related features,
i.e. grain making a fast shaking motion and gait making a slower opening and closing motion (cf. the
onomatopoetic associations of opening/closing the mouth in pronouncing ‘wahwah’).
Needless to say, we also often find uses of shape expressions designating more
traditional western music-theory-related sonic features in innumerable writings
on musical analysis, such as:
• Melodic features, such as contours, various kinds of patterns
• Harmonic features, both single chords and composite chord
progressions
• Modality, not as abstract pitch space (or scales) but as shapes of
interval constellations, referred to as ‘physiognomy’ by Lutosławski
(Norwald 1969)
• Rhythmical patterns and textures as shapes
In summary, we could say that most sonic features of musical experience could
be represented as a shape, bearing in mind the idea presented earlier that shape
is a fundamental cognitive strategy for making sense of the world. Yet there are
also a number of sonic features that are so close to body-motion features, as is
the case for rhythm and texture, that we need to have a look at what is sound
and what is body motion here.
18 Music and Shape
Body-motion features
There will be overlaps in many (perhaps most) cases between these catego-
ries, meaning that music-related body motion will also often be multifunc-
tional: some motion by a musician may, for instance, be both sound-producing
and communicative, such as an upward hand motion to prepare a fortissimo
chord on the keyboard, at the same time serving as an upbeat signal to the
other musicians, in addition to demonstrating a high level of energy to the
audience.
Besides observing body motion in performance that is not strictly sound-
producing, we may also readily observe body motion imitating sound-producing
Key-postures, trajectories and sonic shapes 19
when people listen to music, something that we have seen in so-called sound-
tracing, when listeners spontaneously draw (on a digital tablet or in the air) the
shape of sounds that they hear, an example of which can be seen in Figure 1.3.
More extensive study of sound-tracings, including statistical processing of cor-
relations between tracings and sound features, suggests that pitch contours are
quite robustly perceived as shapes, but also that dynamic and timbral features
1.2·104
Frequency (Hz)
0
0 4.841
Time (s)
FIGURE 1.3 The spectrogram of a distortion guitar sound with a downward glissando followed by a
slow upward expansion (top), and so-called sound-tracings of this sound by nine listeners (bottom).
The sound-tracings were made on a digital tablet by the listeners immediately after hearing the sound
for the first time, and should reflect something of how they spontaneously perceived the overall shape
of this sound.
20 Music and Shape
may likewise be spontaneously traced as shapes as long as there is not too much
competition between the features (Nymoen 2013).
The point here is that listening to or imagining music activates mental
images of some kind of music-related body-motion, and that these images are
one of the main sources for shape concepts in musical experience. Taking the
consequences of such close links between sonic and body-motion features, the
question arises as to the true nature of musical features such as rhythmical and
textural patterns: Are they sonic or body-motion patterns? For instance, is a
dance pattern (waltz, tango, samba) a sonic or body-motion pattern? Similarly,
is chunking in music based on sonic cues (sometimes referred to as qualitative
discontinuities in the sound) or on body-motion patterns? Our understanding
is that music includes both sonic and body-motion features, and that these fea-
tures are united in multimodal shape images although they actually emerge
from various constraints at work in the production of musical sound.
Constraint-based shapes
The fact that before the advent of electronic music technology music tradition-
ally was made by body motion in interaction with physical instruments or the
human vocal apparatus means that, in addition to body-motion constraints,
various instrument constraints imposed by physics are reflected in the resultant
sonic shapes. Observing that musical expression is ‘on top of’ instrumental and
body-motion constraints by no means diminishes the endless volitional expres-
sive capacities of music, but it should remind us to take various constraints on
sound production into account when we talk about shape in music.
To begin with, musical instruments have constraints, both in the mode of
excitation and in the subsequent energy dissipation: hitting a metal plate with
a hammer is an impulsive type of body motion, resulting in a sound with short
attack followed by a long decay. The perceived sonic shape is constrained here
by the size, shape and material of the metal plate and the hammer, and by the
force and duration of the impact. Instrumental and vocal sounds typically
have such overall envelope shapes, but may also have various internal textural
features as a direct physical response to excitations, for example the rough or
grainy sound of a deep double bass (bearing in mind the presentation of grain
earlier), or the hollow smooth sound of a high harmonic (flageolet) tone on
a violin.
In our music and shape context, it is interesting to consider so-called physi-
cal model sound synthesis as a way of thinking that takes physical constraints
into account, such as in a mathematical model that simulates the physical exci-
tation and resonance features of ‘real’ instruments or the human voice where
the resultant sonic shapes are constrained by the physical parameters of the
model. The point is that the behaviour of the physical model results in ‘real
Key-postures, trajectories and sonic shapes 21
world’ emergent sonic shapes, fitting with our ecological schemas of how sound
unfolds, in contrast to an abstract synthesis model such as additive synthesis,
where in principle any number of sinusoid components, with any frequency,
duration, fluctuations and so on, may be combined, and where there is really
no connection to the outside world except via those images we might project
onto the sound from previous experiences of similar features, by what is called
‘anthropomorphic projection’.
Instrumental or vocal performances in turn have their sets of constraints,
not just those we typically associate with different instruments—their idioms
or clichés (the things that are easy to play and sound well on an instrument)—
but also more general body-motion constraints that we believe contribute to
the shape of musical sound. Body-motion constraints, both biomechanical and
more neurocognitive (sometimes difficult to tell apart), effectively limit possible
body-motion range, speed and duration, and also necessitate rests and shifts
in posture to avoid fatigue and/or strain injury. Also, the fact that all human
body motion takes time, because it is not possible to move instantly from one
position to another, means that there always will be transition time between
positions. This in turn means that music-related body motion is continuous
(although it may at times appear as abrupt) and hence may result in fusion or
contextual smearing of otherwise singular sound onsets, apparent as so-called
phase-transitions and coarticulations.
Phase-transition designates changes in behaviour due to changes in some
parameter such as the speed and/or amplitude of body motion (Haken, Kelso
and Bunz 1985). In our context this means that otherwise singular motion-
units may fuse into a superordinate unit if the speed is increased, and con-
versely, a rapid motion may become split into distinct units if the speed is
decreased, as would be the case of a 3/4-time waltz pattern going from three
beats per measure to one beat per measure with increasing tempo, and con-
versely, from one beat per measure to three beats per measure with decreasing
tempo, similar to the transitions between sustained, impulsive and iterative
sounds mentioned above.
Coarticulation means that there is a fusion and contextual smearing of body
motion so that otherwise singular actions fuse into more superordinate trajec-
tories; in other words, body motion creates a context where the present state
of an effector (finger, hand, vocal tract) is determined by what was just done
as well as what is to be done next (Rosenbaum 1991). This means that there
are so-called carryover and anticipatory effects at work in sound p roduction,
something that has been quite extensively studied in linguistics (Hardcastle
and Hewlett 1999) but less so in music (Godøy, Jensenius and Nymoen 2010;
Godøy 2014). This coarticulatory fusion also has consequences for the sound
produced, contributing to a similar contextual smearing of sound and of
motion, resulting in continuous trajectories that in turn are one of the sources
of shape experience in music.
22 Music and Shape
FIGURE 1.4 The score of the first two bars of the last movement of Beethoven’s Piano Concerto
No. 1 (top), and graphs showing the position, velocity and acceleration of the vertical motion of the
right-hand knuckles, wrist (RWRA) and elbow (RELB) in the performance of these two bars. We
clearly see the up–down motion at the downbeats, i.e. at what we call the goal-points, as well as the
relative high velocity at these points, typical of so-called ballistic motion.
Motion-sound chunks
On the basis of our own and others’ research, then, we believe that there are sev-
eral elements of musical instruments, body motion and human cognition that
converge in singling out meso timescale motion-sound chunks as primordial for
the experience of shape in music, elements that may be summarized as follows:
• A number of findings in research on human motor control, memory
and attention point to the meso timescale as special in terms of
meaning in both perception and action.
• More specifically in music, the meso timescale is also sufficient for
perceiving a number of musically salient features such as rhythm,
texture, dynamics, timbre, melodic, harmonic and modal features,
style and genre, and sense of motion and affect.
24 Music and Shape
In the context of music and shape, the meso timescale motion-sound chunks
are clearly carriers of salient shape experiences in music:
• All sounds are included in some action trajectory, with various
principles of human motion such as phase-transition and
coarticulation contributing to emergent effects of fused body-motion
and sonic shapes; thus, there is a contextual smearing of otherwise
singular motion and sound elements within the fused chunk.
• This contextual fusion is evident in most musical features, but
in particular in tightly welded units such as various ornaments
(Pralltriller, mordent, turn, etc.) and other figures (all kinds of
rhythmical patterns such as waltz, tango, samba and so on) where
the speed and density of motion and sonic events typically are so
high that anticipatory cognition is required, so that these figures are
conceived and performed as singular, holistic body-motion shapes.
Motion-sound scripts
Although there is converging evidence that the meso timescale is crucial for per-
ceiving very many musical features, we also clearly experience music at longer
timescales: people go to performances of symphonies and operas, participate
in various long-lasting music-related events and rituals, or report long-endur-
ing trance-like experiences of music. Yet the perception of large-scale forms in
music seems not to be a well-researched topic. What we have is a substantial
number of western music analysis texts that assume the efficacy of large-scale
forms, but the little perceptual-empirical material that we have come across
suggests that we should be rather sceptical of such claims until further notice
(see for example Eitan and Granot 2008).
Lacking more systematic research in this area, we could assume from our
motor theory perspective that general principles of goal-directed motor cogni-
tion apply here, so that we understand long sequences as a series of key-postures
Key-postures, trajectories and sonic shapes 25
with intervening continuous motion trajectories and may also mentally quickly
run through a long stretch of music, just as we mentally run though a long walk
or a whole journey by a series of landmarks or junctions. This would essentially
amount to understanding large-scale musical works as extended motion-sound
scripts, as a series of concatenated and/or overlapping motion-sound chunks,
creating a sense of long-range continuity in musical experience. In addition to
the features of meso-timescale chunks, the macro timescale may often, by its
longer extension, have new dramaturgical and/or narrative features. We could
also speculate that such macro-level motion-sound scripts in turn could be
envisaged as having shapes, shapes that we could glimpse in an instant, just
as we could envisage a long walk or journey; in other words, the same prin-
ciple of ‘all-at-once’ overview images applies here too, as a kind of compressed
‘trailer’ or ‘story board’ for the whole work, as in the famous statement by Paul
Hindemith that ‘If we cannot, in the flash of a single moment, see a composi-
tion in its absolute entirety, with every pertinent detail in its proper place, we
are not genuine creators’ (2000: 61).
What we do know from our research on music-related body motion is that
we can see some salient global features over longer stretches, such as quan-
tity of motion (essentially a physical measure based on total displacement of
the body or parts of the body within a unit of time), recurrent patterns of
5000
Frequency (Hz)
0
0 Time (s) 62.84
FIGURE 1.5 The top part shows motiongrams (i.e. video-based summary images of motion
trajectories; see Jensenius 2013 for details) of three different successive dance performances by
the same dancer to a twenty-second excerpt from Lento from György Ligeti’s Ten Pieces for Wind
Quintet (Ligeti 1998), and the bottom part shows for the purpose of reference three repetitions of the
spectrograms of this excerpt. From this macro timescale view of the dancer’s body motion, we can
clearly see the overall shape (curve out from initial position and back) and mode of motion (mostly
calm but with a few abrupt elements).
26 Music and Shape
References
Godøy, R. I. and M. Leman, 2010: Musical Gestures: Sound, Movement, and Meaning
(New York: Routledge).
Godøy, R. I., E. Haga and A. Jensenius, 2006: ‘Playing “air instruments”: mimicry of
sound-producing gestures by novices and experts’, in S. Gibet, N. Courty and J.-F.
Kamp, eds., Gesture in Human-Computer Interaction and Simulation: 6th International
Gesture Workshop, Lecture Notes in Artificial Intelligence 3881 (Berlin: Springer), pp.
256–67.
Godøy, R. I., A. R. Jensenius and K. Nymoen, 2010: ‘Chunking in music by coarticulation’,
Acta Acustica united with Acustica 96/4: 690–700.
Goebl, W., S. Dixon, G. De Poli, A. Friberg, R. Bresin and G. Widmer, 2006: ‘ “Sense” in
expressive music performance: data acquisition, computational studies, and models’, in
P. Polotti and D. Rocchesso, eds., Sound to Sense, Sense to Sound: A State of the Art in
Sound and Music Computing (Berlin: Logos Verlag), pp. 195–242.
Grafton, S. T. and A. F. Hamilton, 2007: ‘Evidence for a distributed hierarchy of action
representation in the brain’, Human Movement Science 26: 590–616.
Haken, H., J. Kelso and H. Bunz, 1985: ‘A theoretical model of phase transitions in human
hand movements’, Biological Cybernetics 51/5: 347–56.
Hardcastle, W. J. and N. Hewlett, eds., 1999: Coarticulation: Theory, Data and Techniques
(Cambridge: Cambridge University Press).
Hindemith, P., 2000: A Composer’s World: Horizons and Limitations (Mainz: Schott).
Hogan, N. and D. Sternad, 2007: ‘On rhythmic and discrete movements: reflections,
definitions and implications for motor control’, Experimental Brain Research 181/1:
13–30.
Hunt, A., M. Wanderley and M. Paradis, M., 2003: ‘The importance of parameter map-
ping in electronic instrument design’, Journal of New Music Research 32/4: 429–40.
Husserl, E., 1991: On the Phenomenology of the Consciousness of Internal Time, 1893–1917,
trans. J. B. Brough (Dordrecht: Kluwer Academic).
Jensenius, A. R., 2007: ‘Action–sound: developing methods and tools to study music-related
body movement’ (PhD dissertation, University of Oslo).
Jensenius, A. R., 2013: ‘Some video abstraction techniques for displaying body movement
in analysis and performance’, Leonardo: Journal of the International Society for the Arts,
Sciences and Technology 46/1: 53–60.
Jensenius, A. R. and R. I. Godøy, 2013: ‘Sonifying the shape of human body motion using
motiongrams’, Empirical Musicology Review 8/2: 73–83.
Klapp, S. T. and R. J. Jagacinski, 2011: ‘Gestalt principles in the control of motor action’,
Psychological Bulletin 137/3: 443–62.
Liberman, A. M. and I. G. Mattingly, 1985: ‘The motor theory of speech perception
revised’, Cognition 21: 1–36.
Ligeti, G., 1998: Ten Pieces for Wind Quintet, on London Winds, György Ligeti Edition,
Vol. 7: Chamber Music (Sony SK 62309).
Loram, I. D., H. Gollee, M. Lakie and P. J. Gawthrop, 2011: ‘Human control of an inverted
pendulum: is continuous control necessary? Is intermittent control effective? Is intermit-
tent control physiological?’, The Journal of Physiology 589/2: 307–24.
McGurk, H. and J. MacDonald, 1976: ‘Hearing lips and seeing voices’, Nature 264: 746–8.
Norwald, O., 1969: Lutosławski (Stockholm: Norstedt).
Key-postures, trajectories and sonic shapes 29
Nymoen, K., 2013: Methods and technologies for analysing links between musical sound and
body motion (PhD dissertation, University of Oslo).
Peeters, G., B. L. Giordano, P. Susini, N. Misdariis and S. McAdams, 2011: ‘The timbre
toolbox: extracting audio descriptors from musical signals’, Journal of the Acoustical
Society of America 130/5: 2902–16.
Petitot, J., 1985: Morphogenèse du Sens I (Paris: Presses Universitaires de France).
Petitot, J., 1990: ‘Forme’, in Encyclopædia Universalis (Paris: Encyclopædia Universalis).
Risset, J.-C., 1991: ‘Timbre analysis by synthesis: representations, imitations and variants
for musical composition’, in G. De Poli, A. Piccialli and C. Roads, eds., Representations
of Musical Signals (Cambridge, MA, and London: MIT Press), pp. 7–43.
Rosch, E., C. B. Mervis, W. D. Gray, D. M. Johnson and P. Boyes-Braem, 1976: ‘Basic
objects in natural categories’, Cognitive Psychology 8: 382–436.
Rosenbaum, D., 1991: Human Motor Control (San Diego, CA: Academic Press).
Rosenbaum, D., R. G. Cohen, S. A. Jax, D. J. Weiss and R. van der Wel, 2007: ‘The problem
of serial order in behavior: Lashley’s legacy’, Human Movement Science 26/4: 525–54.
Schaeffer, P., 1966: Traité des objets musicaux (Paris: Éditions du Seuil).
Schaeffer, P. (with sound examples by G. Reibel and B. Ferreyra), [1967] 1998: Solfège de
l’objet sonore (Paris: INA/GRM).
Schäffer, B., 1976: Introduction to Composition (Warsaw: PWM Edition).
Sethares, W. A., 2007: Rhythm and Transforms (Berlin: Springer).
Smith, B., ed., 1988: Foundations of Gestalt Theory (Munich and Vienna: Philosophia
Verlag).
Thom, R., 1983: Paraboles et catastrophes (Paris: Flammarion).
Xenakis, I., 1992: Formalized Music, rev edn. (Stuyvesant, NY: Pendragon Press).
Reflection
Lucia D’Errico, guitarist and graphic designer
CHARGE
RECOIL ADVANCE
MELODIC
HARMONIC
BASS
RHYTHMIC
DISCHARGE
the musical frenzy of the party, the right hand of the piano weaves a suspended,
almost religious obbligato: ‘Dazwischen schluchzen und stöhnen/Die liebliche
Engelein’ (‘in between, sobbing and groaning,/the lovely little angels’). One
single sonic sensation contains the bodily giddiness of the happy couple and
the dejected inertia of the onlooker.
These four force lines need not be intended as vectors that cause a move-
ment throughout time, but rather as internal potentialities. Advance and recoil
are not forces that establish a chronological order; they interact with it, gen-
erating micro variations and perturbations inside a steady sequential grid.
Advance is not acceleration, but longing. Recoil is not ritardando, but lingering.
Additionally, this bodily shape, so complex and changing in itself, moves inside
another shape, which I would call architectural: the diachronic dimension of
music. Again, it is not an architecture one can see, but rather a space to cross
with blind eyes. This, depending on the levels of complexity, might resemble a
palace, a hut, or even a garden or a desert; it might have varying temperature
and light (but no optical shapes!). As a listener, I am led to move in unexplored
spaces. As a performer, it is I who is trying to lead someone else through an
architecture I know well. As a composer, I conceive this architecture first and
32 Music and Shape
then try to inhabit it until I am ready to distinguish and remember all of its
details.
Strange as it may sound, something very similar happens in my work as
a graphic designer. There are no optical shapes beforehand: there are forces,
which organize themselves on the empty canvas. The result is not predeter-
mined, but issues from the coagulation of these physical drives into visual ele-
ments. It is not a question of reproducing the visible, but of making visible
(Paul Klee). I ignore the subject I want to design, since it is dictated afterwards
by the arrangement of vectors I perceive somatically. For the same reason, the
habit of organizing music in an optical way as a timeline is as serviceable as it
is misleading. A musical experience is not the sonic rendering of a linear score.
On the contrary, a score should be nothing but the code, the deciphering of
which might recreate a planned spatial and haptic experience in the listener
through sound.
2
Music—as pertaining to the very act of shaping sounds over time during a
performance—engages most of our senses. As audience members in a concert,
we hear the musical sounds, we see the musicians on stage, and we feel the
rhythmic beat, only to realize that we have been tapping our finger to it, and
perhaps we taste a moment of sweetness during an intensely emotional passage.
Although the latter seems metaphorical, it also seems an apt description, sug-
gesting an underlying mapping from sound experience to taste (Knöferle and
Spence 2012). Even sitting at home and listening to a record in solitude with
eyes closed necessarily entails a multimodal experience as we map features of
the musical sound onto other domains, particularly the spatial and visual. That
is the central argument of the chapter. We feel the melodic line ascending and
descending; we feel we are moving or being moved forward, gently at times or
with sudden force; we sense the brightness or gloominess of some passages; or
perhaps we conjure up internal images that the music invoked in us and that
now become an integral part of our listening experience. How do we map music
onto other domains and why do we do it so readily? In this chapter, I address
the former question in some depth by reviewing studies on individuals’ draw-
ings and gestures in response to musical sounds. I introduce these multifaceted
shapes of sound and music as a way of studying music perception and cogni-
tion empirically, and outline methodological issues and challenges. To begin
with, however, I take a very brief look at some potential explanations for why
these cross-modal mappings may exist in the first place.
33
34 Music and Shape
How sound and music are mapped onto the visual and visuo-spatial domains—
with paradigms other than drawing or gesturing—has been reviewed at length
elsewhere (Eitan 2013; Spence 2011) and is not discussed here. However, it is
important to review the experimental paradigms underlying the vast majority
of empirical findings to date to be able to put drawing and gesturing approaches
into context.
To a large extent, increasing knowledge of cross-modal correspondences is
based on reaction-time paradigms that were developed by Garner in the 1960s
around the same time that the cognitive revolution gained momentum, with the
underlying metaphor of the human mind as a computer processing incoming
information.1 According to this view, sensory input from different modalities
is integrated at various levels of processing ranging from early sensory/percep-
tual levels to late semantic levels (for a review see Marks 2004). The speed with
which this processing occurs can be measured in behavioural experiments in
which participants respond to features of a dimension of a modality by press-
ing buttons which have been assigned certain feature values. In the simplest
case, there is only one modality involved, and features are varied only along
one dimension. For instance, participants may be asked to indicate as quickly
as possible whether the pitch (i.e. the relevant dimension) of a sound is high
or low, while the loudness (i.e. the irrelevant dimension) is kept constant. This
task—which has been termed ‘speeded identification’—often serves as a base-
line condition, involving two possible stimuli and two possible responses. If
the irrelevant dimension is varied as well (e.g. loudness: soft and loud), we get
four possible stimuli (high/soft, high/loud, low/soft, low/loud) while the num-
ber of possible responses is still two. In the latter scenario—‘speeded classifica-
tion’—participants’ task is to ignore the variation in the irrelevant dimension
(i.e. loudness) and indicate the feature value (high versus low) of the relevant
dimension (i.e. pitch). While these examples concern a single modality, there is
extensive research combining dimensions from several modalities (for a review
see Spence 2011). Whenever there are greater reaction times in comparison to
a baseline condition due to the variation of features in an irrelevant dimen-
sion or stimulus, this is referred to as ‘Garner interference’. On the other hand,
36 Music and Shape
Children’s drawings of sound and music have been studied extensively, creat-
ing a large body of empirical evidence and proving influential for studies with
adults. They are thus reviewed here in some depth before moving on to adults’
drawings of sound and music.
trained and untrained children aged seven to eight years and ten to eleven years
to listen to the first movement of Bach’s Brandenburg Concerto No. 2 and to
describe their listening experience both verbally, by creating a ‘listening map’,
and kinaesthetically, by moving their body, revealed that the most commonly
addressed ‘perceptual topics’ included ‘instrument, register, continuous motion,
formal sections, repetition, dynamics, tempo, contour, and pattern’ (ibid.: 36–7).
The type of visualization was dependent on age: the younger group created less
differentiated mappings—drawing pictures, the contour or the instruments—
whereas the older group used words and combinations of shapes to represent
both extramusical properties (e.g. mood) and musical parameters such as the beat.
Regarding the kinaesthetic responses, both groups depicted a broad variety of
musical parameters such as ‘beat, subdivided beat, articulation, melodic rhythm,
embellishment, duration, style, phrase, subphrases and motivic fragments, con-
tour, form, and pattern’ (ibid.: 42). Perhaps expectedly, both the visual and the
kinaesthetic responses were more differentiated than the verbal responses.
If the assumption that some musical experiences defy linguistic descriptions
is correct, the same should hold for adults. Indeed, some of the studies aimed
at uncovering aspects of children’s musical understanding through visual rep-
resentations have included adult participants as well. Davidson et al. (1988)
reported that invented notations of ‘Happy Birthday’ by seven-year-olds are
comparable to those of ten-year-olds and untrained adults. Moreover, it was
revealed that children older than nine years, as well as musically untrained
adults, show very stable figural representations, while only participants able to
read music display fully developed metric representations (Bamberger 1982).
Smith et al. (1994) found similar drawings of rhythmic sequences across groups
of musically untrained children and trained and untrained adults. In the next
section I focus on adults’ drawings of sound and music in more detail.
the strategy they had applied for the pure tones and chose to create pictorial
representations based on associative ideas.
Finally, a drawing study within a clinical setting is worth mentioning here.
De Bruyn and colleagues (2012) asked a group of participants with Autism
Spectrum Disorder (ASD) and a group of controls to draw along with vari-
ous musical excerpts on an electronic graphics tablet, focusing on either the
rhythmic structure or the melodic contour. Results revealed that both groups
performed equally well in the rhythm condition, but participants with ASD
performed slightly better in the melody condition. Overall, the results are inter-
preted as evidence that patients with ASD have no difficulty imitating struc-
tural aspects of the music.5
All of the drawing studies reviewed here can be regarded as involving special
(two-dimensional) types of gestures as well—gestures with the side effect of
creating a visible trace on paper or screen. Next, I turn to empirical evidence of
‘proper’ three-dimensional gestures in response to sound and music.
Küssner et al. (2014) ran a similar experiment with adult musically trained
and untrained participants who were presented with a series of pure tones con-
currently varying in pitch, loudness and tempo. The authors found that—just
as with drawing approaches—musically trained participants show more accu-
rate pitch–height mappings. It was also revealed that the bias for increasing–
decreasing contours does not hold for musically trained participants. There
were multiple strategies for representing the loudness, depending on the com-
plexity of the stimulus: if only loudness was changed, participants associated
the loudness with both the y axis and the z axis; in more complex stimuli, loud-
ness was associated with muscular energy, operationalized as fast shaking-
hand movements. Tempo was associated with the speed of hand movement
and elapsed time with the x axis (see also Küssner 2014). Moreover, this study
revealed interaction effects between the concurrently manipulated auditory
features—such as pitch and loudness affecting the association between tempo
and speed of hand movement—suggesting that gestural mappings of isolated
musical parameters should not automatically be generalized to more complex
auditory stimuli such as music.
Another study investigating isolated musical parameters has been carried
out by Nymoen et al. (2011), in which they asked participants to move a rod in
response to pitched and nonpitched sounds. It was revealed that pitch was most
strongly associated with the vertical axis and loudness with speed and hori-
zontal movements. Using a more restricted instruction, Kozak, Nymoen and
Godøy (2012) asked their participants to carry out either smooth or discon-
tinuous circular hand movements in response to sound stimuli manipulated in
rhythmic complexity, attack envelope, pitch, loudness and brightness. Focusing
on individuals’ ability to synchronize with the sound stimuli, they found that
discontinuous movement patterns resulted in better synchronization, with
musically trained participants performing more accurately in some trials only,
but never worse than untrained participants. Moreover, smooth attack enve-
lopes resulted in more motion, regardless of musical training.
Investigating gestural responses to everyday sounds— and more specifi-
cally, action-and nonaction-related sounds—Caramiaux et al. (2014) tested
the hypothesis that action-related sounds would give rise to sound-producing
gestures, whereas nonaction-related sounds would entail the representation
of their sonic shape. Confirming their hypothesis, the authors discovered that
speed profiles of participants’ movements were more similar for nonaction-
than for action-related sounds. It was suggested that the identification of the
source of an action-related sound (e.g. pouring cereal into a bowl) leads to
more idiosyncratic hand gestures than tracing the sonic shapes of nonaction-
related sounds.
There are very few studies investigating how adults represent music—that is,
‘real’ musical excerpts as opposed to a set of musical features (pitch, loudness,
timbre, etc.)—with three-dimensional hand gestures. Haga (2008) asked three
Shape, drawing and gesture 45
effort and shape categories. Results revealed that all the movement features
clearly differentiated between the two types of excerpt. For instance, if the
average value for ‘acceleration’ was high for the heroic passages, it was low for
the lyric passages. Moreover, there was an effect of musical training, show-
ing that trained participants achieved higher values for the movement features
‘size’ and ‘height’. This suggests that they moved more and filled more space
with their gestures during the experiment, possibly because they were more
familiar (and comfortable) with the music. Regarding the analysis of the lin-
guistic expressions, there was much agreement among the participants as to
how well a particular adjective described the expressive qualities of the music.
Furthermore, it was found that the extremes of the movement features corre-
lated with the extremes of the adjective scales such that an excerpt which was
rated, for instance, as conveying the expressive qualities ‘big’, ‘broad’, ‘thick’
and ‘exalting’ also gave rise to a high value for the movement feature ‘size’. The
authors interpret their findings as evidence for the sharing of expressive quali-
ties of music in linguistic expressions and body movements.
Having reviewed both drawing and gesture studies and shown the diversity
of contexts in which they were carried out, I now focus on some methodologi-
cal issues and how they can be addressed in future studies.
Methodological issues
When cross-modal mappings of auditory stimuli are studied, the outcome will
depend to a large extent on the specifics of the experiment such as the choice
of stimuli, the experimental setting and the instruction given to participants.
By discussing some of the issues involved I hope to provide a helpful, if by no
means exhaustive, overview for researchers who wish to carry out experiments
on cross-modal mappings of sound and music.
MUSIC VERSUS SOUND
This dichotomy is not specific to the study of cross-modal mappings but can
be found in any other field in which researchers have to face the problem of the
whole versus its parts. Unlike psycho-acousticians who exclusively work with
highly controlled, synthesized sound stimuli, music researchers are particularly
concerned with the unravelling of cross-modal mappings of real music, and a
broadly accepted way to study these is to investigate its constituent parts such
as pitch, loudness or timbre. As I have shown above, the problem with study-
ing characteristics of musical sounds in isolation (e.g. change in pitch) is the
creation of an ontological gap: we cannot be sure that findings from studies
using synthesized pure tones in order to investigate cross-modal mappings of
pitch apply equally to situations in which we listen to the changing pitches of
Shape, drawing and gesture 47
a musical performance. There are too many other factors involved in the latter
that render generalizations problematic. On the other hand, the choice of ‘real’
musical excerpts as experimental stimuli gives rise to a number of confounding
variables since it is unavoidable that other musical qualities such as dynamics
or articulation—or at least timbral qualities—will be co-varied with pitch. This
makes it difficult, if not impossible, to study causal links. I therefore suggest
that researchers should, whenever possible, include both types of stimuli in
their experiments (e.g. Eitan and Timmers 2010; Küssner and Leech-Wilkinson
2014) in order to get a better idea of the extent to which findings from highly
controlled psycho-acoustical stimuli hold true for musical excerpts, and also of
the extent to which findings from studies using musical excerpts can be repli-
cated by manipulating the musical sound feature of interest in isolation.
Although arguably simplistic compared to real musical excerpts, pure tones can
be synthesized with varying degrees of complexity. However, most studies so
far—at least those concerned with music cognition—have included pure tones
whose features were manipulated in isolation. There is scope for many more
studies using controlled pure tones (or more naturally sounding ones, such as
MIDI sounds) whose features are concurrently varied in a systematic manner
(for recent examples see Eitan and Granot 2011; Küssner et al. 2014). As men-
tioned above, in most cases music consists of the dynamic co-variation of sev-
eral musical parameters. These co-variations may, to some extent, be recreated
48 Music and Shape
in the synthesis of pure tones, achieving more ecologically valid stimuli while
keeping possible confounding variables at a minimum.
Apart from the ‘liveness’ aspect, there is evidence that individuals’ motor activ-
ity during listening affects their cross-modal mappings of music. For instance,
it has been suggested that the motor behaviour during listening influences chil-
dren’s visual representations of musical excerpts (Fung and Gromko 2001). A
group of children allowed to move with props or in sand while listening to the
music produced visualizations that included more detailed representations
Shape, drawing and gesture 49
The nature of the task, including experimental stimuli but also the exact word-
ing of the instruction and participants’ interpretation of it, determines what
is being assessed during an experiment. As Rusconi and colleagues (2006)
pointed out in a critique of some classic psychophysical experiments investi-
gating pitch–height mappings, there is a crucial difference between spontane-
ous and mandatory mappings. Spontaneous cross-modal mappings are seen as
occurring automatically, independent of the context and possibly without our
being aware of it, whereas mandatory mappings require our full consciousness
and deliberate action. At best, the latter are used to refine some finding well
supported by empirical evidence; at worst, they introduce highly artificial cat-
egories to an experiment, leading to meaningless responses.
Besides mandatory cross-modal mappings, which are restrained by a lim-
ited choice of response categories, there are also what might be called elabo-
rate responses. Whether spontaneous or not,8 they constitute free, unrestricted
responses to some stimulus, for instance by drawing a sound or a piece of
music. While such paradigms provide richer data than, for instance, reaction
time measures, they often also require some unstandardized analysis proce-
dure, complicating the comparisons between studies. Whichever paradigm
researchers apply after weighing advantages and disadvantages, it is important
that they are aware of the kind(s) of cross-modal mappings they are measuring.
Future directions
Cross-modal mappings of sound and music have been studied for a long time,
dating back to the work of Carl Stumpf (1883), who investigated metaphori-
cal mappings of pitch, for instance. Matching tasks and reaction-time para-
digms dominated psychological studies on cross-modal correspondences in the
twentieth century and have given rise to an impressive amount of empirical
evidence (Spence 2011), with ample opportunity for future studies. However,
the advent of the embodied cognition research programme (Shapiro 2007) has
led to a rethinking of cognition and put considerable emphasis on the role of
the body and its interaction with the physical environment in cognitive pro-
cesses. Consequently, epistemologies are changing and new paradigms are
being developed that consider more carefully the role of the body in psycho-
logical experiments. In musicology, the formalization of an embodied music
cognition theory (Leman 2007) has given a new impulse to studying music cog-
nition with the direct involvement of the body. Leman’s ‘graphical attuning’
and Godøy’s (2006) ‘sound-tracing’ represent new ways of studying sound and
music cross-modally.
This progress would not have been possible, of course, without the devel-
opment of new technologies. Electronic graphics tablets and motion-capture
systems allow researchers to measure participants’ responses to sound and
music with unprecedented precision. Importantly, and in line with the notion
of embodied, goal-directed actions in real time, they provide insights into
the shaping of cross-modal correspondences rather than its final products—the
shapes. However, such approaches come with issues pertaining to both the equip-
ment and the analysis techniques. For instance, some motion-capture systems
are still very expensive or require custom-made software (Küssner et al. 2011).
Another problem is that participants have to wear markers (Maes et al. 2014),
move large joysticks (Leman et al. 2009) or hold a remote controller in their
hand (Küssner et al. 2014) in order to indicate the shapes of auditory stimuli
with their bodies. Thus, one of the challenges for (music) researchers will be to
develop tools that are both less intrusive and less costly.
Crucially, techniques for analysing tracings of sound and music need to be
developed further. First attempts have been made for the analysis of drawings
(Noyce, Küssner and Sollich 2013), as well as for free three-dimensional gestures
(Nymoen et al. 2013), identifying techniques such as Gaussian Processes, non-
parametric and canonical correlations, and pattern recognition classifiers. Due
to music’s unfolding nature over time, the issue of analysing time-dependent
Shape, drawing and gesture 51
Conclusion
Acknowledgements
This work was supported by King’s College London and by the AHRC
Research Centre for Musical Performance as Creative Practice (grant number
RC/AH/D502527/1).
References
De Bruyn, L., D. Moelants and M. Leman, 2012: ‘An embodied approach to testing musi-
cal empathy in participants with an autism spectrum disorder’, Music and Medicine
4/1: 28–36.
Dolscheid, S., S. Shayan, A. Majid and D. Casasanto, 2013: ‘The thickness of musical pitch:
psychophysical evidence for linguistic relativity’, Psychological Science 24/5: 613–21.
Egermann, H., M. Pearce, G. Wiggins and S. McAdams, 2013: ‘Probabilistic models of
expectation violation predict psychophysiological emotional responses to live concert
music’, Cognitive, Affective, & Behavioral Neuroscience 13/3: 533–53.
Eitan, Z., 2013: ‘How pitch and loudness shape musical space and motion: new findings
and persisting questions’, in S.-L. Tan, A. Cohen, S. Lipscomb and R. Kendall, eds.,
The Psychology of Music in Multimedia (Oxford: Oxford University Press), pp. 161–87.
Eitan, Z. and R. Y. Granot, 2006: ‘How music moves: musical parameters and listeners’
images of motion’, Music Perception 23/3: 221–48.
Eitan, Z. and R. Y. Granot, 2011: ‘Listeners’ images of motion and the interaction of
musical parameters’, paper presented at the 10th Conference of the Society for Music
Perception and Cognition (SMPC), Rochester, NY, USA, 11–14 August 2011.
Eitan, Z. and R. Timmers, 2010: ‘Beethoven’s last piano sonata and those who follow croc-
odiles: cross-domain mappings of auditory pitch in a musical context’, Cognition 114/3:
405–22.
Elkoshi, R., 2002: ‘An investigation into children’s responses through drawing, to short
musical fragments and complete compositions’, Music Education Research 4/2: 199–211.
Espeland, M., 1987: ‘Music in use: responsive music listening in the primary school’, British
Journal of Music Education 4/3: 283–97.
Fitts, P. M. and C. M. Seeger, 1953: ‘SR compatibility: spatial characteristics of stimulus
and response codes’, Journal of Experimental Psychology 46/3: 199–210.
Fung, C. V. and J. E. Gromko, 2001: ‘Effects of active versus passive listening on the qual-
ity of children’s invented notations and preferences for two pieces from an unfamiliar
culture’, Psychology of Music 29/2: 128–38.
Galantucci, B., C. A. Fowler and M. T. Turvey, 2006: ‘The motor theory of speech percep-
tion reviewed’, Psychonomic Bulletin & Review 13/3: 361–77.
Godøy, R. I., 1997: ‘Knowledge in music theory by shapes of musical objects and sound-
producing actions’, in M. Leman, ed., Music, Gestalt, and Computing: Studies in
Cognitive and Systematic Musicology (Berlin: Springer), pp. 89–102.
Godøy, R. I., 2003: ‘Motor-mimetic music cognition’, Leonardo 36/4: 317–19.
Godøy, R. I., 2006: ‘Gestural-sonorous objects: embodied extensions of Schaeffer’s con-
ceptual apparatus’, Organised Sound 11/2: 149–57.
Godøy, R. I., 2010: ‘Gestural affordances of musical sound’, in R. I. Godøy and M. Leman,
eds., Musical Gestures: Sound, Movement, and Meaning (New York: Routledge).
Godøy, R. I., E. Haga and A. R. Jensenius, 2006: ‘Exploring music-related gestures by
sound-tracing: a preliminary study’, paper presented at the 2nd ConGAS International
Symposium on Gesture Interfaces for Multimedia Systems, Leeds, UK, 9–10 May 2006.
Gromko, J. E., 1994: ‘Children’s invented notations as measures of musical understanding’,
Psychology of Music 22/2: 136–47.
Gromko, J. E., 1995: ‘Invented iconographic and verbal representations of musical sound:
their information content and usefulness in retrieval tasks’, The Quarterly Journal of
Music Teaching and Learning 6: 32–43.
54 Music and Shape
Haga, E., 2008: ‘Correspondences between music and body movement’ (PhD dissertation,
University of Oslo).
Hair, H. I., 1993: ‘Children’s descriptions and representations of music’, Bulletin of the Council
for Research in Music Education 119: 41–8.
Hargreaves, D. J., 1978: ‘Psychological studies of children’s drawing’, Educational Review
30/3: 247–54.
Hooper, P. P. and E. R. Powell, 1970: ‘Influences of musical variables on pictorial connota-
tions’, Journal of Psychology 76/1: 125–8.
Huron, D. and D. Shanahan, 2013: ‘Eyebrow movements and vocal pitch height: evidence
consistent with an ethological signal’, The Journal of the Acoustical Society of America
133/5: 2947–52.
Huron, D., S. Dahl and R. Johnson, 2009: ‘Facial expression and vocal pitch height: evi-
dence of an intermodal association’, Empirical Musicology Review 4/3: 93–100.
Kerchner, J. L., 2000: ‘Children’s verbal, visual, and kinesthetic responses: insight into their music
listening experience’, Bulletin of the Council for Research in Music Education 146: 31–50.
Knöferle, K. and C. Spence, 2012: ‘Crossmodal correspondences between sounds and
tastes’, Psychonomic Bulletin & Review 19/6: 992–1006.
Kohn, D. and Z. Eitan, 2009: ‘Musical parameters and children’s movement responses’, in
J. Louhivuori, T. Eerola, S. Saarikallio, T. Himberg and P. S. Eerola, eds., 7th Triennial
Conference of the European Society for the Cognitive Sciences of Music (Jyväskylä: ESCOM).
Kozak, M., K. Nymoen and R. I. Godøy, 2012: ‘Effects of spectral features of sound
on gesture type and timing’, in E. Efthimiou, G. Kouroupetroglou and S.-E. Fotinea,
eds., Gesture and Sign Language in Human– Computer Interaction and Embodied
Communication (Berlin: Springer), pp. 69–80.
Küssner, M. B., 2013: ‘Music and shape’, Literary and Linguistic Computing 28/3: 472–9.
Küssner, M. B., 2014: ‘Shape, drawing and gesture: cross-modal mappings of sound and
music’ (PhD dissertation, King’s College London).
Küssner, M. B. and D. Leech-Wilkinson, 2014: ‘Investigating the influence of musical
training on cross-modal correspondences and sensorimotor skills in a real-time drawing
paradigm’, Psychology of Music 42/3: 448–69.
Küssner, M. B., N. Gold, D. Tidhar, H. M. Prior and D. Leech- Wilkinson, 2011:
‘Synaesthetic traces: digital acquisition of musical shapes’, presented at the Supporting
Digital Humanities Conference: Answering the unaskable, Copenhagen, Denmark, 17–
18 November 2011.
Küssner, M. B., D. Tidhar, H. M. Prior and D. Leech-Wilkinson, 2014: ‘Musicians are
more consistent: gestural cross-modal mappings of pitch, loudness and tempo in real-
time’, Frontiers in Psychology 5/789, https://doi.org/10.3389/fpsyg.2014.00789 (accessed
9 April 2017).
Lakoff, G. and M. Johnson, 1980: Metaphors We Live By (Chicago: University of Chicago
Press).
Leman, M., 2007: Embodied Music Cognition and Mediation Technology (Cambridge,
MA: MIT Press).
Leman, M., F. Desmet, F. Styns, L. Van Noorden and D. Moelants, 2009: ‘Sharing musical
expression through embodied listening: a case study based on Chinese Guqin music’,
Music Perception 26/3: 263–78.
Levitin, D. J., R. L. Nuzzo, B. W. Vines and J. O. Ramsay, 2007: ‘Introduction to functional
data analysis’, Canadian Psychology/Psychologie canadienne 48/3: 135–55.
Shape, drawing and gesture 55
Smith, K. C., L. L. Cuddy and R. Upitis, 1994: ‘Figural and metric understanding of
rhythm’, Psychology of Music 22/2: 117–35.
Spence, C., 2011: ‘Crossmodal correspondences: a tutorial review’, Attention, Perception, &
Psychophysics 73/4: 971–95.
Spence, C. and O. Deroy, 2012: ‘Crossmodal correspondences: innate or learned?’, i-Perception
3/5: 316–18.
Stumpf, C., 1883: Tonpsychologie (Leipzig: S. Hirzel).
Suzuki, S., E. Mills and T. C. Murphy, 1973: The Suzuki Concept: An Introduction to a
Successful Method for Early Music Education (Berkeley, CA: Diablo Press).
Tan, S.-L. and M. E. Kelly, 2004: ‘Graphic representations of short musical compositions’,
Psychology of Music 32/2: 191–212.
Thompson, M., 2012: ‘The application of motion capture to embodied music cognition
research’ (PhD dissertation, University of Jyväskylä).
Trehub, S. E. and L. Trainor, 1998: ‘Singing to infants: lullabies and play songs’, in C. Rovee-
Collier, L. P. Lipsitt and H. Hayne, eds., Advances in Infancy Research, Vol. 12 (Stamford,
CT: Ablex), pp. 43–78.
Trimble, O. C., 1934: ‘Localization of sound in the anterior-posterior and vertical dimen-
sions of “auditory” space’, British Journal of Psychology: General Section 24/3: 320–34.
Upitis, R., 1987: ‘Children’s understanding of rhythm: the relationship between develop-
ment and music training’, Psychomusicology: Music, Mind & Brain 7/1: 41–60.
Upitis, R., 1990: ‘Children’s invented notations of familiar and unfamiliar melodies’,
Psychomusicology: A Journal of Research in Music Cognition 9/1: 89–106.
Upitis, R., 1992: Can I Play You My Song? The Compositions and Invented Notations of
Children (Portsmouth, NH: Heinemann).
Van Dyck, E., 2013: ‘The influence of music and emotion on dance movement’ (PhD dis-
sertation, Ghent University).
Verschaffel, L., M. Reybrouck, M. Janssens and W. Van Dooren, 2010: ‘Using graphical
notations to assess children’s experiencing of simple and complex musical fragments’,
Psychology of Music 38/3: 259–84.
Walker, P., J. G. Bremner, U. Mason, J. Spring, K. Mattock, A. Slater and S. P. Johnson, 2010:
‘Preverbal infants’ sensitivity to synaesthetic cross-modality correspondences’, Psychological
Science 21/1: 21–5.
Walker, R., 1987: ‘The effects of culture, environment, age, and musical training on choices
of visual metaphors for sound’, Perception & Psychophysics 42/5: 491–502.
Werner, H., 1980: Comparative Psychology of Mental Development, 3rd edn (New York:
International Universities Press).
Whitney, K., 2013: ‘Singing in duet with the listener’s voice: a dynamic model of the joint shap-
ing of musical content in live concert performance’, paper presented at the Performance
Studies Network Second International Conference, Cambridge, UK, 4–7 April 2013.
Zbikowski, L. M., 2002: Conceptualizing Music: Cognitive Structure, Theory, and Analysis
(New York: Oxford University Press).
Reflection
Anna Meredith, composer
Shape is both the most important aspect of my composing and the hardest
thing to describe. Before I write any piece, whether a piece for orchestra or an
electronic track, I draw a sketch of its contour along a timeline; so my drawers
are stuffed with pages of jaggy lines, builds and cuts which help me control my
pacing—one of the most important things to me in my music. One of these
sketches and its associated composition can be accessed at .
As to what the lines mean, that’s harder to pin down. At the risk of sounding
flaky, I think the best description might be that they are tracing the energy of
a piece. So a big diagonal build on my sketch might not necessarily mean ‘get
louder’ or ‘get faster’ but could suggest a way of controlling the musical energy
of an idea as my way of showing the trajectory of a line or fragment I’ve come
up with.
When I’m writing a piece, it feels like this drawing/sketching process is my
way of auditioning my ideas: so if I’ve got something, no matter how little,
I then imagine it going through the dramatic shapes I need for the piece to
see if the material will be appropriate. This involves keeping half an eye on a
stopwatch while striding round my studio tunelessly singing bits of the mate-
rial and muttering things like ‘idea breaks apart and glitches here’ or ‘melodic
line builds until it takes over whole ensemble’, to see if I think it’ll work. Once
I’ve got the right ideas, and am confident that they’ll stand up to the drama I’ve
got planned for them, my next step becomes more of a zooming in, looking at
part of the shapes, working out exactly how I get from A to B and filling in the
detail.
57
3
spatial height (Walker et al. 2010; Wagner et al.1981; see Lewkowicz and Minar
2014 for a critique), pitch and visual shape (e.g. round versus sharp; Walker et
al. 2010), pitch and luminance (Ludwig, Adachi and Matsuzawa 2011), pitch
and physical size (Morton 1994; see also Tsur 2006), and loudness and lumi-
nance (Lewkowicz and Turkewitz 1980). As can be seen from these examples,
mappings of auditory features onto visual-spatial dimensions in particular are
frequent, highlighting a possible central role of notions related to shape.
While experimental studies can suggest the kind of mappings expected to
play a role in music listening, the actual manifestation of cross-modal interac-
tion in music may be confounded by the diversity of mappings that might be acti-
vated simultaneously, and by contextual factors that influence the connotations
activated. We aim to demonstrate that, multiplicity and context-dependency
notwithstanding, an analysis of cross-domain mappings in music, informed by
experimental findings in cross-modal research, can elucidate important aspects
of musical meaning and reference. In particular, we examine the interrelation-
ship between two central pillars of musical meaning: cross-modal and emo-
tional mappings of musical features. Furthermore, we aim to demonstrate how
multiplicity of cross-modal interaction is instrumental in generating complex,
multilayered musical meanings, which in combination may often be most easily
and efficiently summarized by the metaphor of shape. Investigating a musical
setting of a text permeated with references to nonauditory sensory domains
may serve as a useful point of departure for such endeavour. We chose to con-
centrate on Schubert’s well-known (and oft-discussed) setting of Heine’s ‘Die
Stadt’ (from Schwanengesang D. 957), examining both score-based (composi-
tional) and performance-based features, the latter grounded on quantitative
analysis of recorded music.
MULTIPLICITY OF MAPPINGS
In the poem itself, however (Table 3.1), this narrative is revealed only at
the very end. The first stanza describes a city seen from afar at dusk. We are
given no explicit information about the narrator’s identity, emotions, actions or
whereabouts (indeed, a naïve reading of this stanza could ascribe it to a third-
person narrator, gazing impartially at a remote view). We know nothing of a
water-trip or of the grief of lost love. We don’t know who gazes at the town or
what (if anything) it means to him.
What we do obtain is considerable visual information. We know that the
town is seen from afar, at the horizon (Am fernen Horizonte). We know that
its outlines are veiled as a foggy image (wie ein Nebelbild) and rather dark,
shrouded in dusk (In Abenddämmrung gehüllt). We know quite a bit about
space and light, but little (at least explicitly) about anything else that matters.
While the first stanza presents a gaze at a remote and static object (the town),
the second stanza is a close-up shot of the narrator’s immediate surroundings
(water, boat, boatman’s rowing), involving both motion (Windzug kräuselt, rud-
ert) and emotion (traurigem). Furthermore, at the end of this stanza it is clear
that the poem is narrated by its own protagonist in a first-person narration (in
meinem Kahn) and that this protagonist is neither objective nor impartial. Even
the oar strokes are described as ‘mournful’ (traurigem Takte), though we can,
at this stage, only guess what the mourning is about.
In perceptual terms, then, the second stanza contrasts with the first with
regard to distance (far–near) and motion (dynamic–static; Table 3.2). Note
that in addition to dimensions of motion and colour (graue Wasserbahn), this
stanza also involves the tactile modality (feuchter Windzug), consistent with the
close-by perspective it presents. These changes in the depiction of perceptual
realms are in line with the changes in narrative perspective, stressing first-per-
son narrative and strong (though still subdued) emotions.
It is only in the final (third) stanza—indeed, only in its last line—that the
crux of the poem is revealed: we now know what the town means to the nar-
rator and why he keeps gazing at it from dusk to sunrise.1 Appropriately, the
Cross-modal correspondences in a Schubert song 63
agent of this narrative and its emotional revelation is the very source of clar-
ity: the rising sun itself.
The third stanza both parallels and contrasts with the first (see Table 3.3);
no less importantly, it complements it. Both stanzas involve viewing the same
object—the town—and both emphasize the perceptual dimension of visual
brightness. However, the two stanzas contrast with regard to the view itself,
as well as its emotional underpinnings. Visually, the scene is now bright and
painfully clear, highlighted by the glowing, rising sun, and thus contrasted with
the darker, dim view of the opening stanza. Moreover, the last stanza involves
motion and change (particularly upward motion, associated with positive and
active emotions), rather than stasis: the sun is ‘rising from the earth’ (vom
Boden empor).2
These ‘perceptual’ contrasts between the stanzas are accompanied by
emotional and narrative correlates, as the previously veiled connotations
of the distant town now become painfully clear to both reader and pro-
tagonist. However, from another perspective, perceptual metaphor and
emotional import are strikingly incongruous here. As mentioned, visual
brightness (luminosity) and lightness widely serve as metaphors for emo-
tional valence, such that brighter light and lighter colour correlate with
positive valence. This association, evident in verbal metaphor (Stimmung
64 Music and Shape
hellt sich auf—literally, mood brightens up),3 also affects behaviour and
cognition implicitly and automatically, as evidenced in diverse empirical
work (Meier and Robinson 2005; Meier et al. 2007). Similarly, spatial rise
correlates with active and positive emotion, as suggested by both language
metaphors (e.g. Die Stimmung steigt—the mood rises) and nonverbal exper-
imental measures (Casasanto and Dijkstra 2010; Freddi et al. 2013; Meier
and Robinson 2004).
In particular, the rising sun serves as a metaphor for ‘elevated’—hopeful,
cheerful and active—emotions:
• … he who kisses the joy as it flies /Lives in eternity’s sun rise (Blake,
‘Eternity’)
• But soft! What light through yonder window breaks? It is the east,
and Juliet is the sun! (Shakespeare, Romeo and Juliet, II/ii)
Such hopeful or joyful, ‘sunny’ emotions are often associated with renewal or
creation:
• The sunrise is a glorious birth (Wordsworth, ‘Intimations of
Immortality’)
• Was it light that spake from the darkness /Or music that shone from
the word /When the night was enkindled with sound /of the sun or
the first-born bird? (Swinburne, ‘Music: An Ode’)
• Himmelhoch jauchzend, zum Tode betrübt (heavenly rejoicing, then
deathly sorrowing; Goethe, Egmont)
In ‘Die Stadt’, in apparent incongruity with such metaphors and with a host
of empirical studies of the association of light and mood mentioned above, the
rising sun (which, importantly, is the subject and active agent of this stanza: it
‘shows’ the protagonist the town) evokes a ‘dark’ memory and a mood of
mournful, hopeless despair. In the next sections, we investigate how both
Schubert and some of his most prominent present-day performers encounter
this seeming contradiction, as well as other aspects of Heine’s imagery.
Schubert’s setting of Heine’s evocative text (see Appendix 3.1 ) has been
extensively analysed from diverse perspectives (e.g. Clark 2002; Hascher
2008; Kerman 1962; R. Kramer 1994; L. Kramer 2003, 2004; Morgan 1976;
Schwarz 1986; Youens 2007). As noted, our main goal in the present analysis is
elucidating how Schubert’s musical setting encounters cross-modal mappings
and their interactions with emotional expression, as suggested by the text. To
lay the ground for that analysis, however, we first present some observations
concerning the structure of Schubert’s song as it relates to Heine’s text.
Cross-modal correspondences in a Schubert song 65
The vocal sections of ‘Die Stadt’ present an ABA’ design, consistent with
the text’s structure, as described above. The third stanza repeats and comple-
ments the first, while the second stanza contrasts with both. The outer stanzas
(bars 6–14, 27–35), harmonically and melodically closed, are similar to each
other in their harmonic and melodic structures. They also present a similar
rhythmic structure (including the conspicuous dotted rhythms in the piano
accompaniment) and piano texture. The third stanza, however, contrasts with
the first in several conspicuous ‘surface’ features, particularly dynamics (f to ff,
contrasting with the overall pp of the first two stanzas) and register (the piano
accompaniment rises an octave and the bass is doubled). The vocal line also
rises higher than in the opening stanza (to g2, the highpoint of the entire song,
on ‘Liebste’, bar 34) and is more disjunct and angular, presenting the largest
melodic intervals in the song (fourths, fifths, minor sixth, octave; bars 29–31,
33–35). In contrast, the vocal line of both stanzas 1 and 2 presents only seconds
and thirds. Stanzas 1 and 3, then, while structurally almost identical,4 contrast
in conspicuous expressive aspects (dynamics, register, vocal contour, interval
size), the last stanza achieving a more dramatic, decisive closure, complement-
ing the tonally stable yet muted opening.
The musical setting of stanza 2 (bars 14–27), like its text, strikingly contrasts
with those of both outer stanzas. While the settings of both stanzas 1 and 3
depict a harmonically closed structure, a homophonic texture and an arched,
mostly ascending melodic contour, stanza 2 introduces a static yet dissonant
harmony throughout (the ambiguous diminished-seventh chord C–E♭–F♯–A,
which also shapes the melodic line), a florid, arpeggiated accompaniment fig-
ure and a continuously falling vocal contour (descending from the previously
established high point, e♭2, to c1). Together, these features embody a paradoxi-
cal combination of several metaphorical movements: rapid, repetitive surface
motion (the piano figuration), which is yet static (unchanging harmony) and
aimless overall, going nowhere (diminished-seventh chord, harmony devoid of
any clear tonal ‘direction’). This notwithstanding, a constant, steep fall under-
lies the entire stanza (the vocal contour).
Figures 3.1–3.3 quantitatively plot some of the relationships among the
three vocal stanzas, as described above. Figure 3.1 depicts the contour of the
vocal line (top, black line) expressed in terms of the weighted average pitch
per two-bar phrase (weighted according to note duration). Additionally, it
shows the mean absolute melodic interval per two-bar phrase (bottom, grey
line), which indicates how much the pitch of the vocal line varies in successive
two-bar phrases. Figure 3.2 plots the mean intensity (left) and the maximum
intensity (right) per two-bar phrase for three performances of ‘Die Stadt’ (to be
discussed separately later).5 Figure 3.3 plots the mean and standard deviation
of the rhythmic durations present in the vocal line per two-bar phrase.
The figures suggest a complex web of similarities and contrasts between the
three stanzas. Stanzas 1 and 3 are similar in melodic contour, both presenting
66 Music and Shape
75
16
70
12
65
8
60 4
0
55
0 2 4 6 8 10 12
Two-bar phrase
FIGURE 3.1 Mean weighted pitch (black line) and mean absolute pitch interval (grey line) per
two-bar phrase
an ascending contour (with stanza 3 rising higher), which contrasts with the
descending contour of stanza 2 (Figure 3.1, top). With regard to melodic inter-
vals, however, it is stanza 3, presenting larger intervals, which contrasts with
both stanzas 1 and 2. This pairing is also depicted by intensity (Figure 3.2):
stanza 3 presents (in all three performances) considerably higher intensity than
both earlier stanzas (intensity contours, which differ for all three stanzas, also
vary with performance, which will be discussed later). The rhythm of the vocal
line, on the other hand, shows a process of change (Figure 3.3), in which the
stanzas become more rhythmically diverse, in particular through the presence
of longer durations.
These complex interrelationships notwithstanding, the three vocal stan-
zas could present a fairly conventional narrative structure, in which the outer,
stable stanzas frame a central unstable one, with the last stanza intensifying
and dramatizing the concluding tonal closure through louder dynamics, higher
register and larger melodic intervals. Yet Schubert turns this ‘reasonable’ form
upside-down (or rather, inside-out): he frames the vocal sections with introduc-
tory and concluding sections, both identical to the piano part of the central
second stanza, with its harmonically ambiguous diminished-seventh harmony
and florid arpeggiations.
The expressive and structural implications of this framing have been fre-
quently observed and debated in the critical and analytical literature (e.g. Clark
2002; Kerman 1962; Kramer 2003, 2004; Morgan 1976; Schwarz 1986; Youens
2007), and we do not address them at length. Two related outcomes of this
Mean intensity DFD Maximum intensity DFD
Mean intensity IB Maximum intensity IB
Mean intensity TQ Maximum intensity TQ
80 85
75 80
Intensity (dB)
70 75
65 70
60 65
55 60
0 2 4 6 8 10 12 0 2 4 6 8 10 12
Two-bar phrase Two-bar phrase
FIGURE 3.2 Mean intensity (left) and maximum intensity (right) per two-bar phrase for three performers. Intensity was measured from commercial
recordings combining the piano and the vocal line. Interruptions in lines indicate bars that are separated by piano accompaniment intermezzi.
68 Music and Shape
0.8
Duration (crotchets)
0.6
AvDur
StdDur
0.4
0.2
0
0 2 4 6 8 10 12
Two-bar phrase
FIGURE 3.3 Average rhythmic durations (black line) of the vocal line and standard deviation of
rhythmic durations (grey line) within successive two-bar phrases
gambit should be pointed out, however. Structurally, it turns the song from
what could have been a tonally and narratively closed entity (as described
above) to an open one—perhaps (as Morgan 1976 and others suggest) as a
link to other songs in Schwanengesang.6 Narratively, rather than an intermedi-
ate stage connecting dusk (first stanza) and sunrise (third stanza), the nightly
rowing scene of the second stanza is also a frame for the entire song, sup-
plying the material for its opening and closing piano figuration. Due to this
framing, Schubert’s song now takes place in a constant, perhaps eternal limbo,
accompanied by Charon’s constant rowing, leading nowhere; thus, Schubert’s
foggy framing perhaps suggests who the boatman is and what ancient tale—the
Orphean tale of love lost—the narrator (and Heine) is trying to retell.
LIGHT
brighter), pitch height (higher/ brighter) and pitch direction (rising pitch/
brighter). Visual brightness is also associated with aspects of the sound spec-
trum, particularly spectral centroid (higher/brighter). Research also suggests
associations of colour lightness or brightness with modality (major is lighter
and brighter than minor; Bresin 2005, Palmer et al. 2013), tempo (faster music
associates with lighter colours; Palmer et al. 2013), and interval size (larger
melodic intervals associated with more extreme degrees of brightness or dark-
ness; Hubbard 1996).
Schubert’s ‘Die Stadt’ uses the most conspicuous of these correspondences
unequivocally. Thus, the dimensions contrasting the first and second stanzas,
set in dusk, and the third stanzas, depicting sunrise, are those most widely and
conspicuously associated with brightness: sound intensity (which has been
associated with visual brightness even in newborns; Lewkowicz and Turkewitz
1980) and pitch height (associated with colour lightness and brightness both in
humans and in other primates; Ludwig et al. 2011).
Sound intensity also affects the spectral structure of the musical sound
(both piano and vocal), such that louder sound emphasizes higher, ‘brighter’
spectral components; hence, loudness contrasts between the third stanza and
the preceding stanzas entail corresponding differences in spectral ‘brightness’
associated with visual brightness (Griscom and Palmer 2012). To examine
whether the analogy of visual brightness and spectral structure is expressed in
performances of ‘Die Stadt’, we calculated the median spectral centroid for the
three stanzas (piano solo sections excluded) for each performance (Figure 3.4).
Spectral centroid was measured using the Libxtract plugin available in Sonic
1600
Median spectral centroid (Hz)
1400
1200
DFD
1000 IB
TQ
800
600
1 2 3
Stanza
FIGURE 3.4 Median spectral centroid (Hz) per stanza for three performances of Schubert’s ‘Die Stadt’.
Spectral centroid was measured from commercial recordings combining piano and vocal line.
70 Music and Shape
Visualiser.7 As the figure shows, the increase in intensity in the third stanza is
indeed accompanied by a rise in spectral centroid (compared to second stanza)
for all performances, and for two of the three performances (TQ and DFD)
median spectral centroid in the third stanza is also considerably higher than in
the first stanza.
Additionally, larger melodic intervals emphasize cross- modal mappings
of pitch and brightness, producing more extreme (bright or dark) mappings
(Hubbard 1996). Hence, the concentration of the largest melodic intervals in
the setting of ‘Leuchtend vom Boden empor’ (literally, ‘glowing upwards from
the earth’) is telling.
A different type of allusion to light quality (yet unaccounted for by cross-
modal empirical research) involves the diminished-seventh sonority, which
frames the song and underlies its central stanza. Due to its symmetrical struc-
ture, the diminished-seventh chord is the most ambiguous sonority in the tonal
harmonic palette, and may be associated (in enharmonic interpretations) with
virtually every tonal centre. Though in its present context this ambiguity is
not exploited, the chord may serve as an apt symbol of the foggy visual (as
well as emotional) quality shrouding the song. Whether this high-level sym-
bolic association (grounded in tonal syntax, rather than basic perceptual cor-
respondences) also affects listeners’ perception is an intriguing question, which
remains to be empirically explored.
DISTANCE
MOTION
Textually, the three stanzas are clearly distinguished from each other in the
qualities of motion they suggest. The first stanza does not allude to motion
in any direct way. The second, in contrast, is full of motion, and suggests two
simultaneous types of movement: the erratic wind, creating ruffles in the water,
and the measured, ‘mournful’ oar strokes (Takte—also musical beats). The
third stanza is underlined by a single majestic motion: the rise of the sun ‘from
the earth’.
Schubert applies two types of mapping to suggest these motion qualities.
One is the direct analogy between temporal aspects of physical motion (e.g.
pace, regularity) and aspects of rhythm (IOI, metric accent). The other analogy
maps pitch space onto physical space, and thus pitch change (e.g. rise or fall,
steps or leaps) onto physical motion (for reviews of relevant cognitive research
see Eitan and Granot 2006; Eitan 2013). In the second stanza (bars 16–26),
both mappings are applied. The arc of rapid arpeggiation on first beats sug-
gests, through both rhythm and pitch contour, the wind and the water ripples
that the wind generates. The boatman’s Takte (the repeated, steady lowering of
the oars into the water) are alluded to by accented As, repetitively descending
two octaves (second and third beats). In the third stanza it is the pitch/space
analogy, applied in the vocal line, which suggests motion (the rising sun): the
rising vocal contour and the concentration of upward leaps (fifth, fourth) on
‘vom Boden empor’ (bar 30).
As noted above, cross-modal and emotional mappings are often highly cor-
related, in music and elsewhere. Low pitch, for instance, is associated with
darker (low lightness) or dimmer (low brightness) visual stimuli, as well as with
negative, low-intensity emotion (e.g. sadness). Correspondingly, negative emo-
tional states—‘dark’ emotions—are themselves associated with darker or dim-
mer visual stimuli. Similarly, high or ascending visual stimuli, high or rising
auditory pitch, and positive, ‘uplifting’ moods are also cognitively associated.
Cross-modal correspondences in a Schubert song 73
We now examine in more detail three performances of the score and how these
modify or add to the observed cross-modal mappings and affective associa-
tions. We focus here on performers’ local variations in intensity and tempo.
Both types of variations map in various ways onto cross-modal and affective
dimensions. As discussed above, intensity is closely associated with the distance
74 Music and Shape
DYNAMIC SHAPING
Figure 3.2 shows the intensity values per two-bar phrase and their variation
across the three performances. Comparison of the two panels shows a strong
overlap between the profiles in mean and maximum intensity (left and right
panels, respectively), except that the maximum intensity is on average 5 dB
higher than the average intensity (see differences in scale). The intensity profiles
are also very highly correlated for the three singers and clearly separate the
third stanza from the first two, following the forte indication of the score and
the crescendo towards fortissimo in the final phrase, as discussed above.
Focusing on changes in maximum intensity per two-bar phrase within each
stanza, we find that intensity seems to correlate in particular with pitch con-
tour (compare Figures 3.1 and 3.2). Using partial correlations, we can corre-
late measured intensity with the weighted pitch per phrase, after correction for
correlations with a forte indication in the score. This means that the jump in
intensity is accounted for by the forte indication, and the remaining variation is
correlated with pitch height. Table 3.5 shows the resulting partial correlations,
which are strong for IB in particular, followed by TQ and DFD with lower but
still significant partial correlations.
Intensity and pitch height are both associated with visual brightness, a
prominent feature of the text. Intensity reinforces associations related to pitch
of the vocal line. In the first stanza, intensity and pitch rise, which may relate
to the appearance (erscheint) of the city at dusk. In the second stanza, intensity
and pitch descend and dissolve into the rowing motion of the accompaniment,
76 Music and Shape
DFD IB TQ
which itself comes to a standstill in the fermata before third stanza. The soft-
ness of the low pitch disambiguates the low voice as being depleted of energy
rather than ‘full’ or ‘big’, which low voices can also be (Eitan and Timmers
2010). Additionally, it emphasizes the emotional distance of the protagonist.
While the second stanza involves a physical close-up, the protagonist is psycho-
logically distanced and isolated. In the third stanza, intensity is (as discussed)
a main parameter in the change in psychological distance: from remote and
passive to emotionally involved, and from a darkened, veiled mood to pain-
ful clarity. The increase in intensity and pitch within the stanza sustains this
process until the final tones of the singer, in which the source of the emotional
experience is revealed.
Within this general trend of matching intensity to pitch contour, the singers
deviate to varying degrees from a perfect correlation, which may be a way to
communicate the ambiguity of the pitch ‘rises’. DFD deviates most strongly
from the correlation with pitch height. In the first stanza, this is the case for the
last phrase, where he decreases in intensity rather than increasing, which can be
seen as a depiction of the dusk (Abenddämmrung) and the limbering darkness.
In the third stanza, intensity is high from the start. It builds up to a degree but
diminishes within the final phrase, where the loss of the beloved is acknow
ledged, and the return to the dark and subdued mood that follows is anticipated.
Both of these deviations from a matching of intensity with pitch contour seem
to qualify the pitch rises as dark in mood and depressed in emotion. The peaks
in pitch are moderated, emphasizing a distance and darkness of mood.
TEMPORAL SHAPING
As explained before, the duration of sung two-bar phrases was calculated from
the measured phrase onsets and offsets. All two-bar phrases in the first two
stanzas consisted of six crotchets. However, in the final stanza, phrases vary
in score duration: alternating phrases are slightly shorter or longer than six
beats, the longer phrases containing relatively long notes (minims for Boden
and Liebste). To make the phrase durations comparable, the measured dura-
tions were normalized by dividing performed duration by score duration and
multiplying this by 6, to show the duration of all phrases as if they consisted of
Cross-modal correspondences in a Schubert song 77
10
9
Normalized phrase duration
DFD
IB
8
TQ
5
0 2 4 6 8 10 12
Two-bar phrase
FIGURE 3.5 Normalized phrase duration of successive two-bar phrases in the performance by DFD, IB
and TQ. Interruptions in lines indicate bars that are separated by piano accompaniment intermezzi.
Singer DFD IB TQ
and close to significant for IB (Table 3.6). This highlights that tempo is used to
support the increase in proximity, activation and emotional arousal in the third
stanza (particularly in comparison to the first) and possibly (through crosstalk)
also the increase in brightness.
If we look at the local variations in tempo within stanzas, it seems that
tempo is used at times to intensify the effects of other elements of the music
and at times to moderate them. In the first stanza, for example, all three perfor-
mances gradually slow down towards the end of the stanza (Abenddämmrung).
In this stanza, the upward motion in pitch is accompanied by a slowing down
rather than a speeding up of tempo, perhaps associating with the stillness of
the evening and the static quality of the scene.
In the second stanza, tempo again decreases gradually for DFD and TQ,
although less strongly than in the first stanza. This time the slowing down
accompanies a decrescendo and a fall in pitch. The decrease in tempo intensi-
fies the associations related to the descent in pitch and intensity of the vocal
line, increasing a sense of the isolation and depression of the protagonist. In
contrast, IB does not slow down in the second stanza but keeps a steady tempo.
This choice may instead emphasize the steady motion of the oars, highlighting
external scenery rather than psychological process.
In the third stanza, the tempo is faster from the start of the stanza for IB,
while DFD and TQ speed up after a slower first phrase. This growth in motion
coincides with an increase in intensity and underlines the emotional intensity
and turmoil of the stanza. The changes in speed, increasing rhythmic irregu-
larity, further contribute to the emotional intensity of this stanza. IB extends
the phrases containing the longer notes (minims for Boden and Liebste), while
DFD and TQ lengthen in particular the final phrase and thus dramatically
accentuate the memory of the beloved (Liebste verlor).
Of the three performances, IB’s can be seen as providing the most literal read-
ing of the score, being relatively steady in tempo at a local level, and showing a
very strong correlation between intensity and pitch. He emphasizes the global
changes in the poem and the music: there is stronger motion in the second
Cross-modal correspondences in a Schubert song 79
stanza than in the first, and intense emotional arousal in the third. The dra-
matic climaxes in the third stanza are underlined through intensity and rubato,
extending the moments of intense emotion. His performance is seemingly most
bright and active, employing a relatively fast tempo and rising intensity with
rising pitch.
DFD shows the strongest modification of affective and cross-modal map-
pings. His performance matches the global intensification of the poem and the
music with a global rise in the tempo across stanzas and contrasting dynamics,
in particular in the third stanza. However, each overall increase in tempo is
counterbalanced by a considerable decrease in local tempo. His performance
of the first two stanzas can be heard as the darkest, most mournful of the
three: any surges in brightness through rising pitch are darkened by using softer
dynamics and slower tempos. The emotional instability and intensity of the
third stanza is emphasized through sudden and strong tempo changes, while
the temporary nature of the vivid memory of the beloved is highlighted by an
early return to the subdued character of the start.
In TQ’s performance, the difference in character between the first two stan-
zas is less apparent than in the other performances. The second stanza is not
faster than the first, although there is a contrast in speed between the end of
the first and the start of the second. The two stanzas are also very similar in
overall intensity. At a local level, TQ shows similar uses of tempo and dynamics
to DFD and (in some cases) IB. TQ tapers off the increase in pitch in the first
stanza by limiting the growth in intensity. He also limits the decrease in intensity
in the pitch fall of the second stanza, and slows down only slightly in this stanza,
providing a relatively constant affective character. In contrast, emotional inten-
sity is very strong in his performance of the third stanza, with a sudden rise in
dynamics, spectral brightness (indicated by spectral centroid) and fluctuation in
tempo. The emotional climax is sustained until the final notes of the vocal line.
METHODOLOGICAL CONSIDERATIONS
CONCLUSION
This analysis of ‘Die Stadt’ suggested ways through which cross-modal map-
pings contribute to the emotional and interpretative meaning of the song. We
emphasized how basic features of the music, including pitch, intensity and
tempo, are carefully employed to provide a multisensory experience of Heine’s
text. The textual context brought forward metaphors related to light, distance
and motion. Our analysis highlighted musical parallels to these metaphors and
affective connotations that come into play through particular treatment in the
composition and its performances, suggesting that the deceptively simple cross-
modal correspondences examined by experimental psychology may combine
to generate a highly complex, multivalenced web of musico-poetic meanings.
Finally we argued how the process of managing such a complex array of pos-
sibilities can be handled via the notion of musical shape.
Our analysis suggests that cross-modal mappings should be more centrally
included in models of the perception of emotion in music. We see such map-
pings as closely connected to processes captured under the mechanism of ‘emo-
tion contagion’ (Egermann and McAdams 2013) and attributed to relationships
Cross-modal correspondences in a Schubert song 83
References
Adler, M., 2014: ‘Cross-modal interactions and musical representation’, in Hebrew (PhD
dissertation, Tel Aviv University).
Blauert, J., 1997: Spatial Hearing: The Psychophysics of Human Sound Localization (Cambridge,
MA: MIT Press).
Boltz, M. G., 2004: ‘The cognitive processing of film and musical soundtracks’, Memory
and Cognition 32: 1194–205.
Boltz, M., 2013: ‘Music videos and visual influences on music perception and appreciation:
should you want your MTV?’, in S.-L. Tan, A. Cohen, S. Lipscomb and R. Kendall, eds.,
The Psychology of Music in Multimedia (Oxford: Oxford University Press), pp. 217–34.
Bresin, R., 2005: ‘What is the color of that music performance?’, in Proceedings of the
International Computer Music Conference 2005 (Barcelona: ICMC), pp. 367–70.
Casasanto, D. and K. Dijkstra, 2010: ‘Motor action and emotional memory’, Cognition
115: 179–85.
Clark, S., 2002: ‘Schubert, theory and analysis’, Music Analysis 21: 209–43.
Cohen, A. J., 2001: ‘Music as a source of emotion in film’, in P. N. Juslin and J. A. Sloboda, eds.,
Music and Emotion: Theory and Research (Oxford: Oxford University Press), pp. 249–72.
Collier, W. G. and T. L. Hubbard, 2001: ‘Judgments of happiness, brightness, speed and tempo
change of auditory stimuli varying in pitch and tempo’, Psychomusicology 17: 36–55.
Collier, W. G and T. L. Hubbard, 2004: ‘Musical scales and brightness evaluations: effects
of pitch, direction and scale mode’, Musicae Scientiae 8: 151–73.
Coutinho, E. and A. Cangelosi, 2009: ‘The use of spatio-temporal connectionist models in
psychological studies of musical emotions’, Music Perception 27: 1–15.
Coutinho, E. and A. Cangelosi, 2011: ‘Musical emotions: predicting second-by-second
subjective feelings of emotion from low-level psychoacoustic features and physiological
measurements’, Emotion 11: 921–37.
Dahl, S., M. Grossbach and E. Altenmüller, 2011: ‘Effect of dynamic level in drumming:
measurement of striking velocity, force, and sound level’, in Proceedings of Forum
Acusticum, June 27–July 1, 2011 (Aalborg, Denmark: Danish Acoustical Society), CD-
ROM, pp. 621–24.
Dolscheid, S., S. Shayan, A. Majid and D. Casasanto, 2013: ‘The thickness of musical
pitch: psychophysical evidence for linguistic relativity’, Psychological Science 24: 613–21.
Egermann, H. and S. McAdams, 2013: ‘Empathy and emotional contagion as a link
between recognized and felt emotions in music listening’, Music Perception 31: 139–56.
Eitan, Z., 2013: ‘How pitch and loudness shape musical space and motion’, in S.-L. Tan,
A. Cohen, S. Lipscomb and R. Kendall, eds., The Psychology of Music in Multimedia
(Oxford: Oxford University Press), pp. 165–91.
Eitan, Z. and R. Y. Granot, 2006: ‘How music moves: musical parameters and images of
motion’, Music Perception 23: 221–47.
84 Music and Shape
Eitan, Z. and R. Timmers, 2010: ‘Beethoven’s last piano sonata and those who follow croc-
odiles: cross-domain mappings of auditory pitch in a musical context’, Cognition 114:
405–22.
Fabian, D., R. Timmers and E. Schubert, eds., 2014: Expressiveness in Music Performance:
Empirical Approaches Across Styles and Cultures (Oxford: Oxford University Press).
Freddi, S., J. Cretenet and V. Dru, 2013: ‘Vertical metaphor with motion and judgment: a
valenced congruency effect with fluency’, Psychological Research 78/5: 736–48. Available
at doi: 10.1007/s00426-013-0516-6 (accessed 9 April 2017).
Ghazanfar, A. A. and J. X. Maier, 2009: ‘Monkeys hear rising frequency sounds as loom-
ing’, Behavioral Neuroscience 123: 822‒7.
Ghazanfar, A. A., J. G. Neuhoff and N. K. Logothetis, 2009: ‘Auditory looming perception
in rhesus monkeys’, Proceedings of the National Academy of Sciences 99: 15755–7.
Gingras, B., M. M. Marin and W. T. Fitch, 2014: ‘Beyond intensity: spectral features effec-
tively predict music-induced subjective arousal’, The Quarterly Journal of Experimental
Psychology 67: 1428–46.
Griscom, W. S. and S. E. Palmer, 2012: ‘The color of musical sounds: color associates of
harmony and timbre in non-synesthetes’, Journal of Vision 12: abstract 74.
Hascher, X., 2008: ‘ “In dunklen Träumen”: Schubert’s Heine-Lieder through the psycho-
analytical prism’, Nineteenth-Century Music Review 5: 43–70.
Hevner, K., 1937: ‘The affective value of pitch and tempo in music’, The American Journal
of Psychology 49: 621–30.
Hubbard, T. L., 1996: ‘Synaesthesia-like mappings of lightness, pitch, and melodic inter-
val’, The American Journal of Psychology 109: 219–38.
Kerman, J., 1962: ‘A romantic detail in Schubert’s Schwanengesang’, The Musical Quarterly
48: 36–49.
Kramer, L., 2003: Franz Schubert: Sexuality, Subjectivity, Song (Cambridge: Cambridge
University Press).
Kramer, L., 2004: ‘Odradek analysis: reflections on musical ontology’, Music Analysis
23: 287–309.
Kramer, R., 1994: Distant Cycles: Schubert and the Conceiving of Song (Chicago: University
of Chicago Press).
Lake, J. I., K. S. LaBar and W. H. Meck, 2014: ‘Hear it playing low and slow: how pitch
level differentially influences time perception’, Acta Psychologica 149: 169–77.
Lewkowicz, D. J. and N. J. Minar, 2014: ‘Infants are not sensitive to synesthetic cross-
modality correspondences: a comment on Walker et al. (2010)’, Psychological Science
25: 832–4.
Lewkowicz, D. J. and G. Turkewitz, 1980: ‘Cross- modal equivalence in early
infancy: auditory-visual intensity matching’, Development Psychology 6: 597–607.
Litterick, L., 1996: ‘Recycling Schubert: on reading Richard Kramer’s Distant Cycles:
Schubert and the Conceiving of Song’, Nineteenth-Century Music 20: 77–95.
Ludwig, V. U., I. Adachi and T. Matsuzawa, 2011: ‘Visuoauditory mappings between high
luminance and high pitch are shared by chimpanzees (Pan troglodytes) and humans’,
Proceedings of the National Academy of Sciences 108: 20661‒5.
Marks, L. E., 1989: ‘On cross-modal similarity: the perceptual structure of pitch, loud-
ness, and brightness’, Journal of Experimental Psychology: Human Perception and
Performance 15: 583–602.
Cross-modal correspondences in a Schubert song 85
Shapes composed
Reflection
George Benjamin, composer and conductor
A few works in the repertoire have a formal contour so simple that it can be
recalled in toto after a single hearing. Some, like Borodin’s Steppes of Central
Asia, Debussy’s Sunken Cathedral or the first of Berg’s Three Orchestral Pieces,
approach from the distance, reach an apogee and then recede. Others merely
build inexorably from virtual silence to a cataclysm—Grieg’s In the Hall of the
Mountain King, the passacaglia interlude from Ligeti’s Le grand macabre or,
most famously, Ravel’s Bolero.
In most of these, all surface resources of music—register, instrumental
density, velocity and above all volume—are exploited to create the most rudi-
mentary formal outline: an arch or a wedge. Some—Debussy and particularly
Berg—are marked by a much larger degree of internal diversity and intricacy,
though the fundamental structural mould still holds.
Other basic shapes rely less on incremental sonic display and instead employ
different, and more versatile, essential tools in structural definition, namely sym-
metry and repetition. Using these typically involves subtler resources, involving
above all thematic material and—at least until the early twentieth century—
key. Da capo arias and classical minuets are obviously moulded along these
lines, as are, to a lesser or greater extent, variation and rondo forms. Add the
arts of expanded contrast, transition and evolutionary development, and the
same forces also underpin sonata form. When this architectonic blueprint was
combined with the narrative thrust of the novel—or of opera—more dynamic
and unpredictable forms were the result, from Beethoven to Berlioz, Wagner,
Mahler, Debussy, Berg, Carter and beyond.
A complex musical work has many diverse—and simultaneous—shapes.
On the largest imaginable scale, the placing of grand orchestral perorations in
Wagner’s Götterdämmerung has a specific and precisely judged contour, as does
89
90 Music and Shape
the placing of silences, across the three acts. Similarly the large-scale rhythm of
thematic recall, the alternations between varying types of texture, the shadings
and pacing of the highly diverse harmonic palette, the labyrinthine tonal design
and the flirtation with cadence, all of these were supremely interlaced by this
master of dramatic architecture and proportion. Even the contrasts in tempo,
metre and phrase structure; the use of restricted registers and specific timbres
(piccolo, stopped horns, multiple harps, tam-tam and suspended cymbal); the
varying types of word-setting—all of these have a macro form, many of them
intersecting from time to time, some at the very surface of the music, others
more deeply buried within its construction.
The perception of large-scale form requires much guess-work from the
attentive ear during performance; particularly in modern music, the full shape
can be comprehended only as the very last note falls into silence. At first it may
be almost impossible to discern the type of formal play involved in a work,
or its manner of unfolding and its scale. This is one of the challenges—and
delights—for the listener, and the ear searches for exceptions as much as sym-
metries in order to orientate itself along the path of an unfamiliar work.
The arrival of the chorus in the fifth movement of Mahler’s Second, the
first notes of the harp at the very end of the first act of Tristan—these pianis-
simo entries, an hour into the structure, are decisive structural incidents, just as
potent and memorable as the most energetic prestissimo or extreme tutti climax.
The trombones in the Pastoral Symphony, the large bells in the Symphonie fan-
tastique or the gongs in Le marteau sans maître—these timbral signals, mark-
ing the later stages of each work, all share a similar function. In an opera of
such syntactical volatility and compression as Verdi’s Falstaff, the arrival of
symmetrical phrase patterns and continuous set forms in the third act has a
decisive influence on the recalled form, as does the lapping lullaby motion and
unambiguous tonal security in the recognition scene in Elektra.
Of all the factors involved in creating structures on this scale, perhaps the
least easy to grasp and to recall is the harmonic thread; but such works will
live or die according to the success in handling this most intangible of musical
phenomena. This is all the more challenging for a composer working outside
the predefined pathways and goalposts that the tonal system provides.
No other composer has exceeded Berg as a master of large-scale design, and
the plotting of structure across Wozzeck—on and below the surface—makes
most other music seem like child’s play. In particular, the third act has an irre-
sistible momentum and dramatic impulse, yet it also seems underpinned by
the deepest architectural foundations. The first two acts exploit a sequence of
older musical forms as scaffolding to support the frequently jagged, expression-
istic surface of the music—ranging from sonata form to passacaglia, scherzo
to fugue. However, beyond the opening scene—a highly idiosyncratic set of
variations—Act 3 is virtually free of conventional formal background, each
successive scene inventing a sui generis prototype of astounding originality.
Reflection: George Benjamin 91
This same oscillation over a perfect fifth is used at the end of the first act,
though in a darker and fuller register (Figure R.9).
The second act also concludes with the same harmonies, presented however
in a fragmented and gruff way, so deep in register that they are barely percep-
tible (Figure R.10).
Nevertheless, the conclusion of all three acts of Wozzeck ‘rhyme’
harmonically—as is the case, incidentally, in Berg’s other operatic masterpiece,
Lulu—and the complete third act is straddled by a daisy-chain of harmonic
connections below the surface, each recapitulating harmonic sonorities as a
means of closure, and each giving Berg a firm telos at which to aim his extraor-
dinarily diverse and subtle invention (Figure R.11).
Reflection: George Benjamin 95
A great question facing a modern composer at the start of a work is, Where
is the harmony going? Many allow the pitch material to expand without appar-
ent destination or predetermined direction, forging the form by fantasy and
focused improvisation. Others create a goal—going full circle by returning to
an opening harmony or by inventing a pseudo-tonic sonority at which to aim.
Equally it’s possible to envisage a complete path in advance—particularly if
the harmonic vocabulary isn’t too complex—which might be modified radically
as the composition evolves. Some harmonic plans preclude change, with the
resultant piece remaining locked into the identical, static blueprint from begin-
ning to end. Yet others allow the music to pursue a self-generating—or even
arbitrary—harmonic mechanism below the surface, abruptly cutting the music
off when there is no more to say.
While conceiving a new work, a composer today may well envisage a circle,
a series of blocks, a sequence of loops, a perpetually descending spiral—or a
combination of some or all of these—as a metaphor for its harmonic trajec-
tory. These simple, imagined shapes are integral to the compositional process,
not mere pretence or poetic analogy. Such a path may be planned in advance,
though usually, I suspect, it evolves during the act of composition. Regardless
of its provenance, a large-scale work—whatever the era or idiom—needs a firm
and decisive sense of closure at its end. The shape Berg proposes for the conclu-
sion of Wozzeck maintains more than a degree of relevance today.
4
Emotions have shapes, and musical emotions mirror those shapes. This is a sim-
ple enough claim. But the multitude of assumptions packed into this statement
could fill a library. Indeed, it drives the industry of emotion studies, which has
overtaken music psychology and aesthetics (but not yet musicology) since the
humanities’ affective turn a decade or so ago.1 Do emotions have shapes, or is it
the behaviours, intonations and intentions associated with them? Why and how
is music emotional, and is the emotion expressed, induced or perceived? Does
emotion even exist? What is a ‘shape’, and how can a musical shape be captured
analytically? Is it the preserve of the composer or the performer? And so on.
In walking through this jungle, I put together six arguments, all prefabri-
cated: the theory lies in the assemblage rather than in the constituent ideas.
1. I speak not of ‘emotion’ in the round but of ‘emotions’ in the plural,
many comprising discrete basic categories such as sadness, happiness,
fear, tenderness and anger.
2. These categories are ethological, originating in adaptive animal
behaviour.
3. Emotional behaviour is expressed through goals, or ‘action tendencies’.
4. In music, emotional categories are associated with acoustic features
which are readily identifiable.
5. Goal-directed emotional behaviour in music is conceivable when we
think of music as a virtual person, according to ‘persona theory’.
6. I discriminate between emotional expression and induction, on the
basis that a listener discerns a musical process as being ‘expressive
of emotion’ (rather than transitively expressing a composer or
96 performer’s affective intentionality).
Shapes of affect in Bach’s Sonata in G minor 97
Emotions proper are held to have emerged not in the very oldest and innermost
layer of the brain—the ‘corpus striatum’—which controls basic animal rou-
tines such as walking, patrolling, foraging and mating, but in the central lim-
bic system. Often called the ‘emotion brain’, the limbic system arose with the
peculiarly social world of mammals unavailable to reptiles. It involves sociable
behaviours such as mother/infant care-giving, vocal signalling and play, and
is the site of the basic emotions: happiness, anger, fear, desire and sadness.
A crucial feature highlighted by the ‘city metaphor’ is that the more intellec-
tual neocortex—the third and newest layer, which developed over the six mil-
lion years in our evolution from the apes—does not supersede these two older
layers. On the contrary, ‘as with the organization of cities, earlier forms and
developments have continued, and provided for subsequent developments and
elaborations’ (ibid.: 67). In particular, the neocortex elaborates the sociality of
the limbic emotions. This notion of earlier layers persisting through later ones
98 Music and Shape
Part I: The Adagio
CUMMING’S ANALYSIS
FIGURE 4.1 Bach, Sonata for Unaccompanied Violin No. 1 in G minor (BWV 1001), Adagio, bars 1–13.
Shapes of affect in Bach’s Sonata in G minor 101
FIGURE 4.1 Continued
way that ‘the synthesis of its structural elements, when they are heard as embody-
ing aspects of [human] movement (in directionality, force, etc.), … suggest expres-
sive agency’ (149). In short, a gesture is a ‘melodic shaping’ (230), the embodiment
of a sonic quality as a particular musical event. Cumming’s notion of shaping is
enriched by the ontological differences of sign types; I part company with her, how-
ever, when she tries to extend the Peircean method to theories of emotion.
102 Music and Shape
SADNESS IN THE ADAGIO
expression of grief is evolved to elicit emotional support from others, the musi-
cal persona mimics the dejected face, drooping posture and plaints of a sad
person. These are the familiar ‘acoustic features’ of sadness codified by many
psychologists (Juslin 1997; Huron 2008; Gabrielsson and Lindström 2010): slow
tempo, minor-mode key, narrow intervals, legato articulation, variability of tex-
ture, preponderance of descending melodic contours and a high level of dis-
sonance, especially involving the semitone appoggiaturas of the pianto topic
(Monelle 2000). Descending lines suggest loss of physical and mental energy;
narrow intervals and legato articulation imitate low-energy mumbling.
A fresher perspective is afforded by Huron’s connection between sadness
in music and the ‘detail- oriented thinking’ of ‘depressive realism’ (Huron
2011: 48), following the work of Alloy and Abramson (1979), who consider the
impact of emotion on cognition and perception. From this angle, reflection and
self-reflection are seen as behavioural aspects of sadness, an emotion which is
an adaptive opportunity for a wounded organism to recover by taking stock of
the situation. How would this be illustrated analytically, given that psycholo-
gists of emotion have shied away from looking at the ‘structural features’ of
emotion beyond the parametric level? I suggest that ‘detail-oriented thinking’ is
borne out by thematic atomism and formal fragmentation, the way the Adagio
lurches rhapsodically from one contrast to another. Its lurching vicissitudes are
indeed another side of sadness’s lack of goal, just as its aimlessness is worked
out by spurts of spontaneous melismas and maggiore episodes. (For more on
such signalling see Spitzer 2009.) Huron identifies such major-key interludes
in minor-key works with ‘nostalgia’, which he thinks is a flavour of sadness.
We find such episodes at bars 2–3 (a lurch from G minor to B♭ major and
back again) and, more dramatically, at bars 11–13 (shifting from C minor to E♭
major and back to C). The pathos of these nostalgic moments is heightened by
their very interruption, or ‘containment’, to invoke Cumming’s term—part of
sadness’s relentless denial of goal-orientation.
Sharpening the focus on the opening phrase, ‘atomism’ is evinced in the
sheer density of the Adagio’s texture, a ‘thickness’ which demands ‘detail-
oriented’ reflection from the listener. Hence we see the reciprocal relationship
between sadness as a disposition of musical material (dense and fragmentary),
and sadness as a mode of hearing (acute and detail-oriented). Otherwise put,
we don’t just hear sadness, we also hear in a sad way. Density is heard in the
ways the opening phrase both invokes and resists formal and contrapuntal
schemata. Gjerdingen hears it as instantiating a 1–7 … 4–3 ‘Meyer schema’,2
even though this breaks his own rule (2007: 112) that 4–3s shouldn’t overlap
1–7s. At the very least, the schema is deformed. Better to hear it, I suggest, as
a mutual interference of two schemata: a 1–7 … 7–1 (complementary pianti
weeping gestures, with the second F♯ displaced up an octave), and a 5–4 …
4–3, a descending line which will emerge in the fugue subject, but introduced
here with the opening D elided. The tritone leap from C to F♯ leaves the C high
and dry, seeming to foreshadow the subdominant bias of the movement (as in
104 Music and Shape
the C minor ritornello of bar 13). There is a similar tension between the pro-
pensity of analysts to read the Adagio ‘top-down’ as a Schenkerian 8-descent
(Cumming 2000: 233), or ‘bottom-up’ as a descending, rule-of-the-octave bass
pattern (Lester 1999: 34). This either-or binary detracts from the messy, poly-
phonic richness of the Adagio’s texture, for instance the quasi-canonic counter-
point in bars 2–3, where the melody’s E♭–D step is mirrored a little later in the
‘bass’. This quasi-canon—a mensurally distorted canon at the octave (recall-
ing the G major/minor canons in the Goldberg Variations)—is missed by both
Lester and Cumming, but is suggestive of the Adagio’s very self-reflection.
Thus Cumming’s four affective epithets— pathos, reflection, spontaneity
and containment—really hang together as a package of entailments of a single
emotional category, sadness, considered as a type of adaptive behaviour. Pathos
and reflection are, respectively, outward-and inward-facing behaviours: gestur-
ing to observers, reflecting on loss. Spontaneity and containment are comple-
mentary symptoms of goal loss: energy breaks out, breaks down or is blocked.
FIGURE 4.2 Vivaldi, Violin Concerto Op. 3 No. 6, Largo, bars 1–6
B♭ is formally dislocated from the next note of the cycle because it is projected as
an ending of the first phrase (Vordersatz): the note is relatively long (a quaver),
resolves the preceding tonal tension (dominant-seventh harmony) with a tonic
and is articulated by the following demisemiquaver rest. If the B♭ is an ending,
then the lurch up to the E♭ sounds like a new beginning, metrically stronger than
the first beat of the next bar. The A natural, the third beat of the cycle, is metri-
cally weakened by being displaced by a quaver in bar 3; and it is disconnected
from the E♭ because that note had fallen back down to a B♭. D, the fourth note of
the cycle, is reached only two quavers after A: each of the four steps of the cycle is
differentiated by a distinct textural shape and metrical placement. This makes the
‘skeleton’ of the phrase, its grammatical deep structure, quite challenging to hear.
Moreover, this sense of discontinuity is compounded by the abrupt tonal shifts
across bars 2–3 from G minor to B♭ major and back to G minor.
Hence pinpointing bars 2–3 shows how the Adagio’s fifth cycle is highly
deformed, effectively into a series of isolated structural notes, suggesting a
‘detail-oriented’ listening in line with the ‘depressive realism’ of sadness.
106 Music and Shape
There are two diametrically opposite accounts of ‘shape’ that can be drawn
from this example. In the first, the Adagio’s sadness, with its detail-oriented
depressive realism, is ‘adverbial’, being a transformation of a formal model. All
four movements of Bach’s sonata begin with the same Vordersatz–Fortspinnung–
Epilog ritornello model (see Figure 4.3). Focusing on the central cycle-of-fifths
module spotlights the successive transformations, each one of which produces
a different emotion—more on this in Part II. The key point here is that eliciting
contrasting expressive character by transforming a framework is the essence of
variation form. See also, in the history of theory, Heinichen’s (1711) or Niedt’s
([1706] 1721) lessons to budding composers on how to adapt expressive figura-
tions to libretti in order to project differing emotions. This adverbial account
highlights ‘shaping’ rather than ‘shape’.
The opposite account discovers ‘shape’ in the pattern rather than in its
inflection. By ‘pattern’, I mean the shape of the music’s ‘behaviour’. Elsewhere,
I termed such dynamic emotional shapes ‘affective trajectories’ (Spitzer 2013).
Although Oatley and others characterize sad behaviour with loss of goal, the
emotion is not without directionality. Sadness is a strongly aversive emotion;
for an emotion whose essence is loss of goal, the only goal for sadness, para-
doxically, is to stop being sad. This affordance is strangely ignored by psy-
chologists of musical expectancy. Margulis (2005), for instance, theorizes (after
Huron) three classes of expectation: surprise, denial and expectation proper.
But the sadness of the Adagio’s opening phrase, as an aversive emotion, is
surely implicative of an escape from this sadness. The lurch into the tender/
happy B♭ major episode in bars 2–3 is admittedly a ‘surprise’ in its abruptness.
Yet isn’t this flight to the major, to a positive valence, implied by the aversive
quality of the opening? (Conversely, in the many works where such flight to the
major is denied, isn’t this minor-mode standstill registered as a form of ‘con-
tainment’ or even repression?)
It is important to pin this trajectory to fundamentals in order to u ndergird
more complex, Lacan-tinged, explanations (Spitzer 2013). For instance, if
sad music takes separation anxiety as an axiom, then its trajectory seeks to
recreate (recuperate, memorialize, return to) the severed social bonds, typically
in the form of a maggiore ‘dream image’. The dream image at the centre of
the ritornello’s central module, bars 2–3, for all its brevity, is more animated
(the scalar uprush to E♭) and intervallically more expansive (fourths, tritones,
sixths), and it momentarily even trips into a dance lilt. And then this episode
is just as s uddenly snuffed out by the F♯, returning the music to G minor. The
‘shape’, then, is an implicative drive away from materials associated with sad-
ness, towards those expressive of tenderness and happiness, accompanied by a
sudden ‘opening out’ or expansiveness, suggestive of feelings discharged from
within, or liberated from a constraint; and then a sudden return to the ini-
tial state. Importantly, the middle tender/happy state is not separable from this
process, but part of sadness’s trajectory. It is helpful that Huron characterizes
tender/happy music contextualized within sad music as ‘nostalgia’ (although
maggiore episodes are surely not all backwards-facing: see the discourse gen-
erated by Levinson’s identification of the second group of Mendelssohn’s
Hebrides Overture with ‘hope’; Levinson 1990, Karl and Robinson 1995).
The value of such a broad conception of ‘shape’ is that it doesn’t commit
us either to a formal model (such as Fischer’s and Dreyfus’s ritornello model)
108 Music and Shape
PERFORMANCE SHAPES
Before we turn to the rest of Bach’s sonata to explore ‘shape’ and ‘shaping’ at
the level of the cycle, we need to complete—even ‘consummate’—this dialectic
in the reality of musical performance. Isn’t ‘shaping’ what a violinist does with
Bach’s materials? On the other hand, can one speak of performance ‘shapes’
across an interpretation? A common experimental protocol in emotion psy-
chology research is to get a performer to interpret the same phrase in different
ways so as to project varying affective states. Is it thereby legitimate to view
‘adverbial’ compositional processes, such as variations, as ‘performative’ in this
respect, the composer shaping a musical model into a distinct affect just as a
performer shapes the music? If so, then a notion of emotional shape/shaping
may shed new light on the interaction of scores and performances.
In a market saturated with recordings of Bach’s music for unaccompanied
violin, I have selected distinguished versions by Itzhak Perlman, Sergiu Luca
and Gidon Kremer. Although Cumming doesn’t engage with specific perfor-
mances of the Adagio, her Peircean triad voice–gesture–will suggests generic
differences between these three violinists’ approaches. Perlman’s classic 1988
recording epitomizes mainstream late twentieth-century interpretative practice,
playing the piece with large-scale, often symmetrical phrasing. Perlman brings
out the broad formal unfolding of the Adagio, the ‘will’ of the tones. Luca’s
1992 ‘historically informed’ (HIP) recording is focused much more sharply
on the intricate gestures of the Adagio’s rhetorical delivery. It is tempting to
style HIP ‘gestural’, after Cumming, although its rhetorical quality reminds
us that it is difficult to conceive of musical gesture apart from vocality. That
said, the portamento ‘sobs’ prevalent in early twentieth-century practice, as in
Fritz Kreisler’s 1926 recording, may sound even more vocal than HIP. My third
example, Kremer’s 1981 version, is interesting for combining modern tech-
niques with intricate phrasing, yet the latter expressing not HIP sensibilities
so much as rhapsodic waywardness. Taking the Kremer last, I begin with a
point-by-point comparison of the Perlman and Luca versions, concentrating
on the ‘emotional shape’ of the opening ritornello and its ‘architectural’ expan-
sion across bars 11–13, in the light of tempo and dynamic maps of the perfor-
mances (Figures 4.4 and 4.5).3
Luca’s rendering of Bach’s opening projects the wave- like spectral and
dynamic shapes highly characteristic of the period bow (Fabian 2005: 95). The
short baroque bow is conducive to the ‘ “period” stroke’: soft onset and rapid
decay. A spectrogram easily reveals that the higher frequencies crest and fall
across Luca’s bow strokes on the strong beats of bars 1–2, and that the steep
oscillations are matched by the dynamic swells and ebbs. Conversely, spec-
trograms of Perlman’s performance, on a modern bow and instrument, show
his solidly sustained tone and dynamics. Luca’s spectral/dynamic wave shape
is also mirrored in the oscillations of the tempo maps, but not in synchrony
Luca
40 0
30 –20
Tempo (BPM)
Energy (dB)
20 –40
10 –60
0 –80
1.1 1.3 2.1 2.3 3.1 3.3 4.1 4.3 5.1 5.3 6.1 6.3 7.1 7.3 8.1 8.3 9.1 9.3 10.1 10.3 11.1 11.3 12.1 12.3 13.1
bar.beat
30 –20
Tempo (BPM)
Energy (dB)
20 –40
10 –60
0 –80
1.1 1.3 2.1 2.3 3.1 3.3 4.1 4.3 5.1 5.3 6.1 6.3 7.1 7.3 8.1 8.3 9.1 9.3 10.1 10.3 11.1 11.3 12.1 12.3 13.1
bar.beat
with the note swells, and differently between the two players. It is interesting
that both Luca and Perlman begin at similar tempos (21 bpm), and accelerate
to a peak at beat 3 of the first bar (Luca 25.2 bpm; Perlman 23.9 bpm), before
slowing down. Both players also decelerate towards the end of bar 2 (Luca
23.5 bpm; Perlman 16.7 bpm), against the grain of an older performance tradi-
tion (perhaps beginning with Joachim’s 1903 recording) of taking the ‘uprush’
scale at beats 3–4 somewhat faster. In both recordings, then, the ritornello’s
Vordersatz is shaped by a nearly identical tempo wave (Perlman: 21–23.9–16.7
bpm; Luca: 21–25.2–23.5 bpm), helping to project it as a self-contained unit, a
sort of sonic pillar.
Luca and Perlman drift further apart in how they treat the remainder of the
ritornello and the music immediately after it. A lot of my analysis pivots on
the boundary between the Vordersatz and Epilog of Bach’s ritornello, marked
by the B♭–F♯ gesture at bar 3 and the wrench it effects back from major and
minor. The three performances interpret this boundary in different ways.
Luca articulates the four semiquavers at the beginning of bar 3 very care-
fully, with a hint of dotted rhythm on the first of each pair, and a diminuendo
towards the quaver D (from -26 to -34 dB) thereby rendering the louder B♭–F♯
tonal interruption more dramatic (from -34 to -21 dB). On the one hand, this
cuts off the fifth-cycle Fortspinnung module from the Epilog. On the other, there
is surprisingly little deceleration into the long cadenza-like melisma at bar 3
(from 30 bpm at bar 3.2 to 26.6 bpm at bar 3.4), despite the performance tra-
dition in non-HIP recordings of taking the melisma considerably slower. Yet
both aspects bespeak the same tendency of HIP readings to focus on small-unit
articulation and play down broader contrasts. In this respect, Perlman’s read-
ing is markedly different, epitomizing the mainstream tradition’s preference for
seamless legato, uniformity of tone, long-range or block-like contrasts, and the
projection of large-scale structure.
Where Luca separates modules 2 and 3, Perlman’s powerful sense of line
drives through them in a fine art of transition. He maintains a high dynamic
level across the four semiquavers (-25 dB), all articulated evenly and with equal
intensity, swelling successively through the B♭–F♯ gesture to the G resolution
at bar 3.3 (-27 dB), climaxing with the melisma (-31 to -35 dB). The arrival
of this gesture, then, is smoothly mediated and subsumed into the swell into
the melisma: it is part of a wave, rather than a brusque shock. Perlman’s art
of transition is underscored by tempo changes: the B♭ major dream image at
the start of bar 3 is fastest (accelerating from 20.2 bpm at bar 3.1 to 24.1 bpm
at bar 3.2), and his Adagio subsequently decelerates through the B♭–F♯ ges-
ture to the luxuriously paced melisma (21 bpm at bar 3.3 to 15.1 at bar 3.4).
Compared to Luca, Perlman widens the tempo differential between dream
image and melisma: a slight difference of 4.8 bpm with Luca (from 31.4 to 26.6
bpm), nearly double that with Perlman (from 24.1 to 15.1 bpm). Otherwise put,
in Perlman’s recording, the duration of quaver beats from the D through the
Shapes of affect in Bach’s Sonata in G minor 113
30 –20
Tempo (BPM)
Energy (dB)
20 –40
10 –60
0 –80
1.1 1.3 2.1 2.3 3.1 3.3 4.1 4.3 5.1 5.3 6.1 6.3 7.1 7.3 8.1 8.3 9.1 9.3 10.1 10.3 11.1 11.3 12.1 12.3 13.1
bar.beat
Part II: The Cycle
Like sadness, the other basic emotions can be defined in terms of goals and
social relationship. Anger is typically triggered by the frustration of an ‘active
goal’, leading to aggressive behaviour such as fighting. Tenderness, or love, is
associated with ‘physical and mental closeness’ and with nurturing behaviour.
Fear is stimulated by an appraisal of ‘danger or goal conflict’, the subject react-
ing with withdrawal (e.g. fleeing) or freezing (e.g. trembling) behaviours (Oatley
2004: 81–2).
A cursory overview of the sonata’s remaining three movements does suggest
that their material unfolds these respective emotional behaviours. If the second
movement’s fugal opening expresses a kind of tetchy, repressed or even ‘cold’
anger, then the music ‘lashes out’ in the successive eruptions of semiquaver pas-
sagework. These ‘eruptions’ recall James Russell’s ‘script’ for anger (1991: 39),
a more elaborated version of Oatley’s schema: after an offence, a person glares
and scowls, will feel internal tension and agitation and a desire for retribu-
tion, and finally will lose control and strike out. The Siciliana is generically and
topically a lullaby, its tenderness mirroring the intimate and nurturing social
closeness of a dialogue between a mother and child (described by Colwyn
Trevarthen as the ‘primary intersubjectivity’ enacted in their rhythmic turns
of cross-modal dialogue; 1999–2000: 177). The literature on lullabies is exten-
sive, often referring to their cross-cultural features of simplicity, smoothness,
descending contours, relative slowness and short phrasing (Unyk et al. 1992;
Trainor and Hannon 2013). Daniel Leech-Wilkinson has linked the preponder-
ance of falling pitch contours in art-music lullabies to the descending motions
of Infant- Directed Speech (IDS) or ‘motherese’ (Leech- Wilkinson 2006).
118 Music and Shape
FIGURE 4.7 Bach, Sonata for Unaccompanied Violin No. 1 in G minor (BWV 1001), Fuga, bars 1–4
Shapes of affect in Bach’s Sonata in G minor 119
FIGURE 4.8 Bach, Sonata for Unaccompanied Violin No. 1 in G minor (BWV 1001), Siciliana,
bars 1–6
120 Music and Shape
FIGURE 4.9 Bach, Sonata for Unaccompanied Violin No. 1 in G minor (BWV 1001), Presto, bars 1–11
but harmonious, because the pitches of the cycle are shared between the two
voices, and indeed cross over.
The rhythmic and textural uniformity of the Presto—its continuous succes-
sion of semiquavers—makes it initially difficult to pick out the fifth cycle from the
background figuration (Figure 4.9). Interestingly, the cycle is slightly extended
by a further fifth progression: A–D is followed by G–C at bar 11. This is the only
movement where this happens. It is as if the forward-moving harmonic drive of
the music is so great that the cycle’s seemingly endless implicative potential to
rotate around the circle of fifths (B♭–E♭–A–D–G–C–F–B♭ etc.) can hardly be
contained. This harmonic drive compounds the Presto’s rhythmic speed. As well
as panic, the movement expresses another corollary of fear: shock. The Presto
unfolds a series of shocks by subverting its metrical pattern; indeed, this pat-
tern is constantly shifting in unpredictable ways, cognitively ‘wrong-footing’ the
listener (as it symbolically wrong-foots the fleeing subject, as it were). For
instance, the very start of the cycle, the B♭ at bar 6, subverts a pattern of two-bar
phrases established at the opening (Figure 4.10a and b). That is, the fast music
suggests a slower metrical grouping, whereby bars 1–2 constitute one ‘beat’ of a
‘hyper-bar’, bars 2–3 a second beat, and bar 5 the onset of a third beat. It is this
implicit three-beat hyper-bar that is interrupted by the B♭; it introduces a ‘hyper-
metrical’ disruption. Moreover, a ‘metrical reduction’ of the cycle at bars 6–8
(leaving out the semiquavers between its notes) reveals that, by accenting the sec-
ond beat of each group (the crotchets E♭, D and C), it encapsulates the preced-
ing hypermetrical disruption in miniature (Figure 4.10b). Hence not only does
the cycle arrive as a metrical shock to bars 1–5, but it is itself a series of metrical
shocks. And there is another, broader, level at which the Presto expresses fear:
the sheer speed of the music makes it difficult to follow. This literally overwhelm-
ing quality evokes the classic formula of the sublime, which is fear at its most
philosophically elevated level. The Presto evokes sublime fear both as cognitive
overload and as the behavioural reaction to fear, which is to flee.
Shapes of affect in Bach’s Sonata in G minor 121
FIGURE 4.10 (a) Hypermetrical reduction of Bach, Sonata for Unaccompanied Violin No. 1 in G minor
(BWV 1001), Presto, bars 1–6; (b) metrical reduction of bars 6–8, revealing syncopation
Bach’s cycle, then, projects four distinct emotional behaviours. The central
module of the ritornello stereotype serves as a bellwether for each behaviour.
Its atomization in the Adagio suggests the lack of goal— the lethargy—
connected with sadness or depression. Its conflictual disposition in the Fuga
enacts the aggressive conflict often linked to anger, when goals are blocked.
Its fluid and flexible disposition in the Siciliana suggests the tender dialogue
between mother and child in a lullaby, mirroring social closeness. And its
animation—and overflow into an extended fifth cycle—in the Presto evokes
a subject’s physical flight, in extreme fear or panic. All of these emotional
behaviours constitute ‘shapes’, as previously defined, shaping the ritornello
stereotype.
VECTORS
A third dimension—in addition to ‘shape’ and ‘shaping’—is the relational one,
through which the various behaviours define and articulate themselves against
each other. This dimension commutes shape/shaping into a kind of transforma-
tional ‘vector’, nudging it from the domain of emotion proper to that of ‘affect’.
Although the terms ‘emotion’ and ‘affect’ tend to be used interchangeably, in this
instance I follow thinkers such as Brian Massumi (2002), who represents a con-
stellation of ideas drawn chiefly from Deleuze, Bergson and Spinoza. Massumi
122 Music and Shape
Presto’s fear is passive: we are in the grip of a stereotype; the Fuga’s anger is
deliberative, deploying the stereotype with intent.
Viewing Bach’s sonata as a combinatoire of warring passions jibes with how
emotions functioned in Shakespeare and Homer, according to the emotion his-
torian Philip Fisher. In Fisher’s words, the passions, or ‘vehement states’, define
each other by fighting each other:
When used to define and express the substance of the self, the vehement
states—anger, wonder, ambition, jealousy, shame, pity, or fear—draw
on an essentially Greek and especially Homeric theory of substance and
struggle, or, as the Greeks called it, agon. Substances mutually make
each other known, not only because of their differences but because of
moments of conflict. It is at the meeting point where combat takes place
and mutual destruction is possible that each becomes for the first time
visible as what, in itself, it is. A large rock is one substance, the water
of the sea another. At the shoreline where the sea pounds against the
rock, the rock registers in its shape nothing but the consequences of thou-
sands of years of waves cutting into it, even as each individual wave was,
in turn, stopped and broken by the rock’s resistance… The shattering
wave, the pounded rock make visible on each side the nature of sea and
rock, but they do so at the very moment that each of the two is situation-
ally flooded from without by the differences that occur as each limits the
other. (2003: 51)
References
Unyk, A., S. Trehub, L. Trainor and G. Schellenberg, 1992: ‘Lullabies and simplicity: a
cross-cultural perspective’, Psychology of Music 20: 15–28.
Wegman, R., 2003: ‘Johannes Tinctoris and the “New Art” ’, Music and Letters
84/2: 171–88.
Discography
Joachim, J., [1903] 2003: The Great Violinists: Recordings from 1900–1913. (Testament
SBT2 1323).
Kremer, G., 1981: Johann Sebastian Bach: Sonatas and Partitas for Unaccompanied Violin
(Philips 6769 053; CD reissue: ECM New Series 1926–27).
Luca, S., 1992: Johann Sebastian Bach: Sonatas and Partitas for Unaccompanied Violin
(Nonesuch HC-73030; CD reissue: 73030).
Perlman, I., 1988: Johann Sebastian Bach: Sonatas and Partitas for Unaccompanied Violin
(EMI Classical CDS 7 49483 2; reissue: 0 85281 2).
Reflection
Steven Isserlis, cellist
and languor, and so on: one has also to be familiar with their fates, with the
interplay between them, with their transformation over the course of the work.
This knowledge informs every aspect of a performance—tone colours, tempo
relationships, dynamic contrasts, etc. Again, this may sound obvious, but all
too often musicians fail to come to terms with these basic elements. The result is
boring performances—and alas, there are far too many of those! I would liken
these haphazard musicians to travellers walking through a forest, lurching from
tree to tree, appreciating the beauty of each tree, perhaps, but with no idea how
to get to the other side of the forest. Conversely, a performer who understands
the structure of a work will be blessed with the freedom of a bird flying above
that forest, perceiving each detail in all its exquisite clarity, but able at all times
to make out the overall direction of the path. Foreknowledge of the form—the
story—must inform the interpretation from the outset. The actor analogy again
seems apt: at the end of a great performance of Hamlet, the audience should
somehow feel that, despite the many unpredictable twists and turns of the play,
there has been an inevitable trajectory to the hero’s fate.
Within this overall view of the work, one must also, of course, grasp the
microstructure of each phrase. In music, as in speech, every clause, every part
of the phrase, has a centre, with notes leading to and away from it. Just as in
speaking one highlights the most important or unexpected word in a sentence,
so an equivalent event in music will require some sort of emphasis. This can
be achieved with dynamic stress, of course; it can also be done with time—the
so-called agogic accent, in which one lingers on one note, making up the time
on less important notes (‘rubato in tempo’); with colour; or in countless other
ways. If the performance is to sound truly alive, no two consecutive notes should
have exactly the same weight. And then, again as in speech, there is the question
of punctuation: music is full of commas, full stops, semicolons, full colons and
so on. For string players, the bow must be ready to leave the string at all times;
for pianists, hands and feet must similarly be employed in order to allow the
phrase to breathe; and so on through all instruments and voices. If this very basic
aspect of phrasing is overlooked, the music becomes as comprehensible as the
soliloquy of the unfortunate actor—very far from the great Hamlet described
above—who, in a panic, stumbles to the front of the stage and gabbles out:
‘Tobeornottobethatisthequestionwhethertisnobler’, etc. Not a consummation
devoutly to be wished.
In short, the musician’s challenge is to convey the sentences, the paragraphs
and the overall narrative, the personalities, utterances and destinies of the
musical characters, with as much clarity as possible. In even shorter: the aim
is to communicate the meaning of the music—what a surprise! But it requires
thought as well as feeling, study as well as spontaneity. It also requires the
technical mastery that enables the performer to shape each phrase as the music
requires. It’s not that straightforward, in fact—not as easy as it should sound.
5
Zygonic theory
Zygonic theory seeks to answer the question of how it is that music makes
sense: how, in the absence of semantic content, it is structured and forms
abstract narratives in sound that convey meaning over time. The theory is
‘psychomusicological’ in nature, in that it advances a musicological hypothesis
underpinned by psychological principles. Hence it is an epistemological hybrid,
in which the idiographic intuitions characteristic of music theory and analysis
129
130 Music and Shape
FIGURE 5.1 Oboe and cor anglais duet from the third movement of Vaughan Williams’ Fifth
Symphony
complete arrowheads may be filled (in the case of values extended in time) or
open (in the case of singularities). An example in the domain of duration is
shown in Figure 5.3, where a crotchet tied to a semiquaver in the oboe part
is imitated a beat later by a crotchet tied to a triplet quaver in the cor anglais.
Secondary zygons may be deemed to link primary interperspective relation-
ships, where one is thought to imitate another. In Figure 5.3, examples pertain-
ing to pitch and onset are shown. Once more, superscripts and subscripts are
used to indicate the perspective domain in which the relationship exists and its
level. Finally, tertiary zygons may connect secondary relationships in the mind
of the listener. These occur in the domain of perceived time, for instance, where
there is a regular accelerando or ritardando; see also Figure 5.8.
It is believed that zygonic relationships such as those depicted in Figure 5.3
offer a highly simplified representation of certain cognitive events that we may
reasonably suppose take place (typically nonconsciously) during meaningful
participation in musical activity—whether listening, performing or creating
music anew through improvisation or composition. Moreover, the single con-
cept of a zygon bequeaths a vast perceptual legacy, with many possible mani-
festations: potentially involving any perceived aspect of sound, existing over
different periods of perceived time, and operating within the same and between
different pieces, performances and hearings. Zygons may function in a num-
ber of ways: reactively, for example, in assessing the relationship between two
extant values, or proactively, in ideating a value as an orderly continuation from
one presented. They may operate between anticipated or remembered values,
or even those that are wholly imagined, only ever existing in the mind. (There
is, of course, no suggestion that the one concept is cognitively equivalent in all
these manifestations, only that it is logically so.) Even a short passage of music
comprises a large number of perspective values, potentially linked through a
vast network of relationships, whose effect would be perceptually overwhelm-
ing were it not for the fact that the mind seeks (and is able to find) groups of
relationships that give the impression of acting together in coordinated fash-
ion. This issue is considered at length elsewhere (for example, Ockelford 2005).
In one of the first main expositions of zygonic theory (Ockelford 1999), I hinted
that the notion of the creation and cognition of structure through imitation
need not be limited to music (and so to the perceptual domains pertaining to
sound)—potentially having application to painting, sculpture and ballet, for
example. In relation to art, I observed that ‘pictures normally display an inner
coherence that ultimately derives from the repetition of one or more of its per-
ceived aspects, such as colour, size, shape or texture. An abstract drawing that
Shape in music notation 135
Loc
(3,1)
1
Loc
2
Loc
1 (3,1)
Loc
C
1 (3,1)
FIGURE 5.7 A secondary zygonic relationship of location reflects the fact that the difference in
location between dots B and C is deemed to exist in imitation of the difference between A and B.
Shape in music notation 137
Loc
3 Loc
×1.5
Loc (6.75,2.25)
2
1
Loc
×1.5
2 Loc
(4.5,1.5)
1
Loc
1 (3,1)
tal)
izon
(hor
Loc
2 (h)
Loc
1 +3x
(h)
Loc
+3x +y
1
(v)
Loc
+y 1
(v)
Loc
ical)
(vert
1 Loc
FIGURE 5.9 The perceived orderliness inherent in a straight line modelled in zygonic terms
All images can ultimately be considered as being made up of dots like those
shown in Figures 5.4–5.8. One way of modelling the orderliness of a straight
line is to regard it as describing consistent change in the horizontal and vertical
dimensions, as shown in Figure 5.9. The filled arrowheads symbolize a theoreti-
cally infinite number of relationships that are the same. The connecting (dashed)
lines indicate relationships that work in parallel. A cluster of dots perceived as
forming a single gestalt may be considered to be related to others that are the same
or similar through the compound perspect ‘shape’. Where imitation is thought
to be present, the interperspective relationships between shapes may be deemed
to be zygonic (Figure 5.10). Finally, consider that humanly created images may
be regarded as imitating the visual qualities of objects in ‘real life’ (just as music
may incorporate environmental sounds such as bird-song; see Ockelford 2012b).
138 Music and Shape
Shape
How is it, then, that sounds can be represented through visual images, and
what forms do such analogues take? Scores work on the principle that system-
atic relationships are possible between shape and sound: a shape is taken to have
a meaning such that, in the mind of the performer, the visual image dictates, to
a greater or lesser extent, which sound is to be made, and when. The zygonic
conjecture provides a theoretical framework for modelling how this process
works, and evinces three possible mechanisms through which the visual repre-
sentation of musical sounds can occur. As we shall see, these have a somewhat
convoluted relationship with the threefold Peircean typology of signs: icon,
index and symbol (Peirce [1867–71] 1984: 57; [1893–1917] 1998: 461).
The first form of representation is equivalent to Peirce’s notion of ‘icon’,
whereby a sign denotes its object by virtue of a shared quality. Zygonically
speaking, this is a ‘regular’ mapping, in which the relationship between per-
spective values in different domains involves a function that is not specific to
the perceptual domains concerned—rather, pertaining to a feature or quality
that is abstracted from the perceptual surface and is common to both. To see
this principle in action, consider that, as differences are domain-specific, so
single values of difference cannot be mapped systematically between domains
(see Figure 5.11).4 In contrast, ratios, being abstract, are not bound by the con-
text in which they occur. Hence they permit regular mapping between domains.
See, for example, Figure 5.12. Ratios may occur between differences that are
expressed through secondary relationships. In Figure 5.13 the regular inter-
domain relationships are at the tertiary level. In Peircean terms, the forms of
representation set out in Figures 5.12 and 5.13 are forms of icon that he called
‘diagrams’, ‘which represent the relations, mainly dyadic, or so regarded, of
the parts of one thing by analogous relations in their own parts’ (Peirce [1893–
1913] 1998: 274).
Pitch
+M2
494 Hz
?
P/Loc
Loc
P
2
What would
440 Hz imitation of
pitch by location
mean in these
circumstances?
FIGURE 5.11 Single interperspective values of difference cannot be imitated between domains;
therefore, systematic mapping and iconic representation in Peircean terms are not possible.
140 Music and Shape
Length
×2
×2
Duration
Length
D/length
1
Sound sources Perceived as
2
Imitation of
the ratio between
two durations by
the ratio between
two lengths is possible
FIGURE 5.12 Domains whose perspective values are capable of conveying a sense of size can bear
cross-modal imitation of ratios at the secondary level and therefore have the capacity for iconic
representation.
Pitch
523Hz
+m2
Loc
+y
×0.5
1
×0.5
Loc
494Hz P/Loc
P
2
2
+2y
3
+M2
Loc
Tertiary zygonic
1
imitation of ratio
P
440Hz
FIGURE 5.13 Iconic representation of pitch in terms of location through tertiary-level imitation
1977: 28), since in the vertical dimension two interperspective differences exist
between each dot and the lines, between which a ratio can be gauged and trans-
ferred cross-modally through a tertiary relationship. To the extent that this rela-
tionship can be regarded as imitative, so it can be classed as zygonic.
Consider that the constellations provide temporal performance information
too: according to Stockhausen ‘The points should be played/sung in a short
terse manner, corresponding rhythmically to their graphic layout’ (ibid.: 28).
Although the tempo is not prescribed, this implies tertiary imitation of onsets
as shown in Figure 5.14. And, finally, note that the size of dots in Stockhausen’s
score is linked to intensity: ‘There are points of 5 different thicknesses to which
5 degrees of loudness should correspond. The thickest point is played/sung at ff.
The smallest point corresponds to the respective tutti volume, and all other
points are relatively louder’ (ibid.: 29; emphasis in original). Because both size
and intensity can be compared through primary interperspective ratios, so
cross-modal relationships are possible at the secondary level. Again, these may
be deemed to be zygonic if the changes in musical dynamics are taken to exist
through imitation of their visual representations.
So much for ‘regular’ cross-modal relationships, which offer a coherent
way of linking visual images to musical sounds, and which may convey a
sense of derivation. There are other, ‘irregular’, possibilities too, which may
be ‘indirect’ or ‘arbitrary’. Examples of the former are to be found in the
charts that were part of some scores in the second half of the twentieth
century and beyond, which show wind players the required disposition of
their fingers in order to produce certain combinations of pitches simultane-
ously. For instance, an image such as that shown in Figure 5.15 may appear
FIGURE 5.15 Indirect connection between graphic and sound (Photographic image © Felicity
Ockelford, 2014)
Shape in music notation 145
Visual correlate
Auditory correlate 494Hz
Relationships like this are not imitative in nature, since they stem directly
from the idiosyncratic wiring of an individual’s neural circuitry; nor do they
readily fit within the threefold Peircean typology of signs. Once externalized,
however, imitation would of course be possible, and the colour, shape or other
image could function as a symbol in Peircean terms.
In anticipation of our discussion below concerning forms of music
notation that are accessible to touch readers, we now consider the extent
to which the principles of perceived structure in two-dimensional visual
images may transfer to the tactile figures used by blind people from the sec-
ond half of the twentieth century, when two main ways were developed to
convert lines and shapes into tactile form. The first, known as ‘thermoform-
ing’, constitutes a vacuum-moulding process using thin sheets of plastic,
which permit the shape and texture of small objects to be copied and stored
in a relatively easily manageable form. The second employs ‘swell-paper’,
which, when heated, produces raised lines and shapes in response to black
images (Edman 1992; Ockelford 1996a). Unfortunately, both approaches
are labour-intensive and time-consuming. However, refreshable haptic dis-
plays, akin to video screens for touch, are increasingly making analogues
of visual materials more readily accessible to blind people (Rastogi and
Pawluk 2013).
To the extent that the salient data from visual images can be detected when
they are reproduced in tactile form, so the kinds of relationships described and
illustrated above pertaining to dots, lines and shapes may be perceived in the
domain of touch (Révész 1950; Lechelt, Eliuk and Tanne 1976; Heller 1991;
Lederman and Klatzky 2009). It is my contention that these may be imitative
and therefore function zygonically. The perceptual and cognitive challenges of
assimilating information by touch alone should not be underestimated, how-
ever: readers of tactile scores can perceive what lies beneath their moving fin-
gertips only at any given point in time, which makes distances and angles hard
to judge. Moreover, two-dimensional images larger than a square centimetre or
so have to be mentally reconstructed from series of sensations gleaned through
a painstaking process of digital scanning, making considerable demands on
memory. Nonetheless, complex musical information can be encoded and
decoded in tactile form, as we shall see.
To summarize: in this section, using zygonic theory, we have identified
four ways in which qualities of musical sounds and shapes (or their tactile
equivalents) may be related systematically in cognition. Such relationships
inform the design of musical scores and have enabled them to function in
various ways to represent sounds and to instruct performers how and when
to produce them. The fourfold taxonomy of sound–shape relationships and
its connections with Peirce’s tripartite classification of signs is shown in
Figure 5.19.
Shape in music notation 149
sound-shape relationships
regular synaesthetic
(iconic) irregular
indirect arbitrary
(indexical and iconic) (symbolic)
FIGURE 5.19 Taxonomy of the possible types of relationship between musical sounds and visual
images
Welch (1991), following Walker (1981, 1985, 1987), investigated the mental
images that congenitally blind children produced in response to auditory
stimuli by having them depict the variation in pitch of quasi-musical sounds
on a thin plastic membrane known as ‘German film’, on which is it possible
to produce raised lines using a stylus. As Walker had done before him, Welch
found that blind children systematically associated changes in pitch with the
vertical position on the page (Welch 1991: 220), in just the same way as their
FIGURE 5.21 Regular cross-modal mapping between sound and score, and score and sound
152 Music and Shape
sighted peers do. Whether this striking similarity of mental imagery, irrespec-
tive of vision, resulted from a central cognitive processing mechanism that
works across a number of perceptual modalities, or was merely a consequence
of a common musical metalanguage used in the education of both blind and
sighted children (in which notes are said to be ‘high’ or ‘low’, for example),
remained a moot point. Evidence for the former view is to be found in a range
of psychological work: for example, that of Pratt (1930) and of Roffler and
Butler (1968), whose research (including, in the latter case, participants who
were blind) led them to conclude that every tone has an intrinsic spatial char-
acter, a finding supported by the investigation of Rusconi et al. (2006), which
also showed that the internal representation of pitch is spatial in nature; the
experiments of Mudd (1963) and Thorpe (2015), discussed above, in which a
common pattern of pitch-position mapping was found in most participants;
and the empirical enquiry by Küssner and Leech-Wilkinson (2014), who
found that the majority of their research participants (particularly trained
musicians) represented pitch with height when using a real-time drawing para
digm. Other research, however, in the field of ethnomusicology—pointing to
the fact that, in certain cultures, pitch is not conceived of as high and low,
but ‘small’ and ‘large’ in Bali and Java, for example (Brinner 2008), ‘young’
and ‘old’ in the Amazonian basin (Seeger 2004), and ‘thick’ and ‘thin’ among
Farsi, Turkish and Zapotec speakers (Shayan, Ozturk and Sicoli 2011)—pro-
vides support for notion that pitch/space metaphors are not universal, but
language-based (see Zbikowski 2002: 66–76). However, the position is not
clear-cut: Walker’s comparative study (1987) involving children from indig-
enous Canadian ethnic groups including the Inuit, Haida, Secwepemc and
Tsimshian found that, overall, participants displayed a proclivity for asso-
ciating pitch with vertical placement (rather than pattern, shape or horizon-
tal length). This apparent contradiction may be explained by the finding of
Eitan and Timmers (2010), that diverse cross-domain mappings for pitch
exist latently in western participants in addition to the verticality meta-
phor—a conclusion supported by the work of Antovic (2009) with Serbian
and Romani children, and Dolscheid et al. (2013). Dolscheid’s research
team, based in Nijmegen, found that Dutch speakers’ tendency to describe
pitches as ‘hoog’ (high) and ‘laag’ (low) could be overridden through training,
whereby they could learn to conceive of pitch in the way that, as we noted
above, Farsi speakers do—as ‘naazok’ (thin) and ‘koloft’ (thick). Dolscheid
et al. took this to support the Whorfian hypothesis (Whorf [1956] 2012) that
language affects, in a fundamental way, the nature of perception and cogni-
tion. The impact of culture on perceptual salience is important too, as the
work of Athanasopoulos and Moran (2013) shows: here, a nonliterate Papua
New Guinean tribe, the BenaBena, produced iconic responses to short musi-
cal stimuli, which focused on hue and loudness rather than the variation in
Shape in music notation 153
pitch that proved to be most significant for western and Japanese musicians.
Clearly, the visual metaphors we intuitively use in conceptualizing music con-
stitute an area ripe for further cross-cultural psychological studies.
A typical example of the responses given by Welch’s (western-encultur-
ated) participants is shown in Figure 5.22. Here the stimulus, which had been
produced by Walker for his experiments (see 1987: 495), comprised two pitch
glides, produced by a linear sweep in frequency from 571 Hz to 800 Hz over
a period of two seconds and its reversal, separated by a second’s silence. The
sounds had been generated using a Roland CS15 synthesizer and were as pure
as it was practicable to make them, with 95 per cent of the total spectral energy
lying at the fundamental frequency. Lines produced on German film are inevi-
tably somewhat jagged due to the way in which the stylus presses into the
plastic, though inspection of the children’s efforts as a whole suggests that the
lines in Figure 5.22 were intended to be completely straight. Zygonic analy
sis indicates that there are potentially two forms of cross-modal imitation
functioning here at the tertiary level, whereby horizontal distance on the page
STIMULUS
Time
(seconds)
0 1 2 3 4 5
RESPONSE
(after Welch 1991)
German film
raised lines
produced
with a
stylus
FIGURE 5.22 A congenitally blind child’s representation of pitch glides on German film
154 Music and Shape
FIGURE 5.23 Cross-modal imitation at the tertiary level assumed to underlie the representation of a
pitch g lide as a straight diagonal line
equates to perceived time, and height corresponds to pitch (see Figure 5.23).
Semiotically speaking, the child’s depiction of pitch is iconic.
STAFF NOTATION
Conventional western music notation, using staves of five lines, clefs, time sig-
natures and a range of arbitrary signs to indicate the duration of notes, rests,
articulation, phrasing, dynamics and other aspects of performance, presents,
in Peircean terms, an intriguing mix of iconic and symbolic representation.
FIGURE 5.24 Western staff notation embeds arbitrary symbols within a semi-regular framework of
pitch and time
156 Music and Shape
Until the twentieth century, blind musicians learned new pieces by ear or by
having someone read out the information contained in a score—approaches
that continue to this day in many cultures and in aural traditions. However,
having to rely on others to render musical information that is freely available to
their sighted peers represents an unwelcome loss of autonomy for blind peo-
ple wishing to access music in notated form. It was Louis Braille himself—the
inventor of the literary braille code—who devised the first workable system
of reading and writing music in tactile form. His initial ideas were published
in 1829 (Lorimer 1996), though it took almost a century for braille music to
become established as the main method of making staff notation available for
blind people (Kersten 1997). A braille transcription of the excerpt from Figure
5.24 is shown in Figure 5.25.
Braille comprises cells of six potential dots, yielding 26 or sixty-four pos-
sible combinations (including the blank cell). These are arranged from left
to right in horizontal lines in the same way as print. Cells are read, one at a
time, as the index finger traces over them. Reading music through braille is not
entirely equivalent to using a print score, however, for a number of reasons.
First, the way that cells are laid out on the page means that they have to be read
in series: the two-dimensional nature of printed music scores, over which the
eye scans horizontally and vertically, is necessarily compressed into single lines.
Second, the limit of sixty-four dot-and-blank combinations makes context-
dependent meanings inevitable, with all the information required to define a
single note often necessarily being conveyed in more than one cell. Together,
these characteristics mean that the iconic elements of print notation—the
portrayal of pitch and time in the vertical and horizontal dimensions of the
page through imperfect tertiary zygonic imitation (see again Figure 5.24)—are
absent, and the visual representation of the ‘gist’ of the music is missing. All
that remains are arbitrary signs that function symbolically in Peircean terms;
this is one reason music in braille is more difficult to learn and use than its print
equivalent (Ockelford 1991a, 1996b).
FIGURE 5.25 Music Time in braille music notation (represented in print form), with explanations of the signs
158 Music and Shape
Matrices of potential dots, ostensibly similar to those used in braille, also char-
acterize guitar chord symbols. The way that these function as signs is very dif-
ferent, however. The four chords used in the excerpt shown in Figure 5.24 may
be visually communicated to guitarists as indicated in Figure 5.26. There are
three semiotic processes at work here. First, there is an iconic link between each
graphic and the positioning of the guitarist’s fingers (Figure 5.15). Second,
each hand position on the strings functions as an index for the chord that is
produced. Third, the distance between the ‘nut’ (represented by the thick black
horizontal line at the top of each graphic) and each of the points where the
fingers press on the strings (shown by the black ellipses) is analogous with the
imaginary interval between the pitch to which the string is tuned and the note
that actually sounds (see Figure 5.27).
FIGURE 5.26 The fingering for the opening four chords of Music Time presented using guitar chord
symbols
Shape in music notation 159
FIGURE 5.27 The three semiotic processes at work as a guitarist performs from a chord symbol
(photographic image © Felicity Ockelford, 2014)
FIGURE 5.28 Fragment of Jamie Roberts’ synaesthetically derived score of Jean-Michel Jarre’s
Oxygène, track 4
FIGURE 5.29 Types of semiosis functioning in a fragment of Jamie Roberts’ synaesthetic score of
Oxygène
Conclusion
This chapter explored the function of shape in music notation and set out a
model, using the principles of zygonic theory, that aimed to show how forms of
cross-domain mapping between musical sounds and visual images may logically
occur in cognition. Four types of relationship between the perceptual domains
pertaining to hearing and vision were identified: regular, irregular (which may
Shape in music notation 161
References
Antovic, M., 2009: ‘Musical metaphors in Serbian and Romani children: an empirical
study’, Metaphor and Symbol 24/3: 184–202.
Athanasopoulos, G. and N. Moran, 2013: ‘Cross- cultural representations of musical
shape’, Empirical Musicology Review 8/3–4: 185–99.
Bamberger, J., 1995: The Mind behind the Musical Ear: How Children Develop Musical
Intelligence (Cambridge, MA: Harvard University Press).
162 Music and Shape
Bamberger, J., 2013: Discovering the Musical Mind: A View of Creativity as Learning
(Oxford: Oxford University Press).
Baron-Cohen, S. and J. Harrison, eds., 1996: Synaesthesia: Classic and Contemporary
Readings (Hoboken, NJ: Wiley-Blackwell).
Baron-Cohen, S., L. Burt, F. Smith-Laittan, J. Harrison and P. Bolton, 1996: ‘Synaesthesia:
prevalence and familiarity’, Perception 25/9: 1073–9.
Boulez, P., [1963] 1971: Boulez on Music Today, trans. S. Bradshaw and R. R. Bennett
(London: Faber and Faber.
Brinner, B., 2008: Music in Central Java: Experiencing Music, Expressing Culture (New York:
Oxford University Press).
Chalmers, D., 1996: The Conscious Mind: In Search of a Fundamental Theory (New York:
Oxford University Press).
Cone, E., 1987: ‘On derivation: syntax and rhetoric’, Music Analysis 6/3: 237–56.
Cross, I., 1998: ‘Music analysis and music perception’, Music Analysis 17/1: 3–20.
Cytowic, R., D. Eagleman and D. Nabokov, 2011: Wednesday Is Indigo Blue: Discovering
the Brain of Synesthesia (Cambridge, MA: MIT Press).
Deutsch, D., 2012: The Psychology of Music, 3rd edn (Waltham, MA: Academic Press).
Dolscheid, S., S. Shayan, A. Majid and D. Casasanto, 2013: ‘The thickness of musical pitch:
psychophysical evidence for linguistic relativity’, Psychological Science 24/5: 613–21.
Edman, P., 1992: Tactile Graphics (New York: AFB Press).
Eitan, Z. and R. Timmers, 2010: ‘Beethoven’s last piano sonata and those who follow croc-
odiles: cross-domain mappings of auditory pitch in a musical context’, Cognition 114/
3: 405–22.
Fauconnier, G., [1985] 1994: Mental Spaces: Aspects of Meaning Construction in Natural
Language (Cambridge: Cambridge University Press).
Gjerdingen, R., 1999: ‘An experimental music theory?’, in N. Cook and M. Everist, eds.,
Rethinking Music (Oxford: Oxford University Press), pp. 161–70.
Harrison, J., 2001: Synaesthesia: The Strangest Thing (Oxford: Oxford University
Press).
Heller, M., 1991: ‘Haptic perception in blind people’, in M. Heller and W. Schiff, eds., The
Psychology of Touch (Hillsdale, NJ: Erlbaum), pp. 129–60.
Kanai, R. and N. Tsuchiya, 2012: ‘Qualia’, Current Biology 22/10: R392–6.
Kersten, F., 1997: ‘The history and development of braille music methodology’, The
Bulletin of Historical Research in Music Education 18/2: 106–25.
Kowler, E., 2011: ‘Eye movements: the past 25 years’, Vision Research 51/13: 1457–83.
Küssner, M. and D. Leech-Wilkinson, 2014: ‘Investigating the influence of musical training
on cross-modal correspondences and sensorimotor skills in a real-time drawing para-
digm’, Psychology of Music 42/3: 448–69.
Lakoff, G., 1987: Women, Fire, and Dangerous Things: What Categories Reveal about the
Mind (Chicago: University of Chicago Press).
Lechelt, E., J. Eliuk and G. Tanne, 1976: ‘Perceptual orientational asymmetries: a compari-
son and visual and haptic space’, Perception and Psychophysics 20/6: 463–9.
Lederman, S. and R. Klatzky, 2009: ‘Haptic perception: a tutorial’, Attention, Perception,
& Psychophysics 71/7: 1439–59.
Lorimer, P., 1996: ‘A critical evaluation of the historical development of the tactile modes
of reading and an analysis and evaluation of researches carried out in endeavours to
Shape in music notation 163
make the Braille code easier to read and to write’ (PhD dissertation, University of
Birmingham).
Mudd, S., 1963: ‘Spatial stereotypes of four dimensions of pure tone’, Journal of Experi
mental Psychology 66/4: 347–52.
Ockelford, A., 1991a: Music and Visually Impaired Children: Some Notes for the Guidance
of Teachers (London: Royal National Institute for the Blind).
Ockelford, A., 1991b: ‘The role of repetition in perceived musical structures’, in P. Howell,
R. West and I. Cross, eds., Representing Musical Structure (London: Academic Press),
pp. 129–60.
Ockelford, A., 1996a: Music Matters: Factors in the Music Education of Children and Young
People Who Are Visually Impaired (London: Royal National Institute of the Blind).
Ockelford, A., 1996b: Points of Contact: A Braille Approach to Alphabetic Music Notation
(London: Braille Authority of the United Kingdom).
Ockelford, A., 1999: The Cognition of Order in Music: A Metacognitive Study (London:
Roehampton Institute).
Ockelford, A., 2002: ‘The magical number two, plus or minus one: some limits on our
capacity for processing musical information’, Musicae Scientiae 6/2: 185–219.
Ockelford, A., 2005: Repetition in Music: Theoretical and Metatheoretical Perspectives
(Aldershot: Ashgate).
Ockelford, A., 2006: ‘Implication and expectation in music: a zygonic model’, Psychology
of Music 34/1: 81–142.
Ockelford, A., 2009: ‘Zygonic theory: introduction, scope, prospects’, Zeitschrift der
Gesellschaft für Musiktheorie 6/1: 91–172.
Ockelford, A., 2012a: Applied Musicology: Using Zygonic Theory to Inform Music
Psychology, Education and Therapy Research (New York: Oxford University Press).
Ockelford, A., 2012b: ‘What makes music “music”? Theoretical explanations using zygonic
theory’, in J.-L. Leroy, ed., Actualités des universaux musicaux (Topicality of Musical
Universals) (Paris: Editions des Archives Contemporaines), pp. 123–48.
Ockelford, A., 2013: Music, Language and Autism: Exceptional Strategies for Exceptional
Minds (London: Jessica Kingsley).
Ockelford, A. and C. Matawa, 2009: Focus on Music 2: Exploring the Musical Interests
and Abilities of Blind and Partially-Sighted Children with Retinopathy of Prematurity
(London: Institute of Education).
Peirce, C., [1867–71] 1984: Writings of Charles S. Peirce: A Chronological Edition, vol. 2
(Bloomington: Indiana University Press.
Peirce, C., [1893–1913] 1998: The Essential Peirce: Selected Philosophical Writings, vol. 2
(Bloomington: Indiana University Press.
Pomerantz, J. and M. Portillo, 2011, ‘Grouping and emergent features in vision: toward a
theory of basic Gestalts’, Journal of Experimental Psychology: Human Perception and
Performance 37/5: 1331–49.
Pratt, C., 1930: ‘The spatial character of high and low tones’, Journal of Experimental
Psychology 13/3: 278–85.
Rastogi, R. and D. Pawluk, 2013: ‘Dynamic tactile diagram simplification on refreshable
displays’, Assistive Technology 25/1: 31–8.
Révész, G., 1950: Psychology and Art of the Blind, trans. H. Wolff (London: Longmans,
Green).
164 Music and Shape
Risset, J.-C. and D. Wessel, 1999: ‘Exploration of timbre by analysis and synthesis’, in
D. Deutsch, ed. The Psychology of Music, 2nd edn (New York: Academic Press),
pp. 113–69.
Roffler, S. and R. Butler, 1968: ‘Localization of tonal stimuli in the vertical plane’, The
Journal of the Acoustical Society of America 43/6: 1260–6.
Roskies, A., 1999: ‘The binding problem’, Neuron 24/1: 7–9.
Rusconi, E., B. Kwan, B. Giordano, C. Umilta and B. Butterworth, 2006: ‘Spatial represen-
tation of pitch height: the SMARC effect’, Cognition 99/2: 113–29.
Sagiv, N. and J. Ward, 2006: ‘Cross-modal interactions: lessons from synesthesia’, in S.
Martinez-Conde, S. Macknik, L. Martinez, J.-M. Alonso and P. Tse, eds., ‘Visual per-
ception—fundamentals of awareness: multi-sensory integration and high-order percep-
tion’, Progress in Brain Research 155: 263–75.
Seeger, A., 2004: Why Suyá Sing: A Musical Anthropology of the Amazonian People
(Champaign, IL: University of Illinois Press).
Shayan, S., O. Ozturk and M. Sicoli, 2011: ‘The thickness of pitch: crossmodal metaphors
in Farsi, Turkish, and Zapotec’, The Senses and Society 6/1: 96–105.
Slawson, W., 1985: Sound Color (Berkeley: University of California Press).
Stevens, S. S., 1975: Psychophysics: Introduction to Its Perceptual, Neural, and Social
Prospects (New Brunswick, NJ: Transaction).
Stockhausen, K., 1977: Sternklang: Park-Music für 5 Gruppen (Kürten: Stockhausen-Verlag).
Stone, K., 1980: Music Notation in the Twentieth Century: A Practical Guidebook
(New York: Norton).
Thorpe, M., 2015: ‘The cognition of pitch patterns and cross-modal spatial structure’ (PhD
dissertation, University of Roehampton).
Walker, R., 1981: ‘The presence of internalised images of musical sounds and their rel-
evance to music education’, Bulletin of the Council for Research in Music Education
66/67: 107–12.
Walker, R., 1985: ‘Mental imagery and musical concepts: some evidence from the congeni-
tally blind’, Bulletin of the Council for Research in Music Education 85: 229–38.
Walker, R., 1987: ‘The effects of culture, environment, age, and musical training on choices
of visual metaphors for sound’, Perception and Psychophysics 42/5: 491–502.
Welch, G., 1991: ‘Visual metaphors for sound: a study of mental imagery, language and
pitch perception in the congenitally blind’, Canadian Journal of Research in Music
Education 33 (Special ISME Research Edition): 215–22.
Whorf, B., [1956] 2012: Language, Thought, and Reality, 2nd edn, ed. J. Carroll, S. Levinson
and P. Lee (Cambridge, MA: MIT Press).
Zbikowski, L., 2002: Conceptualizing Music: Cognitive Structure, Theory, and Analysis
(New York: Oxford University Press).
Reflection
Alice Eldridge, cellist and coder
165
166 Music and Shape
These patterns we concoct for practice can be seen as mental images that allow
us to integrate representations in motor, visual and sonic processes—shapes
in our bodies, eyes and ears. Initially, integration of these mappings requires
conscious effort. As a child I learned to read music by first learning Curwen’s
solfège handsigns (see Beach 1914) and songs about colourful insect characters:
‘C is for Clarence Caterpillar, D is for Dora Dragon Fly’, etc., forging links
between note names and their position in physical, pitch and visual space. As
you learn an instrument, another set of mappings is established, from the notes
on the staff to the fingerings necessary to produce the designated pitches. These
visual–motor mappings rapidly take precedence, a phenomenon neatly illus-
trated by scordatura notation. Bach’s Cello Suite No. 5 is written for scordatura
tuning as shown in Figure R.12 (top right). When I see the top interval of the
chord in bar 2 of the Prelude, I ‘know’ and hear a minor third but quite happily
read and finger a perfect fourth, suggesting that I am not reading ‘B♭’ at all, but
‘first finger in half position on the A string’.
The integral role of muscle memory in musicianship is nothing new: research
into musical implications of motor theories of perception abounds (e.g.
Godøy 2010). Anecdotes from expert musicians deftly illustrate. Improviser
Steve Beresford, speaking before a recent concert of the London Improvisers’
Orchestra, remarked that despite having moved from trumpet to piano decades
ago, he still automatically thinks about melody lines in terms of trumpet fin-
gerings. The sight of musicians (saxophonists and keyboard players espe-
cially) air-fingering in gigs to work out the thrust of a solo they’re hearing is
not uncommon. These are not just ‘air instrument’ performances, mimicking
sound-producing gestures (Godøy, Haga and Jensenius 2006): these habits of
expert musicians suggest that actual or imagined activation of motor schemata
FIGURE R.12 Opening of the Prelude of Bach’s Cello Suite No. 5 in scordatura notation
Reflection: Alice Eldridge 167
Guitarist John Russell talks about free improvisation as the closest he gets
to ‘what music actually is’.3 I might go further and suggest that improvising
with others distils many of the joys of being human, capturing the best bits of
168 Music and Shape
conversation, cooking, dancing and dressing up, meeting people and testing
and exchanging new ideas: instantaneously making something. At such times,
the deliberate planning, monitoring and manipulation of activities give way to
more intuitive processes.
In one improviser’s comments, the lack of conscious engagement is almost
the hallmark of ‘good’ improvising, active consideration of shape coming into
play only when musical spontaneity ebbs:
With some musicians I have a really good connection, the music flows
and I don’t tend to think a lot. I don’t think much about shape (and
sometimes not at all) but it feels like I am—and the other musicians are—
working with it on a more unconscious level. I close my eyes, using my
ears, feeling of the soundwaves in my body, the colours I get when my
eyes are closed and the mood it all puts me in to play and reflect on the
music. With other musicians it can be harder to get this flow and I tend to
start thinking more and also be more conscious about the shape and how
to do it. (Julie Kjær, saxophonist)
References
Beach, C. B., ed., 1914: ‘Curwen, John’, The New Student’s Reference Work (Chicago:
Compton).
Godøy, R. I., 2010: ‘Gestural affordances of musical sound’, in R. I. Godøy and M.
Leman, eds., Musical Gestures: Sound, Movement, and Meaning (London: Routledge),
pp. 103–25.
Godøy, R. I., E. Haga and A. Jensenius, 2006: ‘Playing “air instruments”: mimicry of
sound-producing gestures by novices and experts’, in S. Gibet, N. Courty and J.-F.
Kamp, eds., Gesture in Human-Computer Interaction and Simulation: 6th International
Gesture Workshop, Lecture Notes in Artificial Intelligence 3881 (Berlin: Springer), pp.
256–67.
Griffiths, D., 2008: ‘Scheme bricks’. Software available at https://fo.am/scheme-bricks/
(accessed 9 April 2017).
Limb, C. J. and A. R. Braun, 2008: ‘Neural substrates of spontaneous musical perfor-
mance: an fMRI study of jazz improvisation’, PLoS One 3/2: e1679.
Magnusson, T., 2011a: ‘ixi lang: a SuperCollider parasite for live coding’, in Proceedings
of the International Computer Music Conference, 31 July–5 August 2011, University of
Huddersfield (conference document), pp. 503–6. Software available at https://github.
com/thormagnusson/ixilang (accessed 9 April 2017).
Magnusson, T., 2011b: ‘Algorithms as scores: coding live music’, Leonardo Music Journal
21: 19–23.
Magnusson, T., 2011c: ‘Confessions of a live coder’, in Proceedings of the International
Computer Music Conference, 31 July–5 August 2011, University of Huddersfield (confer-
ence document), pp. 609–16.
McLean, A., 2014: ‘Making programming languages to dance to: live coding with Tidal’, forth-
coming in Proceedings of the 2nd ACM SIGPLAN International Workshop on Functional
Art, Music, Modelling and Design, Gothenburg, Sweden, 1–3 September 2014. Software
available at https://github.com/tidalcycles/Tidal (accessed 9 April 2017).
McLean, A. and G. Wiggins, 2011: ‘Texture: visual notation for the live coding of pat-
tern’, in Proceedings of the International Computer Music Conference, 31 July–5 August
2011, University of Huddersfield (conference document), pp. 612–28. Software available
at https://github.com/yaxu/texture (accessed 9 April 2017).
6
Contemporary enquiries into the art of musical improvisation cross a range of dis-
ciplines from cultural studies, pedagogy, psychology, neuroscience, mathematical
and computer modelling, and quantitative and qualitative analyses to ethnogra-
phy. Even a succinct survey of the state of current research would fill a volume,
as is meticulously demonstrated in Berkowitz’s The Improvising Mind (2010) and,
in the context of jazz, Berliner’s seminal Thinking in Jazz (1994). This chapter
addresses the concept of shape in musical improvisation, and offers insights into
how it might be described, identified or created. Jazz pedagogy is used to pres-
ent key improvisational mechanisms and as a foundation on which a model of
improvisation is constructed. The concept of shape in improvisation is addressed
within this model. Taking examples from jazz repertoire, we offer an approach
to categorizing improvisational shape. Finally, these concepts are adopted in the
detailed analysis of a classical cadenza, demonstrating the wide applicability (and
stylistic neutrality) of this approach to improvisational analysis.
One must acknowledge the complex continua that exist between composi-
tion, performance and improvisation (Benson 2003), as well as the role of intu-
ition, spontaneity and ‘polished improvisation’ in composition. Nonetheless,
certain musical practices rely on—indeed, stylistically require—a significant
degree of spontaneity, or at any rate a purposeful avoidance of premeditation
or prescription. Performance-without-complete-blueprint—as might be found
in a jazz solo, a Hindustani raga, a classical cadenza or ‘free’ improvisation—
may itself involve significant preparation and hard-earned skill, but there
is a relatively higher number of musical choices made ‘in the moment’ than
are found in the typical performance of a score or memorized piece. In these
contexts, the audience and (and until the last moment) the performer(s) can-
not be fully aware of what music will emerge. With this definition it might be
argued that there are no completely improvised musics and—if performed by
170 humans—no music that avoids spontaneity, or at least the possibility of small
The shape of musical improvisation 171
variation, in performance. A fully notated score is never just that, and, even if
notes and relative pitch durations are prescribed meticulously, in the moment
of performance spontaneous expressive (or unavoidable) alterations of tempo,
dynamic, timbre and inflection will emerge. There is a spectrum of allowable
spontaneity in performances of all kinds, from subtle nuance to significant
invention (although never without some type of constraint). This chapter deals
with musical shape that might emerge at the inventive edge of the spectrum,
taking examples from the conventional jazz improvisation repertoire (solo-
ing or melodic interpretation within fixed structural and harmonic outlines)
and the improvised classical cadenza (a stylistically constrained exploration of
thematic materials). Furthermore the structural components that are addressed
are distinct from (or additional to) the superficial structural contexts in which
the improvisation might take place (such as the jazz standard lead sheet with
its given form and harmonic framework, or the rhythmic cycle, tala, of a
Hindustani raga). We are in essence considering the shape that might emerge
from spontaneously selected musical elements.
The formal shaping of an improvisation is addressed widely in improvi-
sational pedagogical literature, for example in Hal Crook’s How to Improvise
(1991) or Berkowitz’s survey of nineteenth-century improvisational treatises
(2010: 15–80). Conversely, research into (usually jazz) improvisation has aimed
to reveal musical strategies and structures—however implicit or intuitive—that
occur in improvisational repertoire. These include—citing just three of many
possible examples—Markov chains in Coltrane’s improvisations on Giant Steps
(Franz 1998), phrasing schemata in Charlie Parker’s blues soloing (Love 2012)
and patterns in the transformation of those musical features less amenable to
standard notational analysis (see for example Benadon’s microtiming analysis
of early jazz improvisation in Time Warps in Early Jazz, 2009).
Despite a foundation of jazz practice, this chapter takes an abstracted model
of improvisation (proposed by Mermikides 2010, which shares important fea-
tures of Pressing’s 1988 model) through which improvisational shapes might
be identified. In the model presented here, a musical object is seen as exist-
ing at a point in multidimensional musical space (M-Space) and possessing an
array of properties available for modification. Improvisation is represented as
the artful motion through this space, and characteristics of this motion may
form larger-scale musical structures. This view of improvisation offers practical
applications for performance (and composition) in a range of styles, as well as a
framework within which to analyse and appreciate the repertoire and practice.
Chains of thought
transformations over time (see for example Damian and Feist 2001: 12–20;
Crook 1995: 8–31; and Berliner 1994: 146–69). More specifically, this object
might be perceived as containing a set of musical properties, with each property
open to modification in future phrases. New material is thus created by alter-
ing a selected parameter, or set of parameters, from previous objects. In other
words, improvisation involves the construction of a musical train of thought
where every subsequent object relates to a preceding one in terms of a changing
set of variable parameters. This concept is best fleshed out with a real-world
example. In a section of John Coltrane’s solo in Acknowledgement (Coltrane
1965: 2:06–2:32), to pick one of countless examples from the repertoire, a
simple motive (C, E♭ and F typically in a quaver, quaver, crotchet—or more
generally short, short, long—rhythm) is manipulated in terms of chromatic
transposition, metric placement and rhythmic subdivision (more specifically,
the duration between note onsets also known as an inter-onset interval, IOI).
Even a first listen to the extract will allow a clear identification of the central
role of this musical ‘seed’ in the ensuing improvisation, with these three pri-
mary degrees of transformation. Occasionally, objects merge—for example, the
last note of one phrase becomes the first of the next—but on the whole this
passage provides an idealized, clear (and readily notated) example of this con-
cept of improvisation: the expressive variation of an established musical object.
Many improvisational passages contain patterns that are less immediately
recognizable than the linear chains-of-thought model that Coltrane provides in
the last example. Figure 6.1 shows how an opening phrase might be developed
through a number of complex transformational pathways.
In this analysis, phrases are generally identified as being related to a previ-
ously occurring phrase, and may themselves combine into larger phrases or
break off into smaller ones. They are labelled accordingly. The types of rela-
tionships between phrases are described by sets of transformational processes
in boxed text.
This improvisational methodology thereby involves the selecting of a par-
ticular subset of musical properties of a phrase, which is then either fixed or
modified by varying amounts in the subsequent phrase, to form a series of
interlinking chains. Even in this short extract, the themes of variation, trans
position and recombination (as identified by Berkowitz in the context of classi-
cal improvisational pedagogy; 2010: 39–80), the creation of logical expectations
and their potential subversion (‘rational deception’; Bach [1753, 1762] 1949:
434) and analogies with syntax (Patel 2003) are revealed.
Some further points for consideration:
• There may be many valid analyses of an improvisation, and the
performer’s conception and the listener’s interpretation of the solo
may differ.
• A single phrase may also form the impetus for any number of
subsequent phrases, along any number of transformational processes.
FIGURE 6.1 An illustration of a complex chains-of-thought improvisation methodology (Mermikides 2010)
174 Music and Shape
Here the discussion turns to jazz improvisational pedagogy and how it might
relate to and inform the concept of shape in improvisation generally. The jazz
pedagogical material that started to emerge around the late 1980s from such
educator/practitioners as Hal Crook, Jerry Bergonzi and Mick Goodrick
FIGURE 6.2 An illustration of musical refractions. In the course of an improvisation, a phrase is manipulated by the selection of one of
many transformational processes (1–8 present a few of countless possibilities). The resulting phrase is in turn open to further modifications.
Improvisation is seen as the realization of a pathway through the multitude of refracting musical possibilities (Mermikides 2010).
The shape of musical improvisation 177
The point of such ‘limiting’ exercises is to force new ideas and avenues of
exploration, liberating the improviser from the overwhelming number of
possibilities of the ‘blank canvas’, focusing on specific musical parameters
that might otherwise be overlooked, and avoiding the habitualized patterns
that arise in the absence of guidelines. In Crook’s words, ‘There is no freedom
without structure’ (1991: 55). One might think of this type of approach as the
training of a particular type of skill: the independent and artful modification,
or maintenance, of coexisting musical parameters. This type of pedagogical
literature aims to develop proficiency (conscious and intuitive) in this area to
create authentically chosen material rather than pat phrases at the moment of
improvisation or clichéd responses to a sequence of chords. Improvisational
skill and strategy are thereby developed through the fixing—or variation—of
specific musical parameters.
Alongside jazz pedagogical material, further support for this ‘limit and
vary’ approach, this time from an academic theoretical standpoint, is offered
by Pressing’s (1988) ‘Improvisation: methods and models’, which offers a
model of improvisation as variegated attention paid to selected parameters and
transformational processes. Values of various musical parameters and types
of transformation of a phrase are defined. The amount of attention paid to
each is described with the currency of cognitive strength. This is unlikely to be
more than conjecture, and it is difficult to imagine how it could be measured.
Personal and anecdotal accounts and neuroscientific reports on improvising,
tentative as they may be, seem to suggest that at any particular moment the
creative improviser is thinking actively about one or two musical goals at most
(Werner 1996; Nachmanovitch 1990; Solstad 1991; Limb and Braun 2008).
178 Music and Shape
Fix
Fix Melodic Shape
Key Area Rhythmic Structure
Vary Vary
Rhythmic Density Diatonic Transposition
Melodic Content Rhythmic Placement
Time-Feel
Fix
Fix Intervallic Structure
Harmonic Vary
Implication Chromatic
Vary Transposition
Attack Envelope Rhythmic
Placement
Fix Fix
Fix Fretboard shape Order of Melody
The 3 Melody Notes Vary Notes
in Terms of Scale Rhythmic Vary
Vary Placement Rhythmic
Imply Parallel Scale String Placement Placement of
Melody Notes
FIGURE 6.5 An illustration of how the fixing and variation of musical topics may forge
improvisational continuations from Phrase α (Mermikides 2010)
featured topics (e.g. a particular melodic cell) and thereby exploring deeply
other variables (permutation, harmonic altitude, segmentation, etc.). A con-
temporary bibliography of jazz pedagogical material has started to resemble
a library of chess books with stacks of general-principle texts alongside titles
dedicated to every conceivable opening, variation of opening and style of end
game. The study of these differentiated skills, in both chess and jazz improvi-
sation, aims to offer the player informed options and intuition at the moment
of performance.
This section has introduced, and given some examples of, the concept of
improvisation as a mutation of preceding phrases, through the variable fixing
and variation of various musical parameters. In this way, a phrase is modi-
fied according to various parameters, and wanders from its starting origin
while maintaining a comprehensible narrative for the listener. The next sec-
tion will look more deeply into this idea of trajectories through this wealth
of possibilities, and how this might enhance an understanding of shape in
improvisation.
182 Music and Shape
A crotchet displacement is more distant; the D for example would now fall on
beats 2 or 4, rather than 1 or 3, a more significant change in character. Any qua-
ver displacement alters the phrases yet more extremely, removing the upbeat
and interfering with any swinging of quavers that may be going on. Semiquaver
shifts, and yet finer rational subdivisions (if appreciable), alter the phrase still
more radically. To complicate matters further, the layout of these phrases is not
static: once a rhythmic displacement has been made, phrases are reordered in
terms of proximity. For example, if phrase α is displaced by a crotchet, its ‘near-
est’ neighbour is now a minim away. Incidentally, this representation can more
readily adopt micro-timing features which ‘fall between the cracks’ of standard
notation. With the concept of musical proximity in mind, more dimensions
may be added and a new musical space may be constructed for exploration
(Figure 6.7).
Orthogonal to this rhythmic placement axis, a note separation axis may
also be postulated, representing the progressive elongation and contraction of
phrase α, with wider note separation in one direction and shorter in the other.
This axis might be arranged with the emphatic top D used as a rhythmic anchor
about which the outer two notes are stretched or compressed. The individual
notes may compress until they form a chord and then extend beyond that point
to form a retrograde transformation of phrase α. In addition to rhythmic place-
ment and note separation, another axis may be added that represents all possi-
ble diatonic transpositions of a phrase (diatonic to the key of C Dorian), higher
in one direction and lower in the other. Chromatic transposition within a tonal
harmony creates a nonlinear pattern of musical distance. However, within a
modal setting, from which phrase α is derived, the hierarchical nature of scale
degrees is less clear. The subjective decision has therefore been made to arrange
the proximity in terms of diatonic transposition very simply, so that proxim-
ity in this dimension is equivalent to similarity of melodic register. Given the
definition of these parameters, variations of phrase α exist in three dimensions,
with potential mutations of the phrase existing side by side in conceptual space.
A sense of musical proximity within these constraints may also be perceived.
Now that the concept of proximity has been established, one might also
imagine additional transformational dimensions emerging from phrase α, such
as a chromatic transposition dimension, axes of various timbral characteristics
(including those achievable only through electronic manipulations), points of
symmetry, intonation, segmentations and so on. An impression of how a musi-
cal phrase exists in multiple simultaneous dimensions of transformation, here
termed M-Space, is shown in Figure 6.8.
The coexistence of these multiple dimensions is possible to conceive, but
difficult to illustrate precisely in one diagram. A conceptual model, whose
precise demarcation can be delegated to a computer, might serve better than
two-dimensional illustrations. Regardless, a logical visualization of a phrase
existing within a radiated sphere of closely related musical material may readily
FIGURE 6.7 Phrase α existing at the centre of a three-dimensional musical space with variously
proximate neighbouring phrases. Phrase α is indicated at the origin of the axes and the musical
distance between it and various close neighbours is shown. The boundary of the grey sphere describes
a boundary of equal proximity and contains phrases within this musical distance. The lower part
of the diagram shows an impression of Phrase α existing at a point within this musical space
(Mermikides 2010).
FIGURE 6.8 An impression of M-Space: phrase α sits at the centre of many simultaneous dimensions of musical transformation. Twelve of these are represented in four
three-dimensional subsets (some of which are continuous rather than discrete values) with some proximate phrases indicated. A phrase may move along any number
of such transformational axes during the course of improvisation. In the top right of the diagram, a phrase shows the result of a small move in all of these subsets
simultaneously (the modification is marked as a grey disc in each transformational subset) (Mermikides 2010).
The shape of musical improvisation 187
Field B
Field A B1
A1
B2
A2 B4
A3
B3
Field C
C2 C1
Field E Field D
C3
C4
D1
E1 E2 D2
D3
FIGURE 6.9 A multi-level depiction of Smith’s solo on The Sermon. Improvisation is seen as a
configuration of fields at varying distances and trajectories in M-Space, with each field containing a
constellation of phrases. Phrases, in turn, may be broken down into a nexus of smaller phrase units as
is shown in reference to E2. Phrase E2 has been placed closer to Fields A and C than B and D, to reflect
features of E2.1. Fields themselves are linked together in terms of timbral, registral and temporal
components and may coexist in a yet greater nexus of relationships with other performers or musical
objects (Mermikides 2010).
The shape of musical improvisation 189
M-Space is illustrated in Figure 6.9. Fields A–E co-exist as part of the same
solo, but their relative proximity is also due to registral, timbral and temporal
considerations. Not shown in Figure 6.9 is the yet more complex interaction of
fields and phrases between other performers and musical objects.
In essence, this solo extract might be described as a ‘field series’ in that a
motive is explored in a number of objects (phrases) before jumping to another
area of M-Space for expressive exploration. Although Figure 6.9 is illustrated
in two dimensions, one must be reminded that the relative distances between
fields, and between their constituent phrases, is the cumulative result of their
relative positions in multidimensional space (through variations of many
coexisting musical parameters). Once the concept of M-Space structures and
their relative positions has been grasped, the listening and analytical process
becomes far clearer. From the straight-ahead to the most avant-garde contexts,
it becomes possible to classify improvisations in terms of which parameters are
fixed and varied, and what types of trajectories through M-Space occur.
Other improvisational structures (or strategies) might also exist. One can
hear, for example, in Wes Montgomery’s solo on No Blues (1965: 1:32–2:02),
one very narrow phrase field being used repeatedly as a pivot to other fields in a
‘call-and-response’ manner. A sharply defined F (in octaves) is used as a motive
central to various other phrases (which might belong to several identifiable
fields). This interjected figure is so clearly defined that the other ensemble mem-
bers are compelled to mark it with their own musical material. In other words,
the soloist’s M-Space structures have infiltrated those of the accompanists, as
should occur in any responsive ensemble environment. One might name this
type of improvisational structure a ‘pivot’, where one narrow area of M-Space
is used as a frequent and significant launching point for other satellite objects.
Alternatively, Coltrane’s Acknowledgement (discussed earlier) displays the
strategy of identifying a narrow field and then furtively exploring that space for
extended periods: one might refer to this improvisational shape as ‘nuclear’. Pat
Metheny’s approach on Unquity Road (1976) is less clearly delineated: phrase
fields exist, but the transitions between them are often blurred, and motion
through M-Space relatively slow, so that the result is a ‘merged’ improvisational
structure. Finally, an improvisation may be characterized by repeated distant
jumps in musical space, an ‘unbounded’ improvisational shape, so that there
is little appreciable relationship between successive musical objects (Sheffield
Phantoms (Bailey 1975) may—to many listeners—provide an example).
As analyses of improvised solos are revealed, it becomes possible to sort con-
stituent passages into these kinds of broad category. Note that these are grouped
in terms of the relationships between the phrases rather than according to the
vocabulary. A pictorial comparison of five improvisational strategies is pre-
sented in Figure 6.10—nuclear, field series, pivot, merged and unbounded—of
which Acknowledgement (Coltrane 1965), The Sermon (Smith 1958), No Blues
(Montgomery 1965), Unquity Road (Metheny 1976) and Sheffield Phantoms
1. Nuclear 2. Field Series 3. Pivot
4. Merged 5. Unbounded
FIGURE 6.10 Five improvisational structures: 1) ‘Nuclear’: phrases, with only occasional small anomalies, fall within one close field with only minor variances. 2) ‘Field Series’:
close phrases are played a few times with variances before repeating the process at a different point in M-Space. 3) ‘Pivot’: one particular narrow field is played often, acting as
a springboard to various satellite fields. 4) ‘Merged’: fields are merged by the use of a transitional phrase of otherwise distinct phrase fields. 5) ‘Unbounded’: a series of phrases
with little proximity of one phrase to any other (Mermikides 2010).
The shape of musical improvisation 191
A cadenza in musical space
With these provisos, and using a closely related approach, we can now turn
to an extant cadenza for Beethoven’s Violin Concerto Op. 61 by the Belgian
violinist Hubert Léonard (1819–90) first published in the 1880s (Léonard
c. 1883). Léonard was an established violinist who had studied with Henri
Vieuxtemps and François Habeneck at the Paris Conservatoire. Beginning
his tours of Europe in 1844, he succeeded Charles Auguste de Bériot as prin-
cipal professor of violin at the Bruxelles Conservatoire in 1847, becoming the
tutor of celebrated violinists of the next generation including Martin Pierre
Marsick, Henry Schradieck and Henri Marteau (Stowell 1992: 65). Léonard
is possibly best known today as an inspiration for Gabriel Fauré’s Violin
Sonata. In 1875, on a long visit to Sainte-Adresse, near Le Havre, Léonard
advised the young Fauré on how to make his composition ‘more playable
and effective’ (Nectoux 2004: 23). Léonard, then, was well respected in his
field, not on a par with extraordinary or esoteric violinists such as Eugène
Ysaÿe (in the following generation) or Niccolò Paganini (of the previous
generation), and better remembered, in the long run, as a pedagogue. It was
in this capacity that he provided an unusual second-violin accompaniment
for the Beethoven concerto, exclusively for teaching purposes, reissued in a
1909 arrangement for violin and piano edited by his student Henri Marteau
(Beethoven 1909).
To fill out some of the context in which Léonard’s cadenza was fashioned,
we should note briefly the types of values embodied in classical-era cadenza
improvisation/ composition as exemplified by Daniel Gottlob Türk ([1789]
1982: 301). Like the ‘limit and vary’ exercises illustrated above, Türk describes
some basic starting-points for cadenza creation, although his presentation is
rather more prescriptive than the examples offered from Goodrick (1987) and
Crook (1991).
• [T]he cadenza … should particularly reinforce the impression the
composition has made in a most lively way and present the most
important parts of the whole composition in the form of a brief
summary or in an extremely concise arrangement…
• The cadenza … must consist not so much of intentionally added
difficulties as of such thoughts which are scrupulously suited to the
main character of the composition…
• Cadenzas should not be too long…
• [M]odulations into other keys … either do not take place at all …
or they must be used with much insight … only in passing…
[O]riginally the harmony of the six-four chord and in any case the
triad that follows it were the basis of the cadenza, but in our time
these harmonic confines are probably too narrow. One can modulate;
only one should not remain in neighbouring keys so long that the
feeling for the main key is extinguished.
194 Music and Shape
Keeping Türk’s guidelines in mind, together with the basic principles outlined
in our discussion of jazz improvisation, we can now look at the relationship
of Léonard’s cadenza to the text of Beethoven’s Op. 61 following a chains-of-
thought model. The cadenza is rich in the standard techniques of cadenzas of
the period, principally motivic and melodic development and transposition.
Yet, looking at his traversal of musical space, we see Léonard playing with the
structural relationships expected from Beethoven’s composition, creating musi-
cal leaps in terms of formal proximity, as illustrated below.
An analogy is to be found in the concept of transformational grammar, as
drawn on already above. As Chomsky (1988) theorized, the ‘transformational’
properties of language arise from the language user accessing a body of know
ledge (‘lexicon’, which is the linguistic analogue to Pressing’s 1984 ‘knowledge
base’) to generate novel sentences, or sequences of words or phrases. The cadenza,
similarly, is a kind of real-time interaction between the source material and the
improviser/composer’s musical abilities. To use terms from Pressing (1984), the
‘referent’ is clear: the thematic, motivic, rhythmic shapes or any musical element
from the written text of the concerto, with a clear emphasis on the main thematic
material. The ‘knowledge base’ is derived from the virtuosity of the performer/
composer and thus dictates the execution—or ‘generation’ in Chomskian terms—
of the cadenza, a novel set of phrases and sequences based on the referent.
Thus, Léonard created a novel reconception of Beethoven’s given material,
passing elements of the original text through the filter of his individual musi-
cal personality and creative interests. He uses generative methods of musical
creation common to the period—variation, transposition, recombination and
‘rational deception’—to reenvision Beethoven’s material (Berkowitz 2010). He
also inserts shapes—conceived here as motives or characteristic passages of
material in the process of change—that seem to come directly from his know
ledge base, perhaps inherited from his composition/improvisation education,
The shape of musical improvisation 195
perhaps quoted from his own composed/improvised work, perhaps newly syn-
thesized with material from Beethoven, although it may be hard to discern an
exact origin. In some ways, perhaps this latter category is the most significant
outcome of the cadenza creation process: the fusion of the composer’s and
the performer’s musical identities through which shapes survive although their
obvious identifying features have changed.
Figure 6.11 represents the first section of Léonard’s cadenza, divided into
L1 and L2 according to their relationship to material from Beethoven’s original
text (the corresponding passages are B1 and B2). L1 illustrates the opening of
Léonard’s cadenza, which extends the first seventeen bars of Beethoven’s text
(B1) by one bar (circled). Structurally Léonard keeps the identical thematic
order, employing isologos (to continue with our earlier terminology). The types
of transformation in this section are harmonic, rhythmic and ornamental: the
first statement of the main theme (bars 1–9 of B1) is harmonized with two-,
three-and four-note chords, where the first half of the second section (bars 10–
13 of B1) is extended by one bar (bar 14 of L1) and embellished by arpeggiated
flourishes in A major and Amaj7 (bars 11 and 13). The statement is thus intensi-
fied (e.g. the opening forte dynamic marks an example of dispaesi), departing
from the sweet character of the original melody but keeping the original line
(isomelos). The descending melodic figure in bars 14–15 of B1 is broken into
chords and rhythmically varied while the final two bars, 16–17, are repeated
nearly identically to Beethoven’s text with the minor addition of a ‘g’ pick-up
for supporting harmony (bars 17–18 of L1).
L2 shows us a slightly more adventurous departure from Beethoven’s
motives. A reinvention of bars 32–34 of B2 starts the passage, keeping the
rhythmic gestures of the original section (isorhytmos). However, melodically
the phrase peaks, not troughs (bar 22 of L2), as Léonard reconfigures expecta-
tions (examples of dismelos and diskinetos). Further, he borrows the rhythmic
detail of what follows in B2 (bars 35–41) creating a transitional harmonic sec-
tion that leads into a very unexpected revision of B3, departing from the related
material in B1 and B2.
Illustrated in Figure 6.12, L3 begins with a rather unexpected harmonic and
ornamental reshaping of the melodic and harmonic material of B3, one of
the most ‘expressive’ moments of the concerto as reported widely by theorists
and embodied in performance across many recordings (Stowell 1998; Fabian
2006). This section acquires greater significance later in the cadenza. Léonard
reuses the highly recognizable opening suspensions, g2 to f ♯ 2 and c3 to b♭2, in the
original harmonic progression (B3, bars 332–35), but transforms the texture
and setting entirely with supporting d2 ostinato trills. Here Léonard is extract-
ing part of a musical shape (isokinetos) but shifting the conceptual outcome
dramatically and in terms of original affect (dislogos). The trills are then used
to modulate further towards a section that is so highly generic as transitional
material (bars 37–42) that it is hard to call it Beethovenian, or even Léonardian.
FIGURE 6.11 Opening section of Léonard’s cadenza (L1/L2) and corresponding sections from Beethoven’s Violin Concerto Op. 61 (B1/B2). L1 is subdivided into two
phrases, with their respective transformations connected by lines in the figure. The motivic elements of B2 are also then connected to their reshaped versions in L2.
FIGURE 6.12 Second section of Léonard’s cadenza (L3/L4A/L4B) and corresponding sections from Beethoven’s Violin Concerto Op. 61 (B3/B4). A harmonic relationship is
indicated by showing the use of the C–B♭ and G–F♯ suspensions in B3 as a motivic shape in L3. The modulating sequence of L3 is then separated by a different box, lower
in the figure. The B4 melody is boxed in the middle of the figure, with corresponding melodic embellishments indicated above and below in L4A and L4B.
198 Music and Shape
L4A brings a virtuosic reworking of the secondary theme (B4) that is highly
evocative of chromatic contrapuntal late eighteenth-or early nineteenth-cen-
tury virtuoso violin technique (embodied variously in the études of Pierre
Gavinies or the caprices of Niccolò Paganini). Fluctuating, like the original
section, from major to minor, this passage offers nothing inherently unusual.
What follows, however, is quite unexpected. As illustrated in Figure 6.13, L5
is an arpeggiated section that denies the expectation that the melodic material
in B4, bars 57–60, will be continued (dismelos). Instead, we hear a harmonic
progression with a pronounced bass melody line, indicated by accent marks in
the score. The melodic line seems familiar yet is difficult to recognize, especially
given the denial of the expected continuation of B4’s melodic material. This
is, in our opinion, the most profound moment of musical surprise or ‘rational
deception’, as theorized by C. P. E. Bach ([1753, 1762] 1949).
Possible sources for this passage can be found among Beethoven’s harmonic
suspensions in the orchestral parts: the inner string writing of bars 51–63
(Beethoven 1968, not depicted here); the bassoon and horn parts in relation to
the principal violin part, bars 304–29 (not depicted here); and, most convinc-
ingly, bars 351–57 (shown in Figure 6.13: B5), where we touch on the A♭–B♭
and C tonalities evident in Léonard’s cadenza. Note particularly the melodic
suspensions in the violin part and the almost Khachaturian-like semitone har-
monic modulations in bars 348–69.1
Yet even this suggestion of derivation is dubious. More likely, Léonard’s
material is the generative outcome of synthesized information, the improviser/
composer mixing the referent and his own knowledge base so thoroughly that
the listener can no longer be sure about the origin of the material. Therefore,
instead of an etymological pursuit through analysis, it seems wiser to propose
that our auditory perception is rationally deceived, both through the harmonic
language and in supposing a specific source for the material: it is impossible
to tell whether we are listening to Beethoven or Léonard, and perhaps that is
the point. What matters is not the amount of influence that is present, nor the
success of Léonard’s transition—which might even be seen as sloppy, depend-
ing on one’s sense of compositional values. The important point is that we
have reached something mystical, inexplicable, the joining of two minds who
lacked the opportunity to meet in person but who meet—in a metaphysical
sense—through sound and transfer of text. The synthesis of these two streams
of consciousness, which crystallizes throughout the cadenza, is made particu-
larly evident in this transitional passage.
Returning to Figure 6.12, let us note that the section succeeding this transi-
tion, shown as L4B, revisits the melodic content of B4 in an even more robust
virtuosic texture, now in the original key areas of D major and D minor.
Léonard provides a resolute confirmation of our expectations in the continu
ation of the melody previously denied (isomelos), embellishing the melodic con-
tent in a homogeneous fashion through to the melodic material of bars 57–60
FIGURE 6.13 Final section of Léonard’s cadenza (L5/L6A/L6B) and corresponding sections from Beethoven’s Violin Concerto Op. 61 (B5/B6). Both L5 and B5 indicate progressions that
embody pronounced melodic properties, using secondary dominant and common-tone modulations, semitone movement in the bass, and consistent fluctuations between major and
minor harmonies. The relationship is considered tenuous. The shapes material from B6 is indicated, the opening section corresponds to L6A, while the last sequence of notes in B6,
circling around ‘a’, corresponds to an extended retransition sequence in L6B.
200 Music and Shape
Cadenza shape
L3
Musical proximity to referenced material
L4
L2
L1
L6
textural
shift
isorytmos isokinetos
dislogos
Transformations
dismelos
harmonic diskinetos disrhytmos
embellishment harmonic tenuous
isomelos embellishment harmonic
isokinetos isokinetos relationship
dispaesi isomelos dislogos
isokinetos disrhytmos
(extended) dispaesi
B1 B2 B4 B6 B3 B5
1 17 28 43 64 89 96 331 357
Section bar length and relative position of
referenced Beethoven material
FIGURE 6.14 Graphic representation of Léonard’s cadenza illustrating the relationship of musical
proximity to Beethoven’s original score. The closer the vertical distance between each corresponding
B and L section, the greater the musical proximity of corresponding material. Bar lengths for each
section correspond to horizontal width. A ‘field series’ organization is indicated by the arrows
connecting each L section, while transformations are briefly indicated in the rounded boxes
connecting each B and L section.
The shape of musical improvisation 201
Conclusion
This chapter has approached the concept of shape in improvisation with the
use of four conceptual stepping-stones, subdivided into four sections.
1. Using jazz improvisation (and its pedagogy) as a reference point,
an improvisation might be seen as a chain (or series of chains)
whereby newly created musical objects hold an appreciable musical
relationship (a general motivic similarity) to a previously established
musical object (or set of objects). The use of the word ‘chain’
suggests a strict linear set of relationships, but this model allows an
intricate nonlinear set of connections, whereby relationships might be
apparent among a wide set of parameters, and relationships between
objects might ‘leap-frog’ over interpolated objects.
2. The links in this chain were more keenly examined, and it was
suggested (with a basis in jazz improvisational pedagogy) that they
can be usefully described in terms of which musical parameters are
pertinently fixed and which are varied. Terms like distimbral, dismelos
and isoplacement emerge quite naturally alongside well-established
concepts such as isorhythm and displacement. This ‘limit and vary’
concept of improvisation allows a way to identify improvisational
mechanisms beyond the ‘surface’ vocabulary employed, as well as
having direct practical applications.
3. Since a series of musical objects in an improvisation might be
identified as having varying degrees of similarity, one might
imagine a multidimensional space of musical proximity (M-Space)
surrounding any musical object with closer objects being more
recognizably similar and further objects being more musically distant
until they have little or no recognizable similarity. Since this musical
space occupies many musical dimensions, it follows that musical
202 Music and Shape
References
Bach, C. P. E., [1753, 1762] 1949: Essay on the True Art of Playing Keyboard Instruments,
trans. W. Mitchell (New York: Norton).
Bäckman, K. and P. Dahlstedt, 2008: ‘A generative representation for the evolution of jazz
solos’, Lecture Notes in Computer Science 4974: 371–80.
Balliett, W., 1959: The Sound of Surprise (New York: Dutton).
Beethoven, L. v, 1968: Violin Concerto in D major Op. 61 (New York: Kalmus).
The shape of musical improvisation 203
Monson, I. T., 1996: Saying Something: Jazz Improvisation and Interaction (Chicago:
University of Chicago Press).
Nachmanovitch, S., 1990: Free Play: Improvisation in Life and Art (New York: Tarcher/
Penguin).
Nectoux, J., 2004: Gabriel Fauré: A Musical Life (Cambridge: Cambridge University Press).
Nooshin, L., 2003: ‘Improvisation as other: creativity, knowledge and power—the case of
Iranian classical music’, Journal of the Royal Musical Association 128/2: 242–96.
Patel, A. D., 2003: ‘Language, music, syntax and the brain’, Nature Neuroscience 6: 674–81.
Persichetti, V., 1961: Twentieth-Century Harmony: Creative Aspects and Practice (New
York and London: W. W. Norton).
Pressing, J., 1984: ‘Cognitive processes in improvisation’, in W. R. Crozier and A. J. Chapman,
eds., Cognitive Processes in the Perception of Art (Amsterdam: Elsevier), pp. 345–67.
Pressing, J., 1988: ‘Improvisation: methods and models’, in J. A. Sloboda, ed., Generative
Processes in Music: The Psychology of Performance, Improvisation, and Composition
(Oxford: Oxford University Press), pp. 129–78.
Roads, C., 2004: Microsound (Cambridge, MA: MIT Press).
Sawyer, K., 1992: ‘Improvisational creativity: an analysis of jazz performance’, Creativity
Research Journal 5/3: 253–63.
Schillinger, J., 1978: The Schillinger System of Musical Composition (New York: Da Capo Press).
Solstad, S. H., 1991: ‘Jazz improvisation as information processing’ (MPhil thesis,
University of Trondheim).
Stowell, R., 1992: The Cambridge Companion to the Violin (Cambridge: Cambridge University
Press).
Stowell, R., 1998: Beethoven: Violin Concerto (Cambridge: Cambridge University Press).
Türk, D. G., [1789] 1982: School of Clavier Playing or Instructions in Playing the Clavier
for Teachers and Students, trans. R. H. Haggh (Lincoln: University of Nebraska Press).
Werner, K., 1996: Effortless Mastery: Liberating the Master Musician Within (New Albany,
IN: Jamey Aebersold Jazz, Inc.).
Wishart, T., 1996: On Sonic Art, ed. S. Emmerson (Amsterdam: Harwood Academic).
Xenakis, I. and S. Kanach, 2001: Formalized Music: Thought and Mathematics in
Composition (New York: Pendragon).
Discography
Shapes performed
Reflection
Max Baillie, violinist
3D Bach . . . and the harmonic comet
The harmonic comet
Picture an anchor embedded into the ground. A comet ignites and takes off
from the anchor up into the air and in its wake leaves an expansive arc which it
completes into a circle when the comet returns to its starting point. In continu-
ous motion, it travels up and into a second revolution. On a third rise from the
anchor the comet seamlessly curves out into a new trajectory creating a second
circle suspended in the air. As you watch, the comet builds an entire structure of
suspended rings in the sky, exploring the unknown space and also returning to
familiar orbits. Eventually, its travels take it back to the anchor where it began.
This image is a metaphor for the journey the ear might travel while hearing
a piece of music—specifically, music that both is tonal and modulates from
one key to another. The comet is its real-time flow, the anchor is the home key,
and the circle that emanates from it describes a short harmonic journey from
stasis towards tension and back. The points at which the comet breaks into a
new orbit are the pivot chords, and if a feeling of movement comes from the
changes between chords within one key then it is the pivots that create the
sense of travel: rather than occupying the same space through a modulation,
I imagine the music as travelling from one space to another. The character of
this travel (which may be anything from a sublime cruise to a frantic search or
even a joyous ramble depending on the piece) is down to everything else in the
score: the metre, tempo, voice-leading, bowing, and so on, but the movement
itself comes from travel between harmonic orbits.
But where do we start in our flat forest of notes (Figure R.13a)? Unlike, for
example, a classical portrait whose structure is clearly assumed, the overwhelm-
ing visual impression in the figures here is of something uniform. At a glance
the image flipped over looks more or less the same (Figure R.13b). Where are
the structural shapes?
Reflection: Max Baillie 209
(a)
(b)
FIGURE R.13 The opening of the Allemande from J. S. Bach’s Partita in D minor for solo violin: (R.13a)
as usual and (R.13b) upside down
Harmonic cartography
For both performer and teacher, playing Bach requires some detective work.
I sometimes find it helpful to think of the solo repertoire as a distillation (as
distinct from a reduction). Clearly, it’s not that there’s anything missing, but
there is an invisible hierarchy; the notes don’t all occupy the same function in
the harmonic scaffold. If we dig this structure out (Figure R.14), the shape of
the harmonic journey begins to emerge from a more or less uniform trail of
notes on the staff.
FIGURE R.14 Allemande, bars 1–8, with a harmonic analysis of tonal centres and harmonic rhythm
Reflection: Max Baillie 211
FIGURE R.15 The passage in R.14 represented as a physical journey through space between related
tonal orbits
FIGURE R.16 Allegro assai, bars 1–8, from J. S. Bach’s Sonata in C major for unaccompanied violin
not only in figuration but in Bach’s original phrase markings, as are bars 7 and 8.
From the beginning the harmonic rhythm swings boisterously from tonic to
dominant in each bar, and yet despite the visible differentiation between bars 5
and 6 we are going up a gear on our rustic C major ride: the harmonic rhythm
halves. Bars 5 and 6 belong very much together by virtue of sailing across one
harmony, and the same is true of bars 7 and 8.
The shape the listener receives, if the performer emphasizes this harmonic
rhythm (Figure R.17), is totally different than if either the figurations or the
visual impression of each bar as a separate entity guides the performer. It’s
unquestionably more convincing to my ears: it’s as though the music has gone
up a gear; the trajectory is the same but the arc is bigger. The listener hears
the lower timescale expand while the upper, the flow of semiquavers, remains
constant, and it’s as though two timescales of music are bound together: magic!
But there are other layers too. To use another metaphor, if the harmony is
the skeleton, what of the flesh and blood? The next stage in building up our 3D
image is to look at how the harmonic layer interacts with the melodic rhythm,
and I’ve found this equally illuminating.
Summary
These are examples of what is in the music but not spelled out by the
score: shapes embedded in the text but not immediately visible. To my ears they
FIGURE R.17 The passage in R.16 showing the harmonic rhythm
FIGURE R.18 The Allegro assai, bars 13–16, showing melodic rhythm
Reflection: Max Baillie 215
play an essential role in making sense of what Bach intended: we must remem-
ber that, although an accomplished violinist, he wrote from the keyboard, with
all its richness of harmony, counterpoint and voice-leading expressed through
the medium of an essentially melodic instrument in these works for solo violin.
As a listener I feel I want to be led on the deeper path, the harmonic journey,
while enjoying these layers in dialogue; and that is also the way I aspire to bring
them to life as a violinist. The result is often that the phrasing becomes clearer
and also simpler: the performer makes longer lines where passages belong
together harmonically, and the music gains its multilayered quality where
melodic and harmonic rhythms interplay. Then there is also the whole world of
melodic contrapuntal writing (as opposed to counterpoint between harmonic
and melodic layers) embedded in Bach’s solo violin music; here the challenge
(and joy) as a performer is in the sense of spinning these as dialogue while also
playing one line of music with a single coherence.
Being sensitive to, and inquisitive of, the depth of the text should be a nat-
ural ingredient in a loving realization of this music. Playing with compelling
sound and presence is not enough: if we don’t dig below the surface, and if
we don’t detach ourselves imaginatively from the staff, our performance will
ultimately be one-dimensional because it will miss out the embedded propor-
tions in the music. The idea of a harmonic comet is one possible metaphor that
enables a player to assign an imagined shape to these proportions and embraces
the idea of the ear as a harmonic traveller. It suggests music as an agent with
a will to explore and with a physical form independent of its undifferentiated
representation on the page. As a way spatially to conceive harmonic patterns
it is an imaginative tool, and when combined with the melodic and rhythmic
layers around this harmonic framework it allows us to bring Bach’s music to
life in 3D.
7
This chapter presents findings from a study of performing musicians and focuses
on some of their practices and beliefs related to musical shaping. Musical per-
formance has been studied in myriad ways, and with a wide range of aims
in mind (for a useful overview, see Gabrielsson 2003). Preparation for perfor-
mance (especially memorized performance) has been examined in considerable
detail, with researchers finding expert practice to be a highly structured activity
in which performers focus on three dimensions of a composition: the basic
dimension, which includes all aspects of the music requiring attention simply
to play the notes of the piece, and which therefore includes technical decisions;
the interpretative dimension, involving decisions about phrasing, dynamics
and tempo; and the performance dimension, which involves every aspect of
the piece that requires attention during performance, including basic, interpre-
tative and expressive performance cues (Chaffin, Imreh and Crawford 2002;
Chaffin et al. 2010). Experts often work on small sections of a piece of music,
determined by the musical structure, before joining these chunks together to
create larger sections as the piece becomes more familiar (Chaffin et al. 2002).
Decisions involved in musical performance preparation have also been exam-
ined, with three main types of performance decision being identified: intuitive,
deliberate and procedural, procedural being previously deliberate decisions
that have become intuitive over time (Bangert, Fabian et al. 2014). Bangert,
Schubert and Fabian (2014) propose a spiral model of musical decision-
making, in which a musician’s decisions switch from being intuitive to deliber-
ate and from there become procedural: the proportion of intuitive decisions
thus increases with expertise. As a performer focuses on new musical features,
the cycle is repeated.
Many of the decisions made by performers concern expressive performance,
the teaching and nature of which has been examined extensively (Brenner and
216
Shape as understood by performing musicians 217
Strand 2013; Davis 2009; Fabian, Timmers and Schubert 2014; Juslin 2003;
Juslin, Friberg and Bresin 2002; Juslin, Friberg and Schoonderwaldt 2004;
Juslin and Madison 1999; Karlsson and Juslin 2008). Particularly useful is the
GERMS model (Juslin 2003), which identifies five essential components of
musical expression. These are: Generative rules, which serve to clarify the musi-
cal structure through timing, dynamics and articulation; Emotional expression,
in which a range of parameters is used by a performer to convey an intended
emotional expression; Random variations, which are unavoidable and essential
for a performance to sound as though it is produced by a human being; Motion
principles, which incorporate the representation of intended and non-intended
biological motion in sound; and Stylistic unexpectedness, which involves the
creation of tension through the violation of expectations. Though these com-
ponents may not all be considered consciously by performers in their decision-
making, this division does provide some understanding of what performers are
doing in order to create an expressive performance.
Some studies examine the use of particular types of language, such as
metaphors, in relation to music performance preparation (Barten 1998;
Woody 2002), but few studies examine the use and meaning of only one word.
Usually, such an exercise would be rather futile, as much of the terminology
employed by musicians has a reasonably well-established definition. Shape,
or shaping, however, appears to have resisted formal definition in relation to
music, and yet seems to be a useful term for performers, as well as for other
musicians. A recent questionnaire study (Prior 2012c) revealed that perform-
ers use the notion of shaping when practising, in rehearsals, when teaching
and when playing music from a wide range of genres. The term was used in
relation to several ideas, from musical structure to musical expression, emo-
tion and tension; and in relation to specific musical features such as phrasing,
melodic line and dynamics. Overall, shape was found to be highly versatile
and multifaceted. This was an interesting finding in itself, but there was no
way in which an in-depth understanding of shaping could be gained through
these data, gathered as they were in an online questionnaire. A subsequent
interview study allowed greater interaction with a small number of partici-
pants and allowed the development of a model of the ways in which musical
shaping may be used by performers and understood by those studying them.
This study, and the model arising from those data, are presented and refined
within this chapter.
Aim and method
The aim of the interview study was to understand how performing musicians
use the idea of musical shape or shaping.
218 Music and Shape
PARTICIPANTS
Ten professional musicians were interviewed; five were violinists and the other
five harpsichordists. The choice of instruments was carefully considered, in
terms of both the researcher’s background knowledge and experience as a
musician and the potential this gave for insight into the techniques discussed by
the musicians, and also in terms of the instruments’ different capabilities, which
seemed likely to prompt interesting variations in the musicians’ conceptions
of musical shaping. Specifically, the differences between the instruments’ abil-
ity to sustain a sound, to produce sounds with a varied dynamic range and to
play chords were noted. A further difference between the instrumentalists was
the violinists’ close knowledge of their own instrument, in contrast with the
harpsichordists’ unfamiliarity with the harpsichord used in the study, a double
manual by Michael Johnson.
Details of the participants can be seen in Table 7.1. They ranged from eigh-
teen to fifty-four years of age, and their experience playing their instrument
ranged from less than ten years to forty years. They were all resident in the
UK, though some participants were originally from Australia, South America,
Ireland and Japan. Many of them had studied performance at universities and
conservatoires, often to postgraduate level. They were all established profes-
sional performers, the majority of their earnings coming from performance,
though some of them also taught or had research interests.
THE INTERVIEWER
PROCEDURE
brought with them or knew well. The schedule contents and order were flexible
to ensure that the interviews felt natural and comfortable for the participants.
At the end of the interview, participants signed the consent form and were
compensated for their time. The interviews were recorded using a Panasonic
SD700 HD Camcorder and a Sony ICD-UX200 Digital Voice Recorder.
DATA ANALYSIS
The interviews generated verbal, musical and gestural data, all of which were
analysed to some extent (Prior 2012a). This chapter focuses mainly on the ver-
bal data, with reference to some of the musical data. The verbal data were ana-
lysed with Interpretative Phenomenological Analysis (IPA). This approach has
been widely used in health psychology, but has also been found to work well in
research in music psychology (McPherson, Davidson and Faulkner 2012: 92) as
it allows participants’ thoughts and experiences to be examined idiographi-
cally and in detail. In particular, IPA is appropriate for situations in which
researchers are conducting exploratory studies investigating how individuals
are making sense of their personal and social world and the processes within
that world (Smith and Osborn 2003). The use of IPA was particularly appro-
priate here because of the complex and potentially idiosyncratic ways in which
expert musicians perceive and understand their work, as well as the potential
for emotional involvement in their practices. What constitutes a ‘good’ musical
performance is, in part, socially constructed, determined not only by techni-
cal expertise but by the tastes of both the individual and the period in which
they are performing (Leech-Wilkinson 2009). It therefore seems appropriate to
examine the processes of musical shaping with a method such as IPA that was
developed within the framework of social constructionism.
Data analysis proceeded according to the guidelines for IPA provided by its
pioneers (Smith, Flowers and Larkin 2009; Smith and Osborn 2003). Following
each interview, the recording was listened to in its entirety and initial notes
were made. The data were then transcribed verbatim, but the recording was
used alongside the text throughout the coding process. Initial coding focused
on a phenomenological approach to the data, identifying the main concerns of
each participant and the meaning these concerns had for them. A second stage
of coding followed with an interpretative approach which attempted to iden-
tify how and why the participant had those concerns and to link the phenom-
enological codes to more abstract ideas. The coding was validated by another
member of the research team. Themes were generated from the coded data,
and a summary was written for each participant in relation to each theme. A
summary diagram was also created for each participant. Each interview was
analysed completely before moving on to the next participant’s data.
During the interviews, participants frequently demonstrated their thoughts
about musical shaping on their instrument or by singing. These data were seen
Shape as understood by performing musicians 221
Although the data gathered provided scope for the consideration of musi-
cal shaping in considerable detail, within this chapter a broad view is taken,
with the aim of creating a data-led model of the use of musical shaping by
musicians. The model (which also acts as a summary of the data) is shown
in Figure 7.1 and is available as well on the companion website , complete
with tables showing examples of each component. On the far left of the model
is the concept or idea of a musical level that can be controlled (or for some
participants, ‘shaped’). Next to this is a column of musical triggers for shap-
ing: features of the music that participants identified as influencing their
shaping decisions. On the far right of the model is the change in sound that
results from the musical levels being controlled or shaped in performance.
These three columns are arranged in approximate size o rder, with the larg-
est features at the top and the smallest at the bottom. One of the remaining
two columns in the model outlines the technical modifications that are used
to create this changed or shaped sound on the two instruments studied. The
separation of the two instruments within this column allows for the fact that
each instrument has limitations that restrict a performer’s ability to control the
changes in sound represented in the final column. Although these technical
approaches could be the participants’ main focus of attention, many partici-
pants appeared to ‘skip over’ these detailed decision-making processes, using
more or less metaphorical ideas like shape heuristically to help them to create
a musically expressive performance, a notion that is represented by the central
column in the model. Because many of the heuristics seemed to be applicable
at multiple levels, they are arranged not in size order, but alphabetically. Each
222 Music and Shape
FIGURE 7.1
Model of musical shaping. In the online version, each component is numbered, and
numbered examples of each component are presented in linked tables. See the companion website:
The first column of the model focuses on the musical levels discussed by par-
ticipants in relation to musical shaping. It became apparent through the inter-
views that shape was a very flexible term in many ways, not least in the scale at
which it could be applied. Table 7.2 shows the participants who discussed using
shaping at each level. Examples from all participants may be found on the com-
panion website, and some of these are discussed later in the chapter; here, the
focus remains on a few specific quotations from participants discussing shape
at multiple scales.
Several participants discussed the use of the term ‘shape’ at more than one
level. Elsie commented that ‘Every note should have some kind of shape. And
every phrase needs to have a shape’,2 and she also explained that her understand-
ing of the large-scale shape of the music affected the ways in which she shaped at
Shape as understood by performing musicians 223
TABLE 7.2 Participants who discussed each musical level (see Table 7.1 for their names, repre-
sented here by initials)
B D E T V Total Ja Ju K N Y Total
Concert 0 ✓ ✓ ✓ 3 3
Whole piece ✓ ✓ ✓ 3 ✓ ✓ ✓ ✓ 4 7
Movement ✓ ✓ ✓ ✓ 4 ✓ ✓ ✓ ✓ 4 8
Section ✓ ✓ ✓ ✓ 4 ✓ ✓ ✓ ✓ 4 8
Phrase ✓ ✓ ✓ ✓ ✓ 5 ✓ ✓ ✓ ✓ ✓ 5 10
Note ✓ ✓ ✓ 3 ✓ ✓ ✓ ✓ ✓ 5 8
smaller levels.3 Victor, too, saw ‘shaping’ as a flexible term that could apply to
several levels of the music. When asked to define shape, Victor used metaphors
of language and narrative:
In contrast, Julian used more technical language to describe the slight variation
in meaning that he felt occurred with the use of shape in different contexts:
JULIAN: I suppose if someone said to me . . . ‘What shape does the
music have to you?’ I’d think instinctively they were talking about the
structure. . . So structure and shape sort of overlap in that capacity. If
you’re talking about a phrase, and you said the shape, I’d be thinking
about, as a player, the sort of technical way you might play it, in
terms of grouping of notes, and the articulations . . . what degrees of
staccato or legato do we want . . . in a particular given phrase. But . . .
with baroque music, one note can have shape, a messa di voce, so you
can just, you know, if you were talking to a violinist or particularly
a singer, and you said ‘What shape does that note have?’ you might
immediately think of the swelling and diminuendo of one note.’5
For Darragh, however, the term ‘shape’ applied specifically to the phrasing
level, with other words being more appropriate for larger or smaller levels of
shaping that were discussed by other participants:
These ideas could be seen to operate on a spectrum of specificity and scale, with
Elsie and Victor at one extreme, using the term ‘shaping’ flexibly at all levels,
Julian in a more central position acknowledging the slight variation of mean-
ing in the word between small and large scales, and Darragh at the opposite
extreme, reserving the idea for the phrasing level and using other terminology
for variations in sound at other levels. Specific examples of shaping at each level
are discussed later in the chapter.
The second column within the model shows the score-based triggers identified
by participants as influencing their shape-related decision-making. Table 7.3
shows the participants who reported using each idea. In the model and in the
table, the triggers are shown in the order that relates approximately to musi-
cal scale, with large-scale ideas at the top and small-scale ideas at the bottom.
Some of the titles of these ideas may seem self-explanatory; however, others
are more complicated, and therefore the categories are discussed briefly and in
order, with a few examples. Full examples are provided online.
View of the score
This particular musical trigger was usually an overarching philosophical stance
adopted by the musicians relating to how they felt they should use the informa-
tion provided on the score by the composer or the editor. Participants discussed
the idea of ‘shaping as being anything that you’re doing to get the music off the
page, and to the listener’,7 with the score providing clues as to how this might be
achieved.8 Other participants discussed the score as their only tangible connec-
tion with a composer and that composer’s intentions, with Elsie commenting,
‘it’s just you and the composer again’,9 and Victor describing the score as a code
that he has to interpret.10
Shape as understood by performing musicians 225
B D E T V Total Ja Ju K N Y Total
View of the ✓ ✓ ✓ ✓ 4 ✓ ✓ 2 6
score
Musical ✓ ✓ ✓ ✓ 4 ✓ ✓ ✓ 3 7
structure
Words on the ✓ ✓ 2 ✓ 1 3
score
Harmony ✓ ✓ ✓ ✓ ✓ 5 ✓ ✓ ✓ ✓ ✓ 5 10
Polyphony ✓ ✓ ✓ 3 ✓ ✓ ✓ ✓ ✓ 5 8
Melodic ✓ ✓ ✓ ✓ ✓ 5 ✓ ✓ ✓ ✓ ✓ 5 10
contour
Rhythm ✓ ✓ ✓ 3 ✓ ✓ ✓ ✓ ✓ 5 8
Patterns ✓ 1 ✓ ✓ 2 3
Dynamic ✓ ✓ ✓ 3 0 3
markings
Articulation ✓ ✓ ✓ ✓ 4 ✓ ✓ ✓ 3 7
or phrase
markings
Musical structure
Some of the interviewees discussed musical structure as something that had an
influence on their musical shaping, with Elsie commenting that she is always
aware of her position within the musical structure as she plays11 and confirm-
ing that this influences her shaping on a smaller scale.12 Others described the
ways they would highlight structural boundaries13 or create a sense of structure
through their playing.14 Darragh discussed structure as something his fellow per-
formers frequently liked to be aware of before making interpretative decisions.15
Words on the score
Performance directions,16 words provided by the composer to convey a pro-
gramme or appropriate imagery for a piece,17 and the lyrics of a vocal piece18
were all reported to have a direct bearing on the musical shaping used by the
performers.
Harmony
Harmony was one of only two triggers to be discussed by all participants,
though not all of them felt comfortable in using this trigger themselves, with
Bridget suggesting that she found other methods more intuitive.19 All four of
the other violinists, however, described how harmony could influence their
shaping decisions, with Victor arguing that much of the expressiveness of a
226 Music and Shape
Polyphony
Participants discussed musical parts played by others in ensembles influencing
their musical shaping,24 as well as their awareness of ‘voices’ within their own
parts, and of the shaping decisions they made to try to highlight those voices
for their listeners.25
Melodic contour
Like harmony, melodic contour was discussed by all participants. Many dis-
cussed mirroring melodic contours with dynamics26 or highlighting the top of a
phrase through timing.27 Others discussed descending melodic lines or tessitura
more generally.28
Rhythm
Participants discussed the ‘shape of the rhythm’,29 the hierarchical relationships
between beats in a bar,30 and the link between those relationships and bowing
patterns.31 Others discussed the appropriate grouping of particular rhythmic pat-
terns32 and how the shaping of a phrase related to its rhythmic (and other) constit-
uents.33 Tina discussed decisions relating to the musical shaping of syncopation.34
Patterns
Victor, Jane and Katharine all discussed patterns (such as harmonic sequences)
in the music that influenced their shaping decisions.35
Dynamic markings
None of the harpsichordists discussed dynamic markings, probably because
there were none present in their scores. Bridget, Tina and Victor discussed
dynamic markings as a trigger for their musical shaping, though Victor sug-
gested that he did not feel he needed to think consciously about applying them.
Rather, he suggested, ‘I think dynamics fall into place’.36
detail, specifically noting the phrasing indicated by the composer and what this
meant for him as a performer.37
TECHNICAL MODIFICATIONS
Violinists Harpsichordists
Some of the metaphorical ideas concerning the music and the appropriate
musical shaping seemed to be expressed through gesture, exposing participants’
multimodal understanding of musical shaping. Participants often discussed
ideas of direction, movement and gesture when talking about their musical
shaping; and, while they did so, they often used gestures in conjunction with
their descriptions or demonstrations of the music. Participants used height to
represent pitch, and vertical gestures to indicate rhythmic features. Arch-shapes
were used to indicate the shape of a phrase, and larger arches, wave patterns or
circular gestures to indicate the shape of an overall piece (Prior 2012a, 2012b).
Further analysis is intended to investigate whether or not there are specific dif-
ferences between the gestures used by violinists and harpsichordists, as well
as correspondences between the gestures used by participants, their verbal
descriptions and their musical demonstrations.
Specific examples of heuristics used by participants are discussed in more
detail later in the chapter; however, Table 7.5 shows the use of a range of heuris-
tic terms in the interviews and their distribution among participants. Because of
their holistic and nonspecific nature, the terms are not listed in order of size, as
other components of the model have been. Instead, they are listed alphabetically.
CHANGES IN SOUND
B D E T V Total Ja Ju K N Y Total
Audience ✓ ✓ ✓ ✓ ✓ 5 ✓ ✓ ✓ 3 8
Breathing ✓ ✓ ✓ 3 ✓ ✓ ✓ 3 6
Composer ✓ ✓ 2 ✓ ✓ ✓ 3 5
Direction ✓ ✓ ✓ ✓ ✓ 5 ✓ ✓ ✓ ✓ ✓ 5 10
Emotions ✓ ✓ ✓ 3 ✓ ✓ ✓ ✓ 4 7
Gesture ✓ ✓ ✓ 3 ✓ ✓ ✓ ✓ 4 7
Imagery ✓ ✓ 1 ✓ ✓ ✓ 3 4
Importance ✓ ✓ ✓ 3 ✓ ✓ ✓ ✓ 4 7
Instrument ✓ 1 ✓ ✓ ✓ ✓ ✓ 5 6
Line ✓ ✓ ✓ 3 ✓ ✓ ✓ 3 5
Natural ✓ ✓ ✓ ✓ 4 ✓ ✓ ✓ ✓ 4 8
Shape ✓ ✓ ✓ ✓ ✓ 5 ✓ ✓ ✓ ✓ ✓ 5 10
Singing ✓ ✓ 2 ✓ ✓ ✓ 3 5
Style ✓ ✓ ✓ ✓ ✓ 5 ✓ ✓ ✓ 3 8
230 Music and Shape
Each column of this model is affected by some overall situational factors, such
as whether the performance decisions are made in private practice, in rehearsal
or in performance; or whether the music involves other performers who influ-
ence the shaping decisions made. Several participants noted the value of per-
formance for generating ideas about musical shaping, an idea supported by
some existing research (Doğantan-Dack 2013). Yoshi suggested that she was
‘more alert’ during performance than when practising, which enabled her to
notice new features of the music and to have new ideas concerning the shaping
of those features. If Yoshi considers those ideas to be ‘too risky’ she ‘saves them
for later’, but there are times when she tries new ideas during a performance.46
Both Tina and Katharine valued spontaneity in their ensemble performances,
and noted how other performers would influence their own shaping during a
concert. Tina discussed the ‘communal’ and ‘spontaneous’ shaping of a Haydn
string quartet, stating that ‘the ideal is that at any point, really, one person
might help to guide it in a particular way, so that . . . the contour of the piece
changes’.47 Similarly, Katharine reported that she particularly enjoyed perform-
ing with ‘someone who takes a few risks, and does something spontaneously
that you can then react to: . . . that’s really nice music-making’.48 Hence, it is
anticipated that the situation in which musical shaping is considered will influ-
ence the extent to which each aspect of the model is used.
The value of the model described above lies not only in its ability to outline
various aspects of shaping as discussed by the participants in the study, but also
its potential for showing the combinations of factors used by participants in
specific situations. With this in mind, specific quotations and musical examples
from the interviews are discussed alongside presentations of the model that
highlight which components are active in each situation. Examples showing
shaping at each musical level are offered.
She continued, playing the music in two ways, and describing her thoughts
about what she was doing:
ELSIE: I’ll play it in two different ways. If . . . I was playing this, within
the context of a larger piece of music, and I played it sort of, um
232 Music and Shape
[plays]. That might sound, sort of OK, but I’m still so involved with
it, do you know what I mean? Um, why not just let it go? [sighs] and
give the audience a chance to go, ‘Oh, that’s really nice’ you know, in
between having been gripped for the first thing, you know, so I could
just [plays] and just [plays]. It could give something, just completely
different. And it’s all to do with where the music lies within the whole
thing.52
These sound examples are available on the companion website, and it is pos-
sible to examine them for differences between the two versions. Using Sonic
Visualiser (Cannam et al. 2010), we identified the main beats of the excerpts
and exported the data for statistical analysis (Table 7.6). The two versions
differed in tempo: the ‘involved’ version ( ) had a shorter mean beat length
and was therefore faster than the ‘letting go’ version ( ). In an interview sit-
uation, the significance of this is difficult to assess; however, the variance of
the beat length also differed, with the ‘involved’ version having a significantly
larger variance than the ‘letting go’ version (Levene’s test of homogeneity of
variance: F (1, 26) = 13.9, p = 0.001). The two versions also differed in Elsie’s
use of dynamics. Although the mean power of each excerpt cannot be judged
reliably from this interview source, the variance of the power showed a con-
siderable difference between the two versions, with the ‘involved’ version hav-
ing significantly greater variance than the ‘letting go’ version (Levene’s test
of homogeneity of variance: F (1, 1739) = 15.9, p < 0.001). Some of this was
achieved by using less bow pressure, though Elsie did not specify any other
technical modifications. When listening, one can hear a slight difference in
the vibrato used in each version, with the ‘involved’ version seeming to have
a slightly faster vibrato that begins more promptly after the start of the note
than the ‘letting go’ version.
When representing the whole of this quotation with the model, we can see
that Elsie is considering the shaping of a piece and movement as a whole, and
that she is considering the musical structure of the whole work as a trigger for
her musical shaping. She is using the heuristics of ‘audience’, ‘emotions’ and
Involved Letting Go
JANE: But I’m thinking of . . . not just the shape that’s up and down, . . .
I was thinking of shapes that swell. Again, it’s my three-dimensional
thing, something that swells out, like a kind of serpent with
swellings in its body! [laughs] . . . So it not just a slippery snake that
goes like that, it’s something that kind of opens out and expands. . .
RESEARCHER: Can you tell me where?
JANE: Where it is, I suppose again, it would come to the harmonic thing
[plays]. That’s a sort of [plays], that’s a ‘here I am’ [plays], a sort
of visible [plays]. That to me is where he’s swelling out . . . puffing
himself up, but still he’s got energy to carry on [plays]. Now that
could be either [plays]; that could be just going away to nothing so,
I suppose the shape of that, thin shape, fat bulbous shape, starting
fairly bulbously, getting thinner, more bulbous as it comes down
again, and then going off to, just disappearing off. . . Which is . . . the
way the harpsichord works; you could do it completely the opposite
on the piano, because of . . . the dynamics, so in a way, the lack of
dynamics, . . . means that you have to follow, what the instrument’s
telling you. . . While on the piano, I could play that [sings] at the end,
but on the harpsichord I can do the [plays]. Some holding, but, could
do that I suppose [plays]. . . And the fact that he’s put er, lines over
each one, shows a kind of gestural [plays], gestural shape [plays].
Slightly rounded at the end there.55
When represented on the model, this quotation highlights the phrase level,
triggers of harmony and melodic contour, heuristics of gesture, imagery,
instrument and shape, technical modifications relating to over-holding, and
timing as a change in sound (see Figure 7.6, available online ).
Victor discussed shaping a phrase in slightly more prosaic terms, though
he too was frequently emotionally invested in the music he was discussing
and playing. He noted the musical triggers of harmony, melodic contour
and rhythm, using the heuristic of the audience (listener) and the metaphor-
ical imagery of communication to convey his ideas. The following quote is
represented in Figure 7.7, available online :
VICTOR: it’s about how the listener will receive something that makes
sense. So how the melody, how the phrase is made up, is completely
unique, and it’s made up of technical considerations of rhythm,
pitch, harmony, of where the top point is, where it’s going, how
fast it’s getting there, . . . how slowly or fast it unravels, how it does.
Shape as understood by performing musicians 235
I think phrasing’s about being able to see that, from this [indicates
score].56
Tina discussed wide-ranging parts of the model when talking about shaping
a phrase. She discussed musical triggers of melodic contour, harmony and
dynamic markings, and heuristics of audience, direction, imagery and line, as
well as changes in sound relating to timing and dynamics. Her quote is repre-
sented in Figure 7.8, available online .:
TINA: Yes, I suppose the shape of a phrase, whether it goes up
or . . . down, for example, the first line, thinking of it generally,
growing up to the top, and down again . . .
RESEARCHER: So is it the pitch you’re thinking about, in terms of the
shape, or—
TINA: Pitch, and, well, the dynamic, which is written in anyway. And
direction, so . . . some sort of forward movement towards the higher
point of it, so sort of trying to reach the top of it and then perhaps
away, and relaxing on the way back down again.
RESEARCHER: Do you mean forward movement in terms of tempo, or a
combination of things, or—
TINA: Um, not exactly tempo, not an accelerando, but a sense of it.
Someone I know describes things as, you play them either in the
present tense, or the future, or the past, so I s’pose, if you play
something in the future, you’re sort of looking forwards . . . um,
which doesn’t exactly mean you play . . . faster, . . . it means you’re
sort of on the front edge of maybe, of what you think the tempo is,
rather than the back edge.
RESEARCHER: Yeah, OK. . . Would you describe that as rubato, or is it
not quite as much as that?
TINA: It’s not, no, not as much as that, just a general sense, I suppose
a sense of ‘line’ through . . . some kind of thread that, your, sort
of, intention, that comes across. . . I suppose if you’re speaking, if
you’re reading something out loud, you make sure the words within a
sentence carry on, even though you have to articulate each word and
things, but you don’t [pauses] pause [pauses] until you get to the end
of the sentence, you make sure you’ve got there, I suppose.57
Darragh discussed a technically and perceptually complex passage from
Bach’s E minor Partita which contains implied polyphony. He noted how for
this particular passage, little conscious shaping was required, an approach
which is supported by recent research (Davis 2009), whereas at the end of the
passage he would begin shaping the music once more. It was apparent from
his playing that he was referring not only to the timing fluctuations within
his performance, but also the dynamic range, the timbre or tone colour, and
vibrato.
236 Music and Shape
When represented on the model (see Figure 7.9, online ), this quotation high-
lights shaping at the phrase level, with musical triggers of melodic contour and
(implied) polyphony, heuristics of breathing, composer and shape, technical
modifications in both hands (the left hand has complex fingering patterns and
shifts, and the bow moves in changing patterns relating to string crossings),
affecting the timbre, the timing and probably the dynamic variation and vibrato.
Yoshi, too, described shaping a single note, providing considerable detail about
the physical interaction between her arms and fingers and the keyboard of the
harpsichord, and about the resulting differences in the sound produced:
Shape as understood by performing musicians 237
YOSHI: I think sometimes, it’s the way you drop. [plays] If you just let
the weight of your fingers drop, or if you do it a little bit more [plays]
instant, not force, but just a little bit of ping on your finger, and then
you get more of a clear start to the sound. And if you, you can use
the flat bit of your finger, then it’s a little bit [plays] um, milder, a
little bit more sort of gentle, sort of plucking. . . I think the weight,
the speed, and also the angle . . . of the fingers will sort of, I think,
[plays] I guess you have more control [plays] when it’s flatter . . .
[plays] rather than that, but then, and then you sort of, sometimes,
just give it a little kick, and that’s a little bit . . . uh, it’s a little bit
more clear at the beginning, and somehow louder as well. [plays]60
When represented on the model, the heuristics of gesture, imagery and instru-
ment are highlighted, reflecting Yoshi’s discussion of the movements she is
making, the metaphorical ideas surrounding those movements (‘ping’, ‘little
kick’, etc.), and the technicalities of the instrument she is playing. She is dis-
cussing the attack of a single note, and therefore this is highlighted in the
musical level and technical modifications areas of the model. The change in
sound discussed concerns the timbre and dynamic of the note produced. This
can be seen in Figure 7.11, available online .
Discussion
It is clear from the examples shown that multiple components of the model are
frequently used by participants at once. Each broad category can be thought
about in isolation or considered in relation to another. Often, technical modifi-
cations may not be thought about on a conscious level, with performers thinking
instead of heuristics to achieve their desired change in sound. Nor are musi-
cal triggers always thought about consciously. Different participants seemed
to favour particular components, suggesting that, over time, performers may
develop their own preferred means of thinking about musical shaping that are
represented in numerous areas of the model. It is worth bearing in mind, how-
ever, that the model was built from data gathered in one interview with each par-
ticipant, and is unlikely to represent the full scope of the shaping experience. It
does, however, provide a picture of some of the ways in which these performing
musicians conceptualize and use the notion of musical shaping.
The model provides a new perspective on performance preparation, partly
because it is focused on musical shaping, rather than on performance prepar
ation in general. Some components seem to correspond with aspects of exist-
ing research findings. In relation to research in expert practice (Chaffin et al.
2002; Chaffin et al. 2010), many of the musical triggers, some of the heuris-
tics, and many of the technical modifications may be involved in the formation
of interpretative performance cues. Some of the heuristics also seem likely to
238 Music and Shape
would discuss shaping in the ways suggested here. A future study might look at
wind or brass players, or singers. Another interesting group might be players
of untuned percussion instruments: we could hypothesize that they might be
focused on rhythm, but to what extent do they shape what they play according
to the melodic and harmonic features of other parts?
Further studies might also establish whether or not the model has the poten-
tial to be generalized to western performers who are less reliant on a score, such
as musicians within the broad popular genre or jazz musicians. Within Chapter
8 of this book, Greasley and Prior argue that the performers of popular music
share responsibility for the shaping of the final sounds of the songs with others,
such as sound engineers, and indeed, classical musicians in recording settings
and certain live performance situations may also recognize this idea. The model
might therefore need to be extended to encompass the performers’ awareness
of and interaction with these other contributors; this is something that neces-
sitates further empirical study.
In its current form, this model offers an understanding of musical shaping
from the perspective of classical performing musicians. While the terms ‘shape’
and ‘shaping’ are commonly used by performers, their meanings have not previ-
ously been defined in relation to music. This model confirms the flexibility of
the term, highlighting its ability to be used in relation to all levels of the musical
structure; the influence of an array of musical triggers on performers’ shap-
ing decisions; the use of shape as one of a number of heuristics for expressive
performance; the technical modifications required to shape a note, phrase, sec-
tion, etc.; and the change in sound that results. At the very least, the data and
the resulting model have allowed some understanding of the commonly used
phrase ‘That was a beautifully shaped performance’, and that understanding
may perhaps help others to achieve that elusive goal.
Acknowledgements
This work was supported by the AHRC Research Centre for Musical
Performance as Creative Practice (grant number RC/ AH/D502527/ 1). The
author is most grateful to David Mackin of Greengate Publishing Services for
producing the figures for this chapter.
References
Bangert, D., D. Fabian, E. Schubert and D. Yeadon, 2014: ‘Performing solo Bach: a case
study of musical decision-making’, Musicae Scientiae 18/1: 35–52.
Bangert, D., E. Schubert and D. Fabian, 2014: ‘A spiral model of musical decision-mak-
ing’, Frontiers in Psychology 5/320, https://doi.org/10.3389/fpsyg.2014.00320 (accessed
9 April 2017).
240 Music and Shape
Barten, S. S., 1998: ‘Speaking of music: the use of motor-affective metaphors in music
instruction’, Journal of Aesthetic Education 32/2: 89–97.
Brenner, B. and K. Strand, 2013: ‘A case study of teaching musical expression to young
performers’, Journal of Research in Music Education 61/1: 80–96.
Cannam, C., C. Landone and M. Sandler, 2010: ‘Sonic visualiser: an open source applica-
tion for viewing, analysing, and annotating music audio files’, paper presented at the
ACM Multimedia 2010 International Conference, Firenze, Italy, 25–29 October 2010.
Chaffin, R., G. Imreh and M. Crawford, 2002: Practicing Perfection (Mahwah, NJ: Erlbaum).
Chaffin, R., T. Lisboa, T. Logan and K. T. Begosh, 2010: ‘Preparing for memorized cello
performance: the role of performance cues’, Psychology of Music 38/1: 3–30.
Davis, S., 2009: ‘Bring out the counterpoint: exploring the relationship between implied
polyphony and rubato in Bach’s solo violin music’, Psychology of Music 37/3: 301–24.
Doğantan- Dack, M., 2013: ‘Familiarity and musical performance’, in E. King and
H. M. Prior, eds., Music and Familiarity: Listening, Musicology and Performance
(Aldershot: Ashgate), pp. 271–88.
Ericsson, K. A., 2006: ‘Protocol analysis and expert thought: concurrent verbalizations
of thinking during experts’ performance on representative tasks’, in K. A. Ericsson, N.
Charness, P. J. Feltovich and R. R. Hoffman, eds., The Cambridge Handbook of Expertise
and Expert Performance (Cambridge: Cambridge University Press), pp. 223–41.
Ericsson, K. A. and H. A. Simon, 1993: Protocol Analysis: Verbal Reports as Data
(Cambridge, MA: MIT Press).
Fabian, D., R. Timmers and E. Schubert, eds., 2014: Expressiveness in Music Performance:
Empirical Approaches across Styles and Cultures (Oxford: Oxford University Press).
Gabrielsson, A., 2003: ‘Music performance research at the millennium’, Psychology of
Music 31/3: 221–72.
Juslin, P. N., 2003: ‘Five facets of musical expression: a psychologist’s perspective on music
performance’, Psychology of Music 31/3: 273–302.
Juslin, P. N. and G. Madison, 1999: ‘The role of timing patterns in recognition of emotional
expression from musical performance’, Music Perception 17/2: 197–221.
Juslin, P. N., A. Friberg and R. Bresin, 2002: ‘Toward a computational model of expres-
sion in music performance: the GERM model’, Musicae Scientiae (Special Issue
2001–2): 63–122.
Juslin, P. N., A. Friberg and E. Schoonderwaldt, 2004: ‘Feedback learning of musical expres-
sivity’, in A. Williamon, ed., Musical Excellence: Strategies and Techniques to Enhance
Performance (Oxford: Oxford University Press), pp. 247–70.
Karlsson, J. and P. N. Juslin, 2008: ‘Musical expression: an observational study of instru-
mental teaching’, Psychology of Music 36/3: 309–34.
Leech-Wilkinson, D., 2009: The Changing Sound of Music: Approaches to the Study of
Recorded Musical Performances, http://www.charm.kcl.ac.uk/studies/chapters/intro.html
(accessed 9 April 2017).
Leech- Wilkinson, D. and H. M. Prior, 2014: ‘Heuristics for expressive perfor-
mance’, in D. Fabian, E. Schubert and R. Timmers, eds., Expressiveness in Music
Performance: Empirical and Cultural Approaches (Oxford: Oxford University Press),
pp. 34–57.
McPherson, G. E., J. W. Davidson and R. Faulkner, 2012: Music in Our Lives: Rethinking
Musical Ability, Development and Identity (New York: Oxford University Press).
Shape as understood by performing musicians 241
Prior, H. M., 2012a: ‘Methods for exploring interview data in a study of musical shap-
ing’, paper presented at the 12th International Conference on Music Perception and
Cognition (ICMPC) and 8th Triennial Conference of the European Society for the
Cognitive Sciences of Music (ESCOM), Thessaloniki, Greece, 23–28 July 2012.
Prior, H. M., 2012b: ‘Multi-modal understandings of musical shape: a comparison of
violinists and harpsichordists’, paper presented at the SEMPRE 40th Anniversary
Conference, Institute of Education, London, UK, 14–15 September 2012.
Prior, H. M., 2012c: ‘Shaping music in performance: report for questionnaire participants
(revised August 2012)’, http://www.cmpcp.ac.uk/wp-content/uploads/2015/09/Prior_
Report.pdf (accessed 9 April 2017).
Smith, J. A. and M. Osborn, 2003: ‘Interpretative phenomenological analysis’, in J. A. Smith,
ed., Qualitative Psychology: A Practical Guide to Research Methods (London: Sage),
pp. 51–80.
Smith, J. A., P. Flowers and M. Larkin, 2009: Interpretative Phenomenological Analysis
(London: Sage).
Willig, C., 2001: Introducing Qualitative Research in Psychology: Adventures in Theory and
Method (Maidenhead: Open University Press).
Woody, R. H., 2002: ‘Emotion, imagery and metaphor in the acquisition of musical perfor-
mance skill’, Music Education Research 4/2: 213–24.
Reflection
Simon Desbruslais, trumpeter
Expressive freedoms
(a)
(b)
FIGURE R.19 Bach, B minor Mass, Gloria II, bars 57–61: a) Gesellschaft edition, followed by b) a
notated interpretation
towards high notes, (3) upper-note trill, (4) no vibrato and (5) slurred semi-
quaver couplets. Figure R.19a recreates the original markings from the Bach
Gesellschaft edition, followed by a transcription of one possible interpretation
(Figure R.19b). Notably, both second and third trumpets also have creative
roles in this style of writing.
This interpretation could be related to the physical baroque instrument.
‘Phrasing- off’ slurred couplets (emphasizing the first note) helps stamina,
accords with extant treatises (such as Quantz [1752] 2001) and creates a
nuanced and layered musical character. The upper (clarino) register is quiet; we
know this both from experimentation with surviving instruments and from the
orchestration of works such as Bach’s Second Brandenburg Concerto, where
trumpet must balance with the concertino of violin, oboe and recorder. Upper-
note lip trills, it may be argued, are easier to play. Particularly when using a
large baroque mouthpiece, vibrato is very hard to create and unnecessarily
exhausting. Finally, semiquavers are easier to perform when slurred in couplets.
While HIP informs performances (the approach described here is represen-
tative of many period trumpeters) the performer is still freer to choose than in
a detailed contemporary score. And although I have reached something of a
norm in my performance of this extract, I remain free to change; the scarcity of
dynamic markings encourages a creative approach to phrase shape.
The next passage (Figure R.20) is something that I have heard hammered out
in modern groups, where the triads are interpreted as loud articulations against
Bach’s complex counterpoint. However, this approach, perhaps influenced by
Reflection: Simon Desbruslais 245
FIGURE R.20 Bach, B minor Mass, Cum sancto spiritu, bars 111–17
the sound of the piccolo trumpet, misses the opportunity to shape this phrase.
A diminuendo towards the final top C (a sounding D) provides an elegant alter-
native, emphasizing instead bar 113 as the dynamic climax.
Though it creates an attractive musical shape, this approach is harder to
perform. Some shapes, however, can make a trumpeter’s life easier. I have heard
many times (and I am sure this is true for other instrumentalists) that one
should make a long note ‘travel’ or ‘go somewhere’. This can be a psychological
tactic to encourage the performer to breathe or bow ‘through’ a note—to work
harder as the note progresses—in order to sustain a long note where the effect
would otherwise be static. However, this can also form dynamic and colour
shapes. The example in Figure R.21 is taken from the third volume of Güttler’s
collection (1970) of Bach’s trumpet music, the definitive text for professional
performers. It is my personal copy, which I have used for many live perfor-
mances on the natural trumpet. I find that the imaginary slurs help to remind
me of the overall shape and direction of the phrase (it is no coincidence that
I am supportive of Schenkerian analysis). I want to see how the pitches relate
to each other: rather than symphonic technique, where pitches are accurately
punctuated with a uniform character, I want to understand, and remind myself
in performance of, the larger musical line.
The extended melodies of the nineteenth century, enabled by the invention of
the chromatic, valved trumpet, encouraged and lent themselves to more extensive
colouration. This period introduced the modern orchestral solo, which became
a platform for performers to show both technical assurance and the individual-
ity of sound colour. However, it was originally the cornet that was assigned the
freedom to perform expressive solos, while the baroque trumpet was demoted
to a less creative role with a function mainly to articulate (Lawson and Stowell
FIGURE R.21 J. S. Bach, Complete Trumpet Repertoire, Vol. III with my annotations (used by kind
permission of Breitkopf & Härtel, Wiesbaden).
246 Music and Shape
FIGURE R.22 Tchaikovsky, Swan Lake Suite Op. 20a, ‘Intrada’, rehearsal mark 13
FIGURE R.23 Pritchard, Skyspace (2012), third movement, notated for piccolo trumpet in A, bars 1–8
(used with permission).
1999: 130–2). This was due in part to the loss of high baroque trumpet technique.
Nineteenth-century orchestral solos often have a sense of expressive freedom, of
which the famous cornet solo in Pyotr Tchaikovsky’s Swan Lake Suite is exem-
plary (Figure R.22). These moments highlight the individuality of the principal
trumpeter’s sound, line and musical shapes. Generally, tied notes encourage both
a small dynamic change and a change in vibrato and colour.
The often complex styles of contemporary music have encouraged a
‘straighter’ approach which mirrors that of much baroque performance.
Playing with vibrato is more exhausting, requiring greater stamina, and the
high demands of contemporary music are more easily met with a straighter
sound. This is not to say that vibrato is completely avoided, but it is not the
primary colour. Similar physical strength is required to play the clarino trum-
pet and contemporary repertoire.
Individual nuances in musical phrase, however, are as important in many
contemporary works as they are in the baroque. While Deborah Pritchard’s
piccolo trumpet concerto, Skyspace (Figure R.23), contains meticulous atten-
tion to notational detail, I add my own character (indeed, I do not think that it
is possible to notate every expressive detail in a score). In the third movement,
although this is not notated, I ‘lift’ the second of each quaver-pair in a manner
related to my experience of baroque music.
Concluding remarks
References
Bach, J. S., 1970: Complete Trumpet Repertoire, ed. L. Güttler, 3 vols (Wiesbaden: Breitkopf
& Härtel).
Lawson, C. and R. Stowell, 1999: The Historical Performance of Music: An Introduction
(Cambridge: Cambridge University Press).
Quantz, J. J., [1752] 2001: On Playing the Flute [Versuch einer Anweisung, die Flöte traver-
siere zu spielen], trans. E. R. Reilly (London: Faber & Faber).
Reflection
Malcolm Bilson, fortepianist
Defining musical shape
In a casual conversation in 2001 a very famous pianist asked me, ‘Why is there
no dot on the upbeat to the first movement of the Beethoven Piano Sonata in
F minor, Opus 2 No. 1’ (Figure R.24). I was taken aback that anyone could
ask such a question, as every eighteenth-century source clearly states that all
upbeats are short and light unless otherwise marked. One doesn’t put an expres-
sive marking on notes that are akin to articles in speech (the, an, of, by, etc.).
I then listened to some twelve recordings by world-famous artists of the last
decades and was astonished to find that almost none played the note short or
light. Most played a heavy, long, even slurred upbeat (clearly not indicated by
the composer). Astonishing though it may seem, very little instruction if any is
given in conservatories around the world concerning the most basic expressive
devices used by composers: How long is a crotchet to be held that has no mark
of any kind (a slur, a tenuto)? What is the meaning of a slur? (Mozart never
wrote a sketch without indicating the slurs—they are the real soul of the music,
realized through the notes.) I made a video on these subjects called Knowing the
Score, which was released in 2006.1
In addition to the more basic questions of notation, one of the topics touched
on in Knowing the Score was performance information that can be gleaned from
composers who have recorded their own works (Bartók, Prokofiev, Elgar and
others). But rather than listening to the recording to see how the composer inter-
prets the score, we can assume that the score represents what these composers
heard, hence: How did they write it down? One of the recorded examples featured
in Knowing the Score was Sergei Prokofiev playing his little Gavotte, Op. 32 No.
3, in what is generally considered a personal, highly idiosyncratic manner. I made
the claim that if we know what a Gavotte is, and follow Prokofiev’s markings
carefully, his rendition will be clearly revealed in his notation. (See Video 1 at .)
248
Reflection: Malcolm Bilson 249
FIGURE R.24 Beethoven, Piano Sonata in F minor, Op. 2 No. 1, first movement, bars 1–9
Mozart and others of the time instruct us how to realize the performance indi-
cations in the scores of the time in order to play in the kind of inflected manner
we observed in the Prokofiev example. There is no music anywhere in the world,
from the simplest folk tunes to the most sophisticated art music, that is played
in an even, uninflected manner, yet such a manner of playing is often accepted
and even cultivated today, as evinced by many recordings of important artists.
I was on the jury of the Leeds International Pianoforte Competition in
2000. What I heard was a phenomenal level of piano playing, and I was often
moved by very beautiful and insightful playing in a variety of repertoires. But
there was no Mozart or Haydn that came even close to the beauties I associate
with that music; it was all smooth and even, virtually uninflected. One work we
heard five or six times was the Haydn Sonata C Major, Hob. 50. In Video 2
you can hear the first few bars performed by the fine Hungarian pianist Dezsö
Ránki. I chose his performance as emblematic of what I heard several times in
Leeds. His performance is by no means unmusical, and represents beautifully
a typical rendition of this score. But in the video I look at the detailed perfor-
mance indications in the score to demonstrate that Haydn’s expressive mark-
ings are at least as important for revealing the musical thoughts as the notes,
yet in Ránki’s, as in most performances today, they are simply glided over in
a smooth, uninflected manner. Musical shape—inflected, passionate and flex-
ible—is inherent in this music, and it is my belief that it can be regained by a
clear understanding of the expressive marks ubiquitous in music of the late
eighteenth and early nineteenth centuries that are today misunderstood or sim-
ply neglected. No one applauds more than I the wonderful new scholarly edi-
tions appearing, giving us access to every little aspect of Mozart’s or Schubert’s
notation. There are now at least seven so-called Urtext editions of the Mozart
Piano Sonatas; but are any of his clear articulations, the essence of his musical
language, being taught in those music schools and conservatories that insist on
their use?
Musical shape is defined by properly inflected realizations of rhythmic
motives as we have shown in this short Haydn example. But two further aspects
of time in musical performance are equally important: tempo fluctuation and
tempo rubato. I am often told that in Beethoven’s music no tempo fluctuation
is allowed, that one must keep a strict beat. Not only is there no basis for this
widely held assumption, but we know that Beethoven changed tempo a great
deal, and indeed much in his music virtually demands it. And tempo rubato,
prized by Caccini in the seventeenth century, Mozart in the eighteenth and
Chopin in the nineteenth, involves independence between a steady accompani-
ment and a freely flowing upper voice, be it violin, the voice or the right hand
at a keyboard. This feature as well is generally discouraged in today’s conser-
vatories and music schools, yet is often the very soul of moving performances
heard on earlier recordings.2
Reflection: Malcolm Bilson 251
Note
This Reflection derives from a lecture given at the Liszt Academy, Budapest, on
4 March 2014. A video of the lecture is available from the companion website
References
Bach, C. P. E., [1753, 1762] 1949: Essay on the True Art of Playing Keyboard Instruments, trans.
W. Mitchell (New York: Norton).
Türk, D. G., [1789] 1967: Klavierschule, facsimile reprint (Kassel: Bärenreiter), pp. 347–65.
8
the limitations of examining within a single chapter a flexible and widely applic
able metaphor such as shape in a genre as diverse as popular music.
SETTING THE BOUNDARIES
The term ‘popular music’ has been so widely used and defined that it is essen-
tial to begin with a brief discussion of the scope of the term as it pertains to
our work. In this chapter we are referring to popular music in contemporary
Britain, Europe and North America, mainly because most of the research to
date has been carried out in these contexts. In distinguishing between folk, art
and popular music within western culture, Philip Tagg (1982) observes that
popular music tends to be produced and transmitted primarily by professional
musicians; is mass distributed mainly through recorded sound;1 is a commod-
ity in an industrialized society; and tends to name composers or authors. Tagg
also notes the general lack of written theory and aesthetics, though this has
since developed (Bennett, Shank and Toynbee 2006; Brabazon 2012; Frith and
Goodwin 1990; Moore 2001; Negus 1996; Scott 2009). A useful definition of
popular music, and one that we will be adopting in the current chapter, is pro-
vided by Shuker (2013: 6):
In sum, only the most general definition can be offered under the general
umbrella category of ‘popular music’. Essentially, it consists of a hybrid
of musical traditions, styles and influences, with the only common ele-
ment being that the music is characterised by a strong rhythmical compo-
nent and generally, but not exclusively, relies on electronic amplification.
Indeed, a purely musical definition is insufficient, since a central charac-
teristic of popular music is a socioeconomic one: its mass production for
a mass, still predominantly youth-oriented market. At the same time, of
course, it is an economic product that is invested with ideological signifi-
cance by many of its consumers.
Musical theatre 3
Jazz/blues 13
Pop 9
Rock/metal 5
Country/folk/gospel 3
Urban (hip hop, soul, RnB, etc.), dance/electronic 3
(house, techno, electronica, etc.)
Contemporary/experimental 4
World 4
Crossover 3
than one genre. Greasley et al. (2013) also highlighted difficulties in definition
because of crossover in musical styles (e.g. ‘Folk Rock’, ‘Country Rock’, ‘Jazz
Rock’), which is why Shuker’s (2013) broad definition of popular music as con-
sisting of a hybrid of musical traditions, styles and influences is useful here.
Readers are invited to draw conclusions from the arguments we present that are
appropriate for the genres and subgenres in which they specialize.
We also make generalizations about the roles played by a number of contrib-
utors (e.g. performer, producer) in shaping popular music. Frith and Zagorski-
Thomas (2012: 5–6) note that ‘deciding who is responsible for what in the studio
is still a matter of record-by-record investigation (much of which remains to be
done) rather than, for example, genre generalization’. This can also be applied
to live music-making in popular music, with its frequent use of technology and
concomitant expert personnel. We aim here to present the potential for each
contributor to shape the music, rather than analysing existing recordings or
presenting any kind of blueprint for musical success.
In this section, we explore the performer’s role in shaping music in live per-
formances, drawing on evidence from popular musicians who responded to a
questionnaire study on musical shaping (Prior 2010, 2012b) and on work in the
popular music field.
Prior’s (2010, 2012b) questionnaire study provides insights into the use of
musical shape or shaping by musicians from a relatively broad range of back-
grounds. More than two hundred participants completed a mixed-response
questionnaire, which sought to establish some of the meanings and contexts
in which the idea of musical shape is used by musical performers. Participants
were asked about their musical background (e.g. main instrument, music
256 Music and Shape
There seemed to be three main ways in which the idea of musical shape was
used by these musicians. First, a few performers discussed their use of shapes
and images to overcome technical difficulties. Several singers described how
shape was helpful for themselves and their students in achieving the correct
pitch, tone colour and expression:
It was taught to me that when visualising your voice you should think of
it as a shepherd’s hook—that it runs from your diaphragm, up through
your body, into your mind, resonating behind the nose, and out through
your mouth. (Professional-standard singer)
An amateur guitarist also discussed the ‘big, round shape’ of the ‘expansive’
timbre he was trying to achieve. A similar technical approach was described by
a guitarist and a pianist. The guitarist described how he would visualize chords
as ‘shapes on the fretboard’ (professional-standard guitarist). The pianist took
this idea further:
While improvising over the tune, I imagine the chords not as abstract
notions, but like architectures—and my movement from one to another
involves drawing different shapes which I select as I go along. (Professional-
standard pianist)
This idea of shapes being formed over time by changing musical features such
as chords or melodic patterns leads to the second way in which the idea of
musical shape was used, that is, in reference to a musical structure or trajec-
tory. This was a more common idea, cited by ten of the participants and often
discussed in relation to composition or improvisation:
Shape would have been used to talk about a large scale structural/expres-
sive trajectory which can help to guide an improvisation. (Professional-
standard pianist)
258 Music and Shape
The idea was also discussed on a larger scale, in relation to the choice of music
over a whole performance:
So each piece had a different shape, to give variety to the performance.
Not just tutti, then individual jazz breaks, then chorus. We tried to vary
the structure and shape of the programme. (Professional double bassist)
We were discussing which track to open our set with, given the style of
the DJ who would be playing before us. We wanted to find an opening
track that would work well after the previous DJ and be significantly
different but also energetic enough not to clear the dance floor. We dis-
cussed it in terms of energy level, often using contour metaphors, which
are absolutely central to DJing (peaks and troughs, building it up, taking
it down). (Amateur DJ)
The way in which a set is compiled can create a very powerful performance, as
exemplified by Fast’s (2006) description of Queen’s performance at Live Aid,
in which the group abbreviated many of their most popular hits to create a
fifteen-minute act with a powerful emotional trajectory designed to enthuse the
audience and thereby raise as much money as possible.
The latter comments made by the questionnaire respondents form a useful
introduction to the third main way in which musical shaping was discussed,
that is, in relation to musical expression. This was discussed on a variety of
scales, in relation to both the whole piece and the shaping of individual phrases.
Sometimes these ideas were discussed in specific technical terms, with refer-
ence to phrasing and breathing, dynamics and tempo fluctuations, all of which
might vary according to the acoustic of the performance space:
How to craft phrases, the beginnings and ends of phrases, the swell of
dynamics, minute tempo changes bar to bar. (Professional- standard
euphonium player)
Tried to shape the line as I heard the song, giving breaks for breath at
what felt like natural points in the line and continuing through places that
needed a sense of continuation and flow. (Professional trombone player)
Shaping the music, rather like a sentence in poetry. Use of dynam-
ics to highlight the phrase. Reacting to new and unfamiliar acoustics.
(Professional-standard percussionist, in this instance conducting a choir)
Shaping popular music 259
Warner 2003), yet only a small number of respondents referred to their use
of technology in relation to musical shaping. Participants mentioned simple
techniques such as ‘reverb’ and ‘fading out’, as well as the use of previously
recorded performances of improvisations to aid the creation of new improvisa-
tions. The latter is best understood through one participant’s own words. He
described the situation as ‘rehearsing in a duo with a saxophonist I regularly
work with’, and the use of shape as follows:
Listening back to previous recordings to give an idea of where the key
ideas lie, the piece being rehearsed will begin from the key idea and prog-
ress to an end-point. The duration of these pieces are usually short and the
‘shape’ of such pieces that have been devised from this method are usually
more focused than ones that last longer in duration. (Professional pianist)
The above comments suggest that the availability of computer programs that
display music and sound as waveforms has added a visual element to these
participants’ understandings of musical shaping (see the Reflections by Savage
and by Reuben later in this volume), rather as notation seems to have done for
classical musicians’ conceptualizations of shaping (see Küssner, Chapter 2 of
this volume). Other technology has also had an influence. Two participants spe-
cifically mentioned mixing and equalization:
From a mixing point of view—EQ-ing tracks to blend together better
is very often a visual thing, i.e. different instruments’ contour shaped so
they don’t all compete for the same frequency ranges. (Amateur electric
guitarist)
When DJing you manipulate the equalization (EQ) of tracks in order to
make them blend as well as possible, which means thinking in terms of
Shaping popular music 261
The demands of the recording studio with its concomitant customized environ-
ment and lack of audience require of performers a different understanding of
musical performance compared to their usual live performance situation (Blake
2009; Gander 2011; Horning 2012; Pras and Guastavino 2011; Williams 2012;
Zak 2009). While studies in the popular field typically present producers as hav-
ing most (or in some cases all) of the control over the finished musical product,
performers have responded creatively to both the technical restrictions (most of
which are now historical) and the opportunities afforded by the studio environ-
ment, and have generated new performance techniques as a result (Doğantan-
Dack 2008; Cook et al. 2009; Frith and Zagorski-Thomas 2012). Many of the
Shaping popular music 263
between Macero and Davis had been a great deal more collaborative—that the
pair had listened to the many hours of takes together in order to make deci-
sions about which sections to edit, splice and cut in the production of the final
album.
Technological advances drive creative practices in the studio (Théberge 1989,
2001; Warner 2003): the microphone offers a crucial example (Horning 2002,
2004). Horning (2004) maintains that ‘the art of microphoning’ (see Canby 1956)
is a skill which evolved as a natural progression from the recording engineer’s
placing of performers before the acoustical recording horn, and one which is
acquired tacitly—by recording engineers and performers alike—through experi-
ence. It is a skill that can be used to achieve unique musical outcomes (Moorefield
2005), and that in some hands has been likened to ‘a painter mixing colours from
a palette’ (Horning 2002: 710). The increased role and responsibility of the engi-
neer for achieving musical balance through careful placement of microphones
led to the development of the multitrack studio, which was instrumental in the
development of popular and rock styles through the potential it offers for the
control and layering of sounds (Théberge 1989, 2001). Multitrack recording
was first used in popular music in the 1950s and is characterized by the separate
recording of multiple sound sources to a number of audio channels to create a
recording. This allows engineers to examine intricate details of timing and tuning
(Blake 2009; Frith and Zagorski-Thomas 2012) as well as the broader perspective
of the musical sound (Zak 2009), an approach which has again led to produc-
ers likening their role to that of an artist painting (Phil Harding, in Frith and
Zagorski-Thomas 2012). Such technology has increased the control of producers
over the recorded sounds, not only because of the detailed level at which they are
able to work, but also because of the necessary separation ‘of the artists from
each other, separation of their performances, and further a separation of the
artists from their song and even their performance’ (Gander 2011: 132). Gander
argues that this empowers the producer to make musical decisions.
Other technological advances have been seen in sampling and computer-
based sequencing, including signal processing, Musical Instrument Digital
Interface (MIDI) sequencing, and sound synthesis (Blake 2009; Katz 2004;
Théberge 2001; Warner 2003). Signal processing enables producers to add spe-
cial effects (e.g. reverb, delay, chorus, flange, compression) to tracks; MIDI
sequencers facilitate enhanced control over layering of sounds; and digital sam-
pling enables the manipulation of sound in a variety of ways down to the finest
detail without any discernible loss of sound quality (Goodwin 1990). The level
of control of the sound that these technologies afford provides many creative
opportunities: one only needs to think of the ‘Amen’ break—a four-bar drum
solo performed by Gregory Coleman in the 1960s song Amen, Brother which
has been used extensively in a range of electronic music styles such as break-
beat, hip-hop, hardcore, jungle, and drum and bass (Butler 2006)—to realize
the potential for the use of samples in creating new records. Some authors,
Shaping popular music 265
given by musicians and producers throughout the literature. Moreover, the bal-
ance of performer, producer and technology will change with every record-
ing, not just as a result of a particular set of personalities involved. Björk, for
example, is said to have described her albums Post and Debut as ‘collections
of duets with the producers who had inspired her: Nellee Hooper, 808 State’s
Graham Massey, Tricky, Howie B.’ (Jonathan Van Meter, in Brackett 2009:
522). In contrast, she states that her later album Homogenic ‘is more like one
flavour. Me in one state of mind. One period of obsessions. That’s why I called
it Homogenic’ (ibid.).
The discussion concerning record producers’ shaping of popular music above
clearly simplifies the myriad influences on the producers themselves. Zagorski-
Thomas (2012) highlights the complex social, commercial and economic fac-
tors contributing to both the availability and the use of recording spaces and
technologies by record producers that can be perceived in the sounds of the
records they produced. The separation of shaping by performer(s) and producer
is somewhat artificial: often, the collaboration between the two is sufficiently
close that separation of the decision-making process is impossible. This situa-
tion is compounded when the performer becomes a sound engineer or producer:
Brabazon (2012) describes a situation where the performer undertakes some
of the production work herself, but later hands her materials over to another
producer to ‘clean up’ and combine with new recorded materials. Some artists
have recorded entire albums themselves: Moby recorded his album Play at home
using Cubase software (ibid.: 62). What is clear is that technological develop-
ments in the studio (whether professional studio or home studio) have been—
and still are—at the heart of the creative process in popular music.
While popular music recordings (e.g. original recording, remix) may be viewed
as a fixed ‘final product’ once released, they are also used by others to create
new musical experiences. One may be the creation of an accompanying video,
mentioned briefly at the outset. Another, which we explore here in more depth,
is the creation of a new live, ephemeral performance through the use of record-
ings by DJs. While some DJs perform solely with their own music productions,
most are typically using combinations of others’ dance music recordings. This
involves a process of selecting records to be performed back-to-back as well as
consideration of the order in which the records will be performed overall. Some
DJs perform ‘live’ sets, creating dance music over the course of the set (which
may be one, two or three hours in duration) by improvising with beat loops
and samples (Collins 2007). In this instance, the music is still pre-recorded,
but the raw material is manipulated, transformed and recomposed in real-time
using programmes such as AbletonLive, a music sequencer and digital audio
Shaping popular music 267
workstation (DAW). There is also live coding wherein programmers write code
in real time which translates into musical output (Collins et al. 2003; Eldridge’s
Reflection earlier in this volume), though a consideration of this is beyond the
scope of this chapter. Here we focus on the use of recordings by DJs, drawing
on our own recent research.
We conducted an interview study exploring DJs’ perspectives on musical
shaping (Greasley and Prior 2013). Using a similar methodology to the inter-
view study with classical musicians mentioned above (Prior 2012a; also her
Chapter 7), we asked DJs to perform a short mix in three conditions: first with
unfamiliar records, second (with the same records) while thinking about musi-
cal shaping or the shape of music, and third (again with the same records) with-
out musical shaping. The first author visited the DJs in their homes, and they
performed on their usual equipment (i.e. decks and mixer set-up). Responses
given by the three male DJs were then compared to the responses of a single
female DJ to the questionnaire study (Prior 2012b) to explore similarities and
differences in the DJs’ perspectives on musical shaping. Results showed a num-
ber of similarities. First, all of the DJs emphasized the importance of record
selection in shaping the overall contour of a performance, confirming previous
literature which has highlighted the centrality of choosing the ‘right’ tracks
in the ‘right’ order (Broughton and Brewster 2002; Straw 1993). As well as
reflecting their own musical preferences, the DJs reported that their choices are
shaped by other factors such as the type of club (e.g. capacity, sound system,
dance floor/seating), the specific night (e.g. single musical style or combination,
night’s reputation), the absence/presence of co-performers, and the audience.
According to the DJs in our sample, the expected audience exerts a power-
ful influence on the records chosen both before and during the performance.
Referring to popular music producers in the recording studio, Hennion (1989)
argued that the audience is never ‘left outside’, that the audience is always in
consideration. Similarly for these DJs, it was apparent that musical choices for
the performance are made with the audience firmly in mind.
A second similarity in perspectives on musical shaping was that all of the
DJs used the functionality on the turntables (e.g. pitch faders) and mixer (e.g.
up-fader, cross-fader, equalization) to manipulate the tempo, volume and fre-
quencies (e.g. bass, mid, treble) to modify the overall sound. The DJs reported
reducing elements of the outgoing tune and increasing elements of the incom-
ing tune, with a particular focus on bassline entries (a combination of which can
be ‘too much’ if equalization is not sufficiently balanced). They also discussed
their use of effects such as echo, flange and reverb to emphasize key structural
points (particularly the ‘drop’, where the bassline reenters after a breakdown).
These findings confirm previous research (Brewster and Broughton 1999, 2012;
Moorefield 2010) in that DJs go beyond the pre-recorded musical materials
they are working with, creating unique compositions, in real time, in the con-
text of the performance (see also Smith 2013).
268 Music and Shape
All of the DJs emphasized that records had inherent shape (one DJ con-
trasted The 45 King’s The 900 Number with DJ Shadow’s Stem/Long Stem to
illustrate this; see Greasley and Prior 2013), and much like the classical musi-
cians in Prior’s (2012a) study, they argued that it was not possible to elimi-
nate shape entirely from a performance because of this. The DJs in our sample
discussed the ways in which they used the existing shape of records, such as
placement of ‘drops’, and equalization to balance overall sound intensity and
musical texture. In trying to perform without shape, the DJs were less likely to
use the existing shape of the record or key structural points, or to employ func-
tions on the turntables and mixer other than for beat synchrony.
Importantly, performing without shaping the music felt ‘unnatural’ to
these DJs, highlighting the seemingly implicit nature of musical shaping (see
Greasley and Prior 2013). This mirrors a key finding in Smith’s recent work
(2013) exploring the compositional processes of hip- hop turntable teams:
she found that decisions about which samples to use and which techniques to
employ were applied unconsciously by the teams in order to achieve the desired
musical outcome. There seems to be a tacit understanding that a ‘good’ per-
formance requires shaping. A further similarity in Greasley and Prior’s (2013)
study was the multimodal understanding of shape that the DJs expressed; they
used gesture and visual diagrams to explain their notions of musical shap-
ing. In particular, DJs reported using the visual diagram of the waveform (i.e.
through Serato, Traktor or the sound recording software) to help them shape
their performances.
The DJs worked with a variety of technology—ranging from Technics
1210 turntables and a Pioneer DJM600 mixer to the latest Denon SC3900
digital media players and a Rane 16 digital mixer—and this highlighted key
differences in their perspectives on musical shaping. The two DJs working
with digital systems discussed their ability to assign cue points on the records
and jump straight to those at any point during a performance. They discussed
how they are able to programme sections of the record or samples, use sam-
ples from the same record simultaneously (as if playing with two identical
vinyl copies at the same time) and loop segments of the musical material.
They reported that during a performance, some of this may have been pre-
pared in advance, while other choices will be made in response to audience
behaviour. The combination of software and hardware allows a greater range
of creative acts; as Poschardt (1998: 365) has argued, ‘technology for them
[DJs] is an integral part of life that offers virtually unlimited creative oppor-
tunities’. In addition, programs such as Serato and Traktor have a synchro-
nization button, which automatically synchronizes the beat on the records
playing. A DJ using this feature is spared the time required to beat match, and
can start applying various effects to modify the sound almost immediately.
The role of technology in DJ practices has been explored by Montano (2010),
Shaping popular music 269
whose research with DJs in the Sydney dance music scene provides evidence
for the practices employed by our participants. He notes that:
The development of technology has enhanced the work of the DJ, so that
tracks can be altered and reshaped in order to fit the specific requirements
of the DJ. Vocals can be added and tracks can be extended or shortened,
allowing the DJ to have more control over the actual ‘sound’ of their set,
which increases the extent to which they can impose their own personal,
unique ‘musical’ identity upon it. (ibid.: 404)
Both of the DJs working with digital set-ups in our study were also scratch
DJs, or as Katz (2012) would call them, ‘performative DJs’, who not only
select recordings but manipulate them in real time for audiences. Scratching is
a specific performance style which involves the use of turntables as a musical
instrument to create and manipulate beats, sounds and samples (typically from
a wide selection of popular music styles) for expressive performance (ibid.;
Hansen 2010; Poschardt 1998; Smith 2013). For these DJs, scratching tech-
niques were fundamental to their conceptualization of a shaped performance.
They emphasized the importance of identifying samples and employing vari-
ous scratch styles (e.g. crab, scribble, hydroplane) and turntablism techniques
(e.g. beat juggling).
In summary, for the DJs we interviewed, shape was related to musical struc-
ture, dynamics, using samples, selecting records and mixing styles. They also
expressed a multimodal understanding of shape including the visual represen-
tation of tracks and mixes, and indicating shape-related ideas using gesture
(see Greasley and Prior 2013 for a full write-up of the study). There were
therefore a number of overlaps between these DJs and the popular musicians
in Prior’s (2012b) questionnaire study, most notably relating shape to struc-
ture, to dynamics and to the multimodal understanding through visualizations
and gesture.
This chapter has explored notions of musical shaping from the perspectives
of performer, producer and engineer, and through the contexts of live perfor-
mance and studio recordings. The practices of popular music are viewed in a
somewhat simplified, layered approach. We began with the performers’ per-
spectives in live performance, while noting the contribution of sound engineers
in this context. Recent questionnaire research has shown that popular musi-
cians use the notion of shape (and images) to overcome difficulties, in reference
to musical structure (or trajectory) and to achieve particular expressive goals
(Prior 2010, 2012b). Only a few references were made by the popular musicians
270 Music and Shape
to their use of technology when thinking about or using the notion of shape.
This was surprising given that technology is an essential element in the defi
nition of musical sound and style, and that musicians’ varying use of tech-
nology reflects diverse aesthetic and cultural priorities (Théberge 1997). Other
research highlights how performers use technology (e.g. microphone, distor-
tion) to achieve particular musical outcomes (Greig 2009; Théberge 2001) and
also how the audience may influence performers’ decisions (Fast 2006; Inglis
2006). Performers and sound engineers in live contexts work towards the pro-
duction of a transient, ephemeral listening experience, and audience members
may participate in this to a greater or lesser extent through singing along and
moving to the music.
We then discussed the potential contributions of performer, sound engineer
and producer in a studio context to create a recording, which is usually viewed
as a fixed, repeatable listening object. In the absence of research specifically
asking these individuals about their understanding of ‘musical shaping’ in the
studio, we made inferences from literature in the popular field. The role of
technology, such as placement of microphones, multitrack recording, MIDI
sequencing and digital sampling, is crucial to creative practice in the studio,
but other influences on the finished product, such as other personnel involved
in the recording industry and economic factors, are also important. It is also
worth noting the potential influence of the recording process on a subsequent
live performance, as highlighted by Blake (2009). It is not always clear whether
the performer(s) or the producer has the greatest influence over the creative
decisions in the studio (particularly given the increasingly blurred boundar-
ies between artist and producer), and we agree with Frith and Zagorski-
Thomas (2012) that assessing relative contribution requires record-by-record
investigation.
Finally, we discussed the use of popular music recordings in new contexts
focusing on the live performances of DJs. DJ performance requires the creative
use of records that have been previously shaped by others; it is not unusual for
DJs to play with more than thirty records in an hour-long DJ performance. In
this way, just as recordings are seen to enable musicians to go beyond the poten-
tial afforded by live performances, DJs go beyond the possibilities afforded by
the simple playback of existing recordings by reshaping them into a new per-
formance, or even a new piece of music, usually creating a transient listening
experience for an audience. These ideas about musical shaping from the differ-
ent perspectives are summarized in Table 8.2.
To illustrate our conceptualization of layers in shaping popular music, it is
helpful to draw on an example in which a track has been performed live, recorded
in a studio, remixed, and then performed with by DJs. Sarah McLachlan’s track
‘Fallen’ from her album Afterglow (which sold more than five million copies
worldwide) is one such. She has performed the track in live contexts (nota-
bly her 2004 summer tour ending in Vancouver) and has also recorded it in
Shaping popular music 271
Who May Shape the Music Means by Which the Music May Be Shaped Final Result
the studio with producer Pierre Marchand. After the artist released the multi-
track stems of the studio recording, it has been remixed by a number of artists
(for example, Josh Gabriel and Dave Dresden produced the ‘Anti-Gravity Mix’)
which was then used in live performances by DJs around the world, and also in
272 Music and Shape
mixtapes (for instance, DJ ATB used the Anti-Gravity remix in his continuous
mix for the DJ2 series). There are many thousands of similar examples wherein
a piece originally written for acoustic live performance has been shaped and
reshaped over time by a range of musicians, producers and DJs.
‘Shape’ has been shown elsewhere to be a versatile term that is used
metaphorically and heuristically by classical performing musicians (Prior,
Chapter 7; Leech-Wilkinson and Prior 2014). Its versatility is also appar-
ent from the data and literature relating to popular music that have been
explored in this chapter. In the same way that some classical musicians can
understand shape to refer to both the large-scale structure of a piece of
music and the moment-to-moment changes they make to a single phrase
or note, shape for DJs seems to apply to an entire set or a track, but also
to small-scale changes made to the sounds produced (e.g. flange, reverb).
The role of the producer has been highlighted, extending further the col-
laborative shaping that may occur in creating a musical product. What may
differ between record production and live performance is the availability of
time to rework materials without public scrutiny, making the shaping pro-
cess of record producers perhaps more similar to that of an artist than to,
say, a dancer’s process (a similarity discussed by some classical musicians).
Indeed, we have noted the extent to which record producers liken their roles
to that of an artist. Phil Harding observes that producing a pop record is
like painting a picture. Susan Horning comments that the placement and use
of microphones has been likened to a painter mixing colours on a palette.
In live sound production or DJ performance, this artistic process is recali-
brated so that it does occur in real time, and yet the analogy persists, albeit
with a focus on the process of collaborative painting rather than a finished
product. In Smith’s work on turntable teams (e.g. Mixologists), one DJ (Beni
G) stated, ‘it’s like two, three or four artists all holding the same paint brush
wanting to paint one picture and they’re all trying to paint it in a slightly
different way’ (Smith 2013: 62). The recurring analogy may provide an alter-
native explanation to our proposition earlier in the chapter—namely, that it
could be due to the historical tendency to snub popular music—as to why so
many titles on record production and DJing start with ‘The Art of . . .’
This chapter has highlighted some of the perspectives on music and shape
that are apparent within popular music. Some insights were gained through
empirical research; others came from examining existing literature in the field.
While participants in the survey and interview studies were able to talk about
shape and to refer to graphic metaphors for their musical sensibilities, it was
less clear that ‘shaping’ is the way they think about their music unprompted or
that ‘shaping’ is something that drives their practices. More research is needed
before arriving at any conclusions in this regard, if indeed this is a question that
can ever be fully answered given musicians’ tacit understanding of the concept.
Nonetheless, the chapter provides a starting point for future research studies,
Shaping popular music 273
highlighting how notions of musical shape may differ with roles and contexts
(e.g. performer in a live setting, producer in the studio). There is a need to
study a broader range of performing musicians in some depth, to investigate
not only the ways in which different instrumentalists and singers conceptualize
music and shape, but also to explore the great diversity of genres. Future work
should also focus on the several roles in the studio (e.g. performer, producer,
sound engineer) given that there is no empirical work on this at present. Finally,
research could focus on the collaboration that comes to the fore in popular
music, attempting to determine how shaping decisions are distributed among
co-performers in a band or group, or between singers, instrumentalists, produc-
ers and sound engineers. It is clear that much remains to be done to understand
more fully the elusive yet prevalent, tacit and yet seemingly fundamental con-
cept of musical shaping within popular music.
References
Barthes, R., 1990: ‘The grain of the voice’, in S. Frith and A. Goodwin, eds., On Record:
Rock, Pop and the Written Word (London: Routledge), pp. 293–300.
Bayley, A., ed., 2010: Recorded Music: Performance, Culture and Technology (Cambridge:
Cambridge University Press).
Bennett, A., B. Shank and J. Toynbee, eds., 2006: The Popular Music Studies Reader
(Abingdon: Routledge).
Bielmeier, D. C., 2013: ‘Determining the relationship between new recording engineers’
perceived skill sets and those observed by their employers’ (PhD dissertation, Argosy
University).
Blake, A., 2009: ‘Recording practices and the role of the producer’, in N. Cook, E. Clarke,
D. Leech-Wilkinson and J. Rink, eds., The Cambridge Companion to Recorded Music
(Cambridge: Cambridge University Press), pp. 36–53.
Boltz, M. G., 2013: ‘Music videos and visual influences on music perception and apprecia-
tion: should you want your MTV?’, in S.-L. Tan, A. J. Cohen, S. D. Lipscomb and R. A.
Kendall, eds., The Psychology of Music in Multimedia (Oxford: Oxford University Press),
pp. 217–34.
Brabazon, T., 2012: Popular Music: Topics, Trends and Trajectories (Los Angeles: Sage).
Brackett, D., ed., 2009: The Pop, Rock and Soul Reader: Histories and Debates, 2nd edn
(New York: Oxford University Press).
Brewster, B. and F. Broughton, 1999: Last Night a DJ Saved My Life (London: Headline).
Brewster, B. and F. Broughton, 2012: The Record Players: The Story of Dance Music Told
by History’s Greatest DJs (London: Virgin Books).
Broughton, F. and B. Brewster, 2002: How to DJ: The Art and Science of Playing Records
(London: Transworld).
Burgess, R. J., 2001: The Art of Music Production (London: Omnibus Press).
Burns, G., 2006: ‘Live on tape. Madonna: MTV Video Music Awards, Radio City Music Hall,
New York, September 14, 1984’, in I. Inglis, ed., Performance and Popular Music: History,
Place and Time (Aldershot: Ashgate), pp. 128–37.
274 Music and Shape
Butler, M. J., 2006: Unlocking the Groove: Rhythm, Meter and Musical Design in Electronic
Dance Music (Bloomington: Indiana University Press).
Campbell, M., C. Greated and A. Myers, 2004: Musical Instruments: History, Technology
and Performance of Instruments of Western Music (Oxford: Oxford University Press).
Canby, E. T., 1956: ‘The sound man-artist’, Audio 44–5: 60–1.
Cohen, A. J., 2009: ‘Music in performance arts: film, theatre and dance’, in S. Hallam, I.
Cross and M. Thaut, eds., The Oxford Handbook of Music Psychology (Oxford: Oxford
University Press), pp. 441–51.
Collins, N., 2007: ‘Live electronic music’, in N. Collins and J. d’Escriván, eds., The Cambridge
Companion to Electronic Music (Cambridge: Cambridge University Press), pp. 38–54.
Collins, N., A. McLean, J. Rohrhuber and A. Ward, 2003: ‘Live coding in laptop perfor-
mance’, Organised Sound 8/3: 321–30.
Cook, N., E. Clarke, D. Leech-Wilkinson and J. Rink, eds., 2009: The Cambridge Companion
to Recorded Music (Cambridge: Cambridge University Press).
Davidson, J. W., 1993: ‘Visual perception of performance manner in the movements of solo
musicians’, Psychology of Music 21: 103–13.
Davidson, J. W., 1994: ‘What type of information is conveyed by the body movements of
solo musician performers?’, Journal of Human Movement Studies 6: 279–301.
Davidson, J. W., 2006: ‘ “She’s the One”: multiple functions of body movement in a stage
performance by Robbie Williams’, in A. Gritten and E. King, eds., Music and Gesture
(Aldershot: Ashgate), pp. 208–26.
Davidson, J. W. and S. Malloch, 2009: ‘Musical communication: the body movements of per-
formance’, in S. Malloch and C. Trevarthen, eds., Communicative Musicality: Exploring
the Basis of Human Companionship (Oxford: Oxford University Press), pp. 565–84.
Davis, R. and S. Parker, 2013: ‘Creativity and communities of practice: music technology
courses as a gateway to the industry’, paper presented at the Proceedings of the Audio
Engineering Society 50th International Conference, Murfreesboro, TN, USA, 25–27
July 2013.
Doğantan-Dack, M., ed., 2008: Recorded Music: Philosophical and Critical Reflections
(London: Middlesex University Press).
Fast, S., 2006: ‘Popular music performance and cultural memory. Queen: Live Aid,
Wembley Stadium, London, July 13, 1985’, in I. Inglis, ed., Performance and Popular
Music: History, Place and Time (Aldershot: Ashgate), pp. 138–54.
Frith, S., 1981: Sound Effects: Youth, Leisure, and the Politics of Rock ’n’ Roll (New York:
Pantheon).
Frith, S., 2001: ‘Pop music’, in S. Frith, W. Straw and J. Street, eds., The Cambridge
Companion to Pop and Rock (Cambridge: Cambridge University Press), pp. 93–108.
Frith, S. and A. Goodwin, eds. 1990: On Record: Rock, Pop and the Written Word (London:
Routledge).
Frith, S. and S. Zagorski-Thomas, eds., 2012: The Art of Record Production: An Introductory
Reader for a New Academic Field (Farnham: Ashgate).
Frith, S., W. Straw and J. Street, eds., 2001: The Cambridge Companion to Pop and Rock
(Cambridge: Cambridge University Press).
Gander, J., 2011: ‘Performing music production: creating music product’ (PhD dissertation,
King’s College London, Department of Culture, Media and Creative Industries). Available
at https://kclpure.kcl.ac.uk/portal/files/5341102/Gander_PhD.pdf (accessed 9 April 2017).
Shaping popular music 275
Gibson, D., 2005: The Art of Mixing: A Visual Guide to Recording, Engineering, and Production,
2nd edn (Boston: Thompson Course Technology PTR).
Goodwin, A., 1990: ‘Sample and hold: pop music in the digital age of reproduction’, in
S. Frith and A. Goodwin, eds., On Record: Rock, Pop and the Written Word (London:
Routledge), pp. 258–76.
Gracyk, T., 1996: Rhythm and Noise: Aesthetics of Rock (London: Tauris).
Greasley, A. and H. M. Prior, 2013: ‘Mix tapes and turntablism: DJs’ perspectives on musi-
cal shape’, Empirical Musicology Review 8/1: 23–43.
Greasley, A. E., A. Lamont and J. A. Sloboda, 2013: ‘Exploring musical preferences: an
in-depth study of adults’ liking for music in their personal collections’, Qualitative
Research in Psychology 10/4: 402–27.
Greig, D., 2009: ‘Performing for (and against) the microphone’, in N. Cook, E. Clarke,
D. Leech-Wilkinson and J. Rink, eds., The Cambridge Companion to Recorded Music
(Cambridge: Cambridge University Press), pp. 16–29.
Griffiths, N. K., 2010: ‘ “Posh music should equal posh dress”: an investigation into the
concert dress and physical appearance of female soloists’, Psychology of Music 38/2:
159–77.
Hansen, K. F., 2010: ‘The acoustics and performance of DJ scratching: analysis and mod-
elling’ (PhD thesis, University of Stockholm). Available at http://www.speech.kth.se/
~kjetil/thesis (accessed 9 April 2017).
Hennion, A., 1989: ‘An intermediary between production and consumption: the producer
of popular music’, Science, Technology and Human Values 14/4: 400–24.
Horning, S. S., 2002: ‘Chasing sound: the culture and sound of recording studios in America
1877–1977’ (PhD dissertation, Case Western Reserve University).
Horning, S. S., 2004: ‘Engineering the performance: performance engineers, tacit knowl-
edge and the art of controlling sound’, Studies of Social Science 34: 703–31.
Horning, S. S., 2012: ‘The sounds of space: studio as instrument in the era of high fidelity’,
in S. Frith and S. Zagorski-Thomas, eds., The Art of Record Production: An Introductory
Reader for a New Academic Field (Farnham: Ashgate), pp. 29–42.
Hsieh, H.-F. and S. E. Shannon, 2005: ‘Three approaches to qualitative content analysis’,
Qualitative Health Research 15/9: 1277–88.
Hughes, T., 2006: ‘Nirvana: University of Washington, Seattle, January 6, 1990’, in I. Inglis,
ed., Performance and Popular Music: History, Place and Time (Aldershot: Ashgate),
pp. 155–71.
Inglis, I., ed., 2006: Performance and Popular Music: History, Place and Time (Aldershot:
Ashgate).
Jarrett, M., 2012: ‘The self-effacing producer: absence summons presence’, in S. Frith and
S. Zagorski-Thomas, eds., The Art of Record Production: An Introductory Reader for a
New Academic Field (Farnham: Ashgate), pp. 129–48.
Kania, A., 2008: ‘Works, recordings, performances: classical, rock, jazz’, in M. Doğantan-
Dack, ed., Recorded Music: Philosophical and Critical Reflections (London: Middlesex
University Press), pp. 3–21.
Katz, M., 2004: Capturing Sound: How Technology Has Changed Music (Berkeley: University
of California Press).
Katz, M., 2012: Groove Music: The Art and Culture of the Hip-hop DJ (New York: Oxford
University Press).
276 Music and Shape
Kooijman, J., 2006: ‘Michael Jackson: Motown 25, Pasadena Civic Auditorium, March
25, 1983’, in I. Inglis, ed., Performance and Popular Music: History, Place and Time
(Aldershot: Ashgate), pp. 119–27.
Krahé, C., U. Hahn and K. Whitney, 2015: ‘Is seeing (musical) believing? The eye versus
the ear in emotional responses to music’, Psychology of Music 43: 140–8.
Leech-Wilkinson, D. and H. M. Prior, 2014: ‘Heuristics for musical expression’, in
D. Fabian, E. Schubert and R. Timmers, eds., Expressiveness in Music Performance:
Empirical Approaches Across Styles and Cultures (Oxford: Oxford University Press),
pp. 34–57.
Lees, G., 1987: Singers and the Song (New York: Oxford University Press).
Massey, H., 2000: Behind the Glass: Top Record Producers Tell How They Craft the Hits
(San Francisco: Backbeat Books).
McIntyre, P., 2012: ‘Rethinking creativity: record production and the systems model’, in
S. Frith and S. Zagorski-Thomas, eds., The Art of Record Production: An Introductory
Reader for a New Academic Field (Farnham: Ashgate), pp. 149–61.
Middleton, R., 1995: ‘The “problem” of popular music’, in idem, Musical Belongings:
Selected Essays (Farnham: Ashgate, 2009), pp. 75–88.
Montano, E., 2010: ‘How do you know he’s not playing Pac-Man while he’s supposed to
be DJing? Technology, formats and the digital future of DJ culture’, Popular Music 29:
397–416.
Moore, A., 2001: The Primary Text: Developing a Musicology of Rock, 2nd edn (Farnham:
Ashgate).
Moorefield, V., 2005: The Producer as Composer: Shaping the Sounds of Popular Music
(Cambridge, MA: MIT Press).
Moorefield, V., 2010: ‘Modes of appropriation: covers, remixes and mash-ups in contem-
porary popular music’, in A. Bayley, ed., Recorded Music: Performance, Culture and
Technology (Cambridge: Cambridge University Press), pp. 291–306.
Moylan, W., 2007: Understanding and Crafting the Mix: The Art of Recording, 2nd edn
(Oxford: Focal).
Negus, K., 1992: Producing Pop: Culture and Conflict in the Popular Music Industry (London:
Arnold).
Negus, K., 1996: Popular Music in Theory: An Introduction (Cambridge: Polity).
Poschardt, U., 1998: DJ Culture (London: Quarter Books).
Poss, R. M., 1998: ‘Distortion is truth’, Leonardo Music Journal 8: 45–8.
Pras, A. and C. Guastavino, 2011: ‘The role of music producers and sound engineers in the
current recording context, as perceived by young professionals’, Musicae Scientiae
15/1: 73–95.
Prior, H. M., 2010: ‘Links between music and shape: style-specific; language-specific; or uni-
versal?’, paper presented at Topics in Musical Universals: 1st International Colloquium,
University of Provence, Aix-en-Provence, France December 2010.
Prior, H. M., 2012a: ‘Report for interview participants’, http://www.cmpcp.ac.uk/
Report%20for%20interview%20participants.pdf (accessed 9 April 2017).
Prior, H. M., 2012b: ‘Shaping music in performance: report for questionnaire participants
(revised August 2012)’, http://www.cmpcp.ac.uk/wp- content/uploads/2015/09/Prior_
Report.pdf (accessed 9 April 2017).
Shaping popular music 277
Scott, D. B., ed., 2009: The Ashgate Research Companion to Popular Musicology (Farnham:
Ashgate).
Shuker, R., 2013: Understanding Popular Music Culture, 4th edn (Routledge: Abingdon).
Smith, S., 2013: Hip-hop, Turntablism, Creativity and Collaboration (Farnham: Ashgate).
Straw, W., 1993: ‘The booth, the floor and the wall: dance music and the fear of falling’,
Public 8: 169– 82. Available at http://pi.library.yorku.ca/ojs/index.php/public/article/
viewFile/30160/27715 (accessed 9 April 2017).
Szwed, J., 2002: So What: The Life of Miles Davis (London: Heinemann).
Tagg, P., 1982: ‘Analysing popular music: theory, method and practice’, Popular Music
2: 37–65.
Tan, S., A. J. Cohen, S. D. Lipscomb and R. A. Kendall, eds., 2013: The Psychology of
Music in Multimedia (Oxford: Oxford University Press).
Théberge, P., 1989: ‘The “sound” of music: technological rationalisation and the produc-
tion of popular music’, New Formations 8: 99–111.
Théberge, P., 1997: Any Sound You Can Imagine: Making Music/Consuming Technology
(Hanover, NH: Wesleyan University Press).
Théberge, P., 2001: ‘ “Plugged in”: technology and popular music’, in S. Frith, W. Straw
and J. Street, eds., The Cambridge Companion to Pop and Rock (Cambridge: Cambridge
University Press), pp. 3–25.
Théberge, P., 2012: ‘The end of the world as we know it: the changing role of the stu-
dio in the age of the Internet’, in S. Frith and S. Zagorski-Thomas., eds., The Art of
Record Production: An Introductory Reader for a New Academic Field (Farnham: Ashgate),
pp. 77–90.
Thompson, D., 2010: Wall of Pain: The Life of Phil Spector (London: Omnibus Press).
Toynbee, J., 2000: Making Popular Music: Musicians, Creativity and Institutions (New York:
Oxford University Press).
Tsay, C.-J., 2013: ‘Sight over sound in the judgment of music performance’, Proceedings of the
National Academy of Sciences 110/36: 14580–5.
Walser, R., 1993: Running with the Devil: Power, Gender and Madness in Heavy Metal Music
(Middletown, CT: Wesleyan University Press).
Warner, T., 2003: Pop Music: Technology and Creativity—Trevor Horn and the Digital
Revolution (Aldershot: Ashgate).
Williams, A., 2012: ‘ “I’m not hearing what you’re hearing”: the conflict and connection
of headphone mixes and multiple audioscapes’, in S. Frith and S. Zagorski-Thomas,
eds., The Art of Record Production: An Introductory Reader for a New Academic Field
(Farnham: Ashgate), pp. 113–27.
Zagorski-Thomas, S., 2012: ‘The US vs. the UK sound: meaning in music production in
the 1970s’, in S. Frith and S. Zagorski-Thomas, eds., The Art of Record Production: An
Introductory Reader for a New Academic Field (Farnham: Ashgate), pp. 57–76.
Zak, A., 2009: ‘Getting sounds: the art of sound engineering’, in N. Cook, E. Clarke,
D. Leech-Wilkinson and J. Rink, eds., The Cambridge Companion to Recorded Music
(Cambridge: Cambridge University Press), pp. 63–76.
Reflection
Steven Savage, record producer and sound engineer
Creating unnatural shapes
When I was asked to write about music and shape for this volume I immedi-
ately thought of the reverb programmes that I use to add ambience to indi-
vidual tracks when I am mixing. Reverb presets often come in the form of
representations of physical space. General categories might include stadiums,
concert halls, churches, theatres, auditoriums, nightclubs, small rooms, etc.
Today’s sampling reverbs, which can translate specific acoustical spaces into
ambiences that can be used on any sound, include such presets as the Sydney
Opera House, St Paul’s Cathedral or the Ryman Auditorium at The Grand Ole
Opry, as well as less renowned, smaller spaces such as a closet, a tiled bathroom
or the interior of a Ford Econoline van. Some programs simulate very specific
types of spaces such as a medium concert hall with stage, which recreates the
effects of a typical concert hall including the anomalies created by the stage
area. Every acoustical space consists of shapes that inform our ear and create
mental images of environments.
How does all this inform my work as a professional music-mixer? One of
my primary tasks as a mixer is the creation of these acoustical environments.
Although some recordists are careful to create realistic environments (primarily
in the western art music world), those of us dealing in popular music tend to
create ‘impossible’ environments. They may be impossible simply because var-
ious elements are in different spaces (the guitar is in a small wood room while
the vocalist is in a concert hall); but they may also be impossible because we
can create ambiences that don’t actually exist in nature—a truncated reverb or
a ‘perfect’ digital delay, for example. Figure R.25 illustrates one such imaginary
acoustic environment in which a three-dimensional model represents the var-
ious qualities produced by spatial constructions that are a part of the mixing
process. The various blocks represent frequency content based on their height
278
Reflection: Steven Savage 279
Frequency
Ambience
Pan
FIGURE R.25 Three-dimensional mixing metaphor. This diagram indicates one way in which shapes
may be used to represent essential conceptual approaches to building audio mixes such as frequency
balance, panning position and the sense of depth created by the addition of ambiences.
(Figure credit: Iain Fergusson)
Shapes seen
Reflection
Mark Applebaum, graphic composer
I wish to point out, however, that it runs parallel with—it has not replaced—
an abiding interest in composing music in which I first pre-hear the result, as
well as the manufacture of scores whose symbols are assiduously defined (by
me or through communion with a common practice, both ancient and recent).
Instead, my point here is that, although the premise for The Metaphysics
of Notation is arguably unconventional (or at least not mainstream), it has
qualities that are common to all music. To wit, my determinate, nonpic-
tographic pieces, and most music for that matter, are also concerned with
form—through their shapes, geometries and contours. The composer Roger
Reynolds used to extol the idea of musical ‘profile’ considered in various
parameters. In lessons he would note the blandness of a particular ampli-
tude profile or the iconicity of a noteworthy rhythmic profile. The word ‘pro-
file’ immediately conjures visual imagery and ideas about shape. Pinocchio
has an especially memorable profile. By extension, one might consider the
‘silhouette’ of Mickey Mouse. Not only is it well known as a successful com-
mercial meme, but it is arguably a shape of intrinsic distinction, and this
gives it the kind of memorable quality that is often desirable to the composer
of musical material. Reynolds’ use of visual analogy is telling but not new.
One recalls the now esoteric Schillinger System of Musical Composition,
which allows a city skyline to be employed as a melodic template. And pic-
tographic notation goes back at least to the ars subtilior composers of the
fourteenth century.
Although I chose not to hear sound when composing the Metaphysics, I was
often conscious of the deliberate analogy between notational shape and musi-
cal discourse. There are two corresponding concerns that I will explore. First
is the idea of devices used for rhetorical development. Second is the idea of
large-scale formal connectivity.
Consider panel 4 of Figure R.26. At its left side a shield appears. (For ease of
expression I’m calling it a ‘shield’; but I think of it equally as ‘a shield-looking
thing that, to the broad-minded interpreter, may or may not evoke shieldness.’
The point is that an interpreter should not be limited by my verbal descrip-
tions.) Although partially obscured behind the first shield, a hook or letter J
appears to rotate from one shield to the next, thus implying (or again, more
accurately, ‘potentially inviting an inference of ’) inversion (Figure R.27).
The incremental clocklike advance— as opposed to a sudden 180- degree
flip—may intimate either some kind of slow motion melodic inversion or the
inversion of a chord from root position to first inversion to second inversion,
etc. Meanwhile, the shields descend in the vertical dimension, thereby sug-
gesting transposition or sequence. Their descent by equal quanta evokes a
FIGURE R.26 The Metaphysics of Notation, panel 4
286 Music and Shape
Returning to panel 4’s wave of circles (see Figure R.28), we see a counterpoint
emerging underneath in the form of the upward sloping diagonal (potentially
a portamento) comprising many small details (reminiscent of ornamentation,
decoration). Unexpected, irregular (syncopated) bits of varying length extend
above or dangle below the slope, thus suggesting accents, chordal congruencies
or multiphonics. This slope feels fundamentally independent of the circles (it is
both contrasting and non-accompanimental), thus establishing polyphony: an
expansion of voices and a richer texture. And the languages of the two voices
are so dissimilar in personality that one envisions Ivesian simultaneity or the
character patterns of Carter.
Just before the slope disappears, two new glyphs appear above it: a small cir-
cle and a small oval, both black. The circle echoes the genesis of the sinusoidal
wave of circles that have since evolved to their mature state of augmentation
and dematerialization. The oval is its squashed permutation, a kind of thematic
metamorphosis or, in its simplest sense, a variation. The circle and the oval are
far enough apart that they might appear atomic, isolated—or, to use Cage’s
language, unimpeded. But, because they are connected by slender lines, we are
compelled to see mutual belonging, a molecular constellation. Cage would call
them interpenetrated, and their connection affects how we understand them
and, presumably, how we might play them.
The downward glissando of materializing rectangles in panel 5 seems to ter-
minate in a point, a seed that grows into a flower (Figure R.31). This consti-
tutes a striking change, the sudden presentation of contrasting material soon
followed by more idiosyncratic references (e.g. a bell, an apple, a telephone).
Arguably, these materials possess stronger, more concrete cultural associations
than their more geometrically platonic neighbours. This might parallel an act
of musical quotation, or perhaps natural mimesis like birdcall in Beethoven or
Messiaen. I’m at a loss to suggest additional meaning for these icons. But by
now the reader is probably able to play this game without my help.
A vertical stripe of decorative embellishments appears next. If it seems
familiar, it is because it constitutes a reappearance of the irregular, syncopated
bits that extended above or dangled below the slope in panel 4. As such, it rep-
resents motivic recurrence, something that could arouse an emotional affect:
after the appearance of many contrasting novelties the return to the familiar
could be felt as a welcome tonic, economical relief, or perhaps even wistful
nostalgia. At the same time, the stripe has changed: it is elaborated by a bul-
bous bottom and contextualized by a heart shape. Taken together, my daughter
instantly identified a ‘heart guitar’. So perhaps a lyric song and instrumenta-
tion change are in order at this point.
To the right of the heart extends a series of dots arranged in two paral-
lel rows. The dots embody syncopated repetition. If you look carefully, you
will notice that the rows contain identical proportions; they are just tempo-
rally displaced as in imitation or canon. I chose dots as a deliberate homage to
FIGURE R.31 The Metaphysics of Notation, panel 5 close-up: contrasting materials, ‘heart guitar’ and canonic dots
292 Music and Shape
Conlon Nancarrow’s temporal canons for player piano, their rolls methodically
punched with holes just so.
Above these piano-roll holes appear odd stalagmites crowned with unique
figuration. They contrast with the limited vocabulary found in the dangling
mobiles underneath, whose sundry angles (one looks like a hockey stick) are
simply axial inversions, retrogrades and retrograde inversions of one another.
Having commented on most of two panels, I will end my analogic exegesis
here. A more thorough evaluation is certainly possible, but this will suffice as an
introduction to the manner in which visual shapes can be considered analogous
to traditional musical devices.
FIGURE R.33 The Metaphysics of Notation, close-up: circle and oval pair inverted across panels 3
and 4
Inversion can also be seen earlier in panel 3 at the bottom, where a kind of
scroll shape adorned with the number 5 is inverted onto the upper part of panel
4 (Figure R.34). The shading of the latter scroll is altered, thereby suggesting
a kind of mode change. Such mutation anticipates other variances: first, the
scroll outline inverts though the number 5 within it repeats without alteration;
but more significantly, eight ‘ribbons’ extend from the panel 3 scroll and five
appear in the panel 4 scroll, only three of which are in common as inverted
reverberations of one another.
The logic continues across panel 4 to panel 5, first by way of two inverted
shields, and then by the aforementioned ‘heart guitar’, which lines up precisely
with the first bit of dangling embellishment in panel 4 (Figure R.35).
Then panels 5 and 6 are conjoined in the vertical plane by one of the dan-
gling angles—a hockey-stick-looking doodle—that points directly to its double
mirror image (it is flipped both horizontally and vertically; Figure R.36).
Continuing downward, we see panels 6 and 7 linked by two pathways (Figure
R.36). First, a vertical chain of circles, themselves sequentially augmented in
size, is seen in flipped form on the other side of the divide. And later in panel 6
there appear twelve equal-sized dots arranged in a ring, like points on a clock.
FIGURE R.34 The Metaphysics of Notation, close-up: ‘scroll’ with number five inverted across panels
3 and 4
FIGURE R.35 The Metaphysics of Notation, close-up: panels 4 and 5 inverted shields, connection to the ‘heart guitar’
Reflection: Mark Applebaum 297
FIGURE R.36 The Metaphysics of Notation, close-up: panels 5, 6 and 7 dangling angles, chain of
circles, dot clock
panel 4’s future. So the vertical logic confounds the horizontal and vice versa.
This structural superfluity purposefully forces the interpreter to choose among
temporal constructs, or to ignore them entirely in favour of a different strategy
for harmonizing the inherent temporality of sound with the intrinsic stasis of
the drawn image.
Similarly a strange temporal puzzle occurs across panels 9 and 10 (Figure
R.37). Starting in the second half of panel 9, a series of tiny repeating dots
curls from the bottom of the page, loops anticlockwise and straightens into a
horizontal comportment where it grows in size, diminishes and finally vanishes
off the right edge of the page. Continuing horizontally onto panel 10, we see
that these dots are the genealogical progenitors of those that begin on the
panel’s left side; after all, they line up in the horizontal plane with panel 9.
The panel 10 dots grow in size, multiply into three larger circles and lead to a
series of waves constituted by various polygons and simple shapes. Eventually
a curl of small shapes emerges near the middle of the panel. It arcs around
clockwise, anticlockwise, and then disappears off the top of panel 10. But this
trail connects—perhaps begets—the aforementioned series of dots on panel 9.
So where is the origin of this infinite, recursive visual rhetoric? It is a paradox
of chronology that is evident in the visual domain but cannot be rationally
represented in the time—one could say the shape—of musical sound.
The composition is not, however, intended for these ‘professionals’. Its fanci-
ful, idiosyncratic curiosities are directed to more ‘abnormal’ players, often ones
who have overcome their conservatoire training. This breed is game for such
creative enterprise, a collective of musicians who, while indeed a minority, form
a remarkably expansive and extraordinarily enthusiastic community.4
But for both the inclined and the averse, my purpose here is simply to
recognize the kinship that this kind of artistic adventure has with tradi-
tional compositional devices. Josquin, Bach and Schoenberg use retrograde;
so do I. Counterpoint can be heard in the music of Palestrina, Brahms and
Ferneyhough; and while it may or may not be heard in Metaphysics (its sounds
are left to each interpreter), it can be seen clearly there. My score and those
of Frescobaldi, Beethoven and Messiaen employ augmentation. Palestrina,
Haydn, Wagner and I are concerned with cadences. Sequence is common to
Du Fay, Mozart, Chopin and my score. The Metaphysics of Notation didn’t
invent the canon; it is found in Ockeghem, Monteverdi and Nancarrow. My
point is that it is worthwhile to note the commonality among our composi-
tional tools, not only the obvious contrasts in notational vocabulary. And there
is another contention here: in all contexts, these compositional tools imply and
embody shape.
At the same time, I cannot stress enough that the aforementioned observa-
tions need not direct an interpreter. The project of the Metaphysics includes the
hope—abundantly fulfilled—that I would experience utterly novel and unex-
pected interpretative solutions to the work’s peculiar challenges. These wide-
ranging outcomes were mainly the consequence of the breathtaking scope of
the players’ imaginations. But I believe that they were also aided by my cautious
avoidance of providing hints. (It is a piece that I have vowed never to perform,
precisely in the hope of not suggesting authorial precedent.) For example, the
best performers do not assume that the score must be read from left to right
(even if my aforementioned description of, say, retrograde relies on such a con-
ception), or top to bottom, or even in a single direction. For that matter, it
doesn’t have to be read linearly at all; some have chosen to interpret entire
panels as a single gestalt—much as a quaver is not read up or down or side to
side but is simply grasped wholly as an indivisible symbol. The score needn’t
be considered in its entirety: some players have set up fixed instrumentation in
front of a single panel as opposed to taking a peripatetic tour of all of them.
A realization can be improvised or carefully predetermined.5 And the score
could stimulate responses that are not even conventionally musical (it has, for
example, been interpreted by spoken word poets and dancers).
So I write this with a degree of wariness, one alleviated mainly by an under-
standing that the intellectually intrepid, curious, creative musicians who are
attracted to Metaphysics will likely read this Reflection and simply ignore it;
they will accept the challenge of inventing their own solutions that are beyond
my limited conception. Cardew’s Treatise Handbook undertakes a seemingly
Reflection: Mark Applebaum 301
Tonality in soundscapes
on the listener’s ears to accumulate the varying sounds. Line is the structure
that provides direction for the listeners, leading them to form an interpretation.
In both painting and music, colour provides energy and excitement. In
music, timbres arising from varying combinations of instruments and pitch
registers are analogous to the different paint colours. The orchestration of the
music is equivalent to the paint palette. As in painting, music may rely on a
well-demarcated contrast between the primary and complementary colours or
may be based on gradual variations, corresponding to the sfumato technique in
painting in which outlines are blurred by blending one tone into another. Both
techniques may have a powerful impact on the emotions conveyed. A skilful
composer arranges and combines all twelve notes together as a painter mixes
colours of paint. As with painting, mixing multiple colours together may result
in a greyish hue. However, this is not necessarily an undesirable result since a
greyish homogeneity forms a quiet background against which other voices or
colours are more emphasized and appear more luminous. For example, Marie
Laurencin’s portraits and the Symphonies Nos. 3 and 4 of Lutosławski both
use this concept of a grey canvas. In my own painting, rather than relying on
premixed complementary colours, I manipulate a few chosen hues of the pri-
mary colours, combining warm and cool variations with each other to form my
complementary colours, sometimes on the palette but, at other times, directly
on the paper or canvas.
Morphology, or shape, in music is formed by rhythms, repeating motives
and varying metres. The technique of counterpoint has been used for centuries
but is still one of the best examples to help understand the concept of musi-
cal morphology. The largest-scale classification for morphology consists of the
basic music forms, such as sonata and dance forms. Relying on simple shapes
within a painting is one method for facilitating the clear manifestation of a
subject. Paul Klee was a maestro in the application of simple shapes. Since he
had a background in music, his paintings provide an excellent example of the
relationship between music and the visual arts.
Space and volume consist of the three-dimensional content of music and
art. In painting, the three-dimensional perspective of a visual art is quite
intuitive. However, when I studied music composition, my teachers also
taught me to design the music with multiple layers, including a foreground,
middle ground and background. The presence of multiple layers is analogous
to perspective. In addition, the rests in music are analogous to the empty
spaces in a painting. The spaces and rests divide the artwork into important
sections and help the viewer or listener understand the structure as a whole.
Volume in the music is most often considered in terms of dynamic change,
piano versus forte. However, it also relates to the density of music, ranging
from minimal to rapid changes. In painting, volume refers to the degree of
contrast between the lightest and darkest regions. Contrast helps the viewer
focus on the composition.
304 Music and Shape
(a) (b)
FIGURE R.38 Watercolour paintings: (a) Red and White and (b) Fireworks
Reflection: I-Uen Wang Hwang 305
space of the painting. Just as the articulation and textures combine to create a
painting, the two piano parts merge naturally to form a single musical image.
The first movement, ‘Red and White’, is slow and tranquil as conveyed by its
namesake painting. Piano 1 and Piano 2 represent the red and white flowers in
the painting, respectively. The delicate flower petals inspire the elegance of the
music. The timing of the interactions of the two piano parts is analogous to the
three-dimensional spatial arrangement of the flowers.
The second movement, ‘Fireworks’, is energetic and lively, expressing the
transparent quality of the watercolour painting in which the layers can be seen
through each other. To achieve this effect, each piano has its individual set
of rhythms, as if representing different firework effects combining to create a
magnificent display.
When I create my art, I do not formally or overtly consider each of the con-
cepts discussed here. This would run the risk of creating music that is too ‘stiff’.
However, as one paints and composes simultaneously, the concepts naturally
and subconsciously arise during the artistic process.
9
What is synaesthesia?
The first recorded case of synaesthesia was that of Georg Sachs in 1812, who
briefly mentioned colour associations to music in addition to a more detailed
account of grapheme–colour synaesthesia (Jewanski, Day and Ward 2009).
However, the earliest detailed account of synaesthetic experiences to music
was that of the Nussbaumer brothers in 1873. Nussbaumer describes how
he, and his brother, experienced colours for different tones. Spatial aspects
(including shape) of the synaesthesia were not described in detail, but the
colour experiences were noted to be ‘inside his head, where it starts at the
temple as a band of colors, which go to the middle of the forehead’ (Jewanski
et al. 2011: 297). In the same year, Lussana (1873) describes colour experi-
ences triggered by the human voice. However, it was not until the 1880s that
multiple cases were documented and contrasted (Bleuler and Lehmann 1881),
and it was in this decade that the term ‘coloured hearing’ took root to describe
synaesthesia more generally.
Colour associations to music tend to be linked either to the pitch class (C, D,
F♯, etc.) including musical keys, or to the pitch height. For those in whom col-
our is linked to pitch class, this is invariably linked to the possession of perfect/
absolute pitch. In these cases, the same note in different octaves is linked to the
same colour (e.g. Carroll and Greenberg 1961; Rogers 1987). In some instances
Music and shape in synaesthesia 309
the colours appear to be derived from the letter-names belonging to the notes
themselves: so, for example, both the letter ‘A’ in a passage of text and the note
‘A’ in a passage of music may be perceived as red. Note that in this case the red-
ness of the note derives from its culturally assigned label, not from any quality
of the sound. For those in whom colour is linked to pitch height, sounds of a
similar pitch height tend to have similar colours, but those that are more distant
(including the same note an octave apart) tend to be less similar. In these cases,
there is a general tendency for notes that are higher in pitch to be lighter in col-
our (more yellow, for example; Ward, Huckstep and Tsakanikos 2006). That
is, although synaesthetes differ greatly in their precise choice of colours, there
tend to be commonalities between synaesthetes (and, indeed, nonsynaesthetes).
Aside from pitch, the colour of musical notes may be influenced by timbre
(ibid.) or, in a few cases, solely determined by timbre (Day 2009). Famous cases
of music–colour synaesthesia include Messiaen (Bernard 1986) and, possibly,
Scriabin (Myers 1915).
Visual experiences aside from colour have been less well documented in
the literature. Zigler (1930) noted synaesthetic shapes associated with instru-
ments (colours were noted too). To document this, a variety of instruments
were brought to the laboratory and played at three loudness levels, in differ-
ent durations (staccato, sustained) and using various pitches. The synaesthetic
shapes had a three-dimensional and dynamic quality to them. The basic shape
tended to be determined more by the instrument than by other characteristics
(loudness, pitch, duration). It was also noted that the two synaesthetes tested
differed greatly in how the shapes were described (see Table 9.1). Increasing the
duration of a note led to elongation of the shape or duplication of the form
(e.g. spherical forms passing across the field of vision in a definite direction).
For instance, one synaesthete saw the forms pass from left to right starting from
two feet away from the face and ending much further away. In orchestral pieces,
the perceived shapes tend to reflect a small number of instruments that are
attended to (so the shapes of one instrument gain prominence). When listening
to a duet, one synaesthete stated that ‘both instruments of a duet united in such
a way that the larger form occupies a lower position and appears to support the
smaller one with which it is connected by a small thread-like projection’.
Karwoski and Odbert (1938) also emphasized aspects of vision other than
colour by requiring their participants to draw or describe visual experiences to
music, including whole pieces as well as isolated notes. Some of their synaes-
thetes reported rather localized visualizations such as ‘cloud, film or veil effects
which may billow vaguely’, or ‘points or limited areas of colour, which may
move or expand or contract with changes in the music’. These can be contrasted
with other synaesthetes who reported ‘developing bands or ribbons of colour’
that ‘almost always show a left-right development’ (ibid.: 9). These could be
single bands of colour that change colour or move with the pitch, or multiple
bands linked to different instruments, rising and falling with changes in pitch.
For such synaesthetes there are, apparently, shapes within shapes (the overall
shape of the melody, and the shape of particular instruments and notes).
In the contemporary literature, Mills, Boteler and Larcombe (2003) docu-
ment the shapes and colours of music and notes in a synaesthetic student (GS).
The synaesthesia operates at multiple levels such that individual notes have a
colour and shape, but longer sections of music exist as ‘landscapes’ or ‘maps’
of coloured lines and blocks: ‘It’s easier actually for me to read my colour map
than to look at the music, because I look at the map I came up with and it’s
kind of like a gauge of how it’s going to sound. Whereas I look at the music
and I can’t always tell what it’s going to sound like’ (ibid.: 1363). The tempo
of the music determines the dynamic changes within the landscape. The pitch
of a note determines its size (high pitch is smaller) and position (high pitch is
higher). Shape is strongly related to timbre, with similar instruments eliciting
similar shapes, and, to a lesser degree, to pitch.
Although synaesthetic experiences are vivid and highly variable, it has been
noted for many years that there are certain ‘rules’ by which sounds and vision
tend to be linked (Marks 1978). For instance, high pitches tend to be judged as
brighter, smaller, higher in position and less rounded in shape (e.g. Marks 2004).
These have recently been termed ‘correspondences’ and their relationship with
synaesthesia itself is debated (Spence 2011). One account suggests that syn-
aesthesia exists on a continuum from strong to weak, in which ‘weak synaes-
thesia’ is essentially normative experience (Martino and Marks 2001). Others
have argued that these correspondences provide a starting-kit for linking sen-
sory dimensions but with synaesthesia being a qualitatively (rather than quan-
titatively) different developmental outcome (Ward et al. 2006). It is certainly
the case that many of these correspondences can be observed in the first few
Music and shape in synaesthesia 311
Sinha (2009: 200–2) describes how she uses her synaesthesia to analyse music
structurally and decompose it into parts (multiple voices):
I have always had the ability to see the form and internal structure of a
piece of music on my internal screen like a chart or a three-dimensional
film… Even without being able to read music I can always tell precisely
from the beam of sound how a piece is constructed, and I always know
exactly where we are within it. The sound beam always comes towards
me at an angle from the right; just in front of me it makes a slight turn
and moves away to the left. (This happens even if the sound source is
located on my left…)… The first few times I heard the song ‘Walking
Down the Street’, without noticing the details, I saw a thick but not
particularly solid beam of sound. It can be most readily compared to
a thick rope formed of a combination of different materials (but not
twisted like a cord)… As soon as I turned my attention to listening to
each voice individually, the beam began to unravel, until the five vocal
parts and the sixth for the rhythm separated out quite clearly and dis-
tinctly… The ray of sound belonging to each one can be described
precisely.
Bass: consists of very long, soft fibres, like mohair—very soft to the
touch.
Some of the themes that emerged above reoccurred during a recent interview
between the author (JW) and a synaesthete (RP), which is worth quoting at
length here, both for its inherent interest and because it covers so many aspects
of synaesthetic experience. RP is a native- German- speaking woman with
grapheme–colour and auditory–visual synaesthesia.
RP: It is like ‘just there’. I have to relax and close my eyes to be aware of
it. It is like being on a train and looking out and being aware of the
reflections on the window. Normally, you would not do this as you
would just look out of the window. But when you do that you see
that there is a second layer, with a different movement.
JW: As you listen to an extended piece of music do you see the part that
you are currently listening to, or do you see the part that you have
been listening to—like the history of the music?
RP: It would be like a path, like you see on Google Earth, so you could
zoom in and see the part that is just happening right now but you
could also zoom out and see, kind of like, the whole piece (like the
melody line)…
JW: But you don’t see ‘the future’, i.e. the music that you haven’t
yet heard?
RP: Except if I know the piece perhaps. That would mean I have seen it
before.
JW: How much level of control do you have, either in terms of zooming
in, or if it is a very complex piece of music I guess you could focus on
different instruments or is [what you see] very fixed?
RP: If there are certain instruments I could focus on one or another and
the rest would, like, go into the background—greyed out or faded out
a little bit. So you can bring something to the foreground… You can
concentrate and filter something out if you wish or you can look at
the whole.
JW: If we think about music at the very small scale, so single notes and
phrases, would these have shapes?
RP: Yes, it has and these are the pictures I sent you—a single note, a
human voice, humming. He does this technique of overtone singing
and this is the moment that the overtone arrives. [Figure 9.1]
JW: If you have an ascending tone then does it [the synaesthesia] go up
in similar ways?
RP: Yes, similar to the melody. It goes up and down according to the
melody.
JW: If you had, say, a flute and double bass how would this separate out
in the line? Would you see different elements?
RP: They would be on different parts of this inner space. It is three-
dimensional. It would be like two beams, or flows, or flying scarves.
The lower instruments would also be on a lower place in that room,
in that space.
JW: What determines what is close to you in the depth plane?
RP: I don’t know. Usually I am able to move around in that space and
take a look from several points of view. But it also may happen that
suddenly it sucks me in and it goes through me. Imagine a huge
(a)
(b)
(c)
FIGURE 9.1 RP’s synaesthetic experience to overtone singing by Wolfgang Saus. This singing
technique filters parts of tones from the overtone spectrum of the voice so that they are perceived as
individual and separate tones. RP describes her experiences as follows: ‘The first picture [Figure 9.1a]
is a snapshot—a frozen tone. A single humming or tone is very dynamic, carrying and expanding. It
has a middle axis that is floating free through space, moving up and down according to the melody.
The wing-like shapes are not wings but more or less radiating beams of energy. The second picture
[Figure 9.1b] is the magical moment when the overtone appears. It takes off like a phoenix. The
overtone sounds similar spherical like a glass harp and for me even looks similar, but there is even
more: he changes the sound-image of the humming sounds below and creates a new unity/common
form. Most times the notes move from left to right, but I can also go around them and view them from
above, below, front or back. The third picture [Figure 9.1c] shows the same scenario from the front.’
Colour versions of these images are available at the companion website .
316 Music and Shape
At the end of the interview, RP was asked whether there was anything else that
she wanted to add.
RP: Are you only talking about the visual stuff rather than what you feel
in your body?
JW: You mean the ‘chills’ or you mean other things?
RP: The chills would be positive but there would also be the opposite. An
example would be that I hate Whitney Houston, I cannot stand it. If
it is anywhere, I go out of the room. I cannot stand the way that the
voice feels in my body. It is like tooth pain but it is not in my teeth, it
is somewhere else, like electricity, because it is too perfect and if I had
to paint it, it would be like the metal chrome; a chrome-coloured
freshly polished car bumper. That is what Whitney Houston looks
like but it also like how it feels—perfect but cold—and I really cannot
stand her voice.
Music and shape in synaesthesia 317
DISCUSSION
There are several general points emerging from the interview that are echoed
elsewhere. First, the experiences themselves are very dynamic: changes in the
synaesthetic visions reflect changes in the temporal characteristics of the music.
This operates at multiple levels, from notes to single voices/instruments within
a more complex piece, to the piece as a whole. To some extent, the synaesthete
can attend to these levels and see the structure of the music come apart.
The interview with RP, and comments of other synaesthetes, also suggest
that their synaesthesia can act as an internal ‘map’ of the music that, to some
degree, makes an external representation of the music (a written score) less
important. There are certain structural similarities between typical synaesthetic
visualizations and western musical notation, including high pitch being higher,
and a tendency to read from left to right. However, in other ways they differ
(e.g. low pitch being bigger and darker), and it may often be the case that these
two musical systems (symbolic, synaesthetic) are not completely aligned.
Finally, the synaesthetic shapes of music typically are not just two-dimen-
sional but have solidity and texture that lends them an almost tactile qual-
ity. They are probably not providing a true auditory–tactile synaesthesia, but
rather there is a sense of what the visualizations would feel like if touched.
One of Zigler’s (1930: 278) synaesthetes describes the surfaces of the shapes in
terms such as ‘hard, soft, velvety, jagged, uneven, smooth, polished, streaked,
crystalline’. Sinha (2009: 201) describes it thus: ‘I can always describe exactly
what it would feel like in the palm of my hand and on the inside of my fingers
if I were to stroke the beam of sound: soft, prickly, pleasant, warm, grassy/airy,
or solid.’
References
Ward, J., S. Moore, D. Thompson-Lake, S. Salih and B. Beck, 2008: ‘The aesthetic appeal
of auditory-visual synaesthetic perceptions in people without synaesthesia’, Perception
37: 1285–96.
Ward, J., J. Simner and V. Auyeung, 2005: ‘A comparison of lexical-gustatory and grapheme-
colour synaesthesia’, Cognitive Neuropsychology 22: 28–41.
Zigler, M. J., 1930: ‘Tone shapes: a novel type of synaesthesia’, Journal of General
Psychology 3: 277–86.
Reflection
Timothy B. Layden, synaesthete
For me every sound has its own shape or form. This sense of shape is like
objects in my periphery. They move around me and change in size and struc-
ture depending on how the sounds change. The experience is more intense
with more complex sounds and when the source of the sound is not visible.
I wonder if it is my brain creating the visual for the sound. The shapes seem to
reflect the sound: liquid sounds often create fluid bubbly shapes; sharp clanging
sounds have more angular shapes like growing crystals; bass sounds are large
and expanding. When there is a loud, seemingly singular sound, this can create
a sense of space around me as if I were inside the shape itself. When many
sounds occur at once, the shapes often combine, creating a complex structure
or a texture. These shapes sometimes blend together, rather as sounds do in the
environment, creating a moving landscape. These experiences are part of how
I sense the world and rarely stand out as distractions. Sometimes, however, a
sudden, unexpected sound will evoke a synaesthetic experience that is distract-
ing, drawing my attention away from whatever I might be doing.
There are certain sounds and combinations of sounds that I particularly
enjoy and seek out. I am a lover of experimental and improvised music, as these
can create unexpected and exciting shapes. I also enjoy the sounds of a busy
urban environment with its bells, screechings, fountains and many pounding
feet. I enjoy industrial sounds at a distance but not close up: when sounds are
too loud and incessant I can become overwhelmed. I love the early morning
calling of birds in spring; their shapes are like a garden of wildflowers.
As an artist I have used my experience of synaesthesia as a source for my
work. I record and create sounds that I use to make music and soundscapes
filled with shapes that I draw, paint, sculpt and write about. Through the pro-
cess of observing my experience of sound shapes, I have become more aware
of them, and the experience has become more detailed. I also feel that I have
become a better artist by using my synaesthesia in my work.
321
322 Music and Shape
FIGURE R.39 Timothy B. Layden, Dark Glistening. A colour version may be seen at .
Musicians are always talking about ‘shape’ in reference to phrasing, but mostly
I think they merely mean that something has shape rather than being ‘shape-
less’—always a derogatory term. Shape in this sense means direction, a start
and a finish with something pleasing in the middle. Seldom are the visual pat-
terns of a draughtsman or an artist relevant to a musician. Nevertheless, only
today I was doing an interview and trying to describe to the journalist a CD
I recently recorded of music by Scriabin and Janáček. Strange bedfellows, it
would seem, despite some parallels in their Slavic origins and their eccentric
visions. But what makes them so different from each other (and therefore fas-
cinating in juxtaposition) is actually related to shape. Scriabin is all seductive
curves (across the phrases and up through the exotic harmonies), whereas
Janáček is angular and fragmented, motives repeated and insisted on like an
army of elbows. Perfumed art nouveau versus bleak if passionate cubism.
323
Reflection
Alex Reuben, filmmaker
like the characters, are also dancing. The effect should be about as interesting
as watching paint dry, and yet audiences seem to like it. I think it’s the combi-
nation of the music and the clashing, bright red and pink that pulls us along,
engages us: it’s a conversation, mirrored by the performers. I composed the
soundtrack with a very simple, three-note phrase which repeats over and over
again—what cognitive scientists might describe as an ‘earworm’. I wanted the
film to be like a TV commercial and had written the music to be catchy. I also
used it in a radio advert. Of course, I’m not consciously thinking of any of this
as I make the films. I’m working instinctively and emotionally. It’s expressive
and I love doing it.
Afterwards I became interested in what was moving me and audiences. I’d
previously collected articles on evolution and began rationalizing what was
going on in my films, and this led me naturally towards research in cognition.4
I learned from the work of neuroscientist Chris Frith that we are far more
engaged in contagion, reacting to the actions and reactions of others, and
that we have less agency than we may like to believe, while the work of neuro
aesthetician Semir Zeki showed me that some of the most popular and engross-
ing paintings are those that carry aesthetic questions—dark spaces where the
viewer must fill in the gap. This taught me something about Que Pasa, that
audiences went on a narrative journey through moving shapes, even when noth-
ing appeared to be happening.
I wondered why, while I feel a very powerful response to space in architec-
ture and the countryside, I have such a strong, physical reaction to paintings.
So in Line Dance (2004)5 I used twenty-four-camera, motion-capture technol-
ogy to create 2D figures in 3D space, a virtual space. My ‘choreogeography’,
co-performed in rehearsal with Afua Aweku, is re-choreographed on computer
with a ‘stick-figure’ model, using data from real dancers. I instinctively felt that
the simple shapes could best reveal movement: I created them in a computer
with the help of skilled technicians. Like the dance, it’s a collaboration, a digi-
tal continuation of the original choreography and improvisation. This digital
choreography is the manipulation of shape and colour. It gives me much the
same sensation as when I am drawing with pencil and paper, an exciting rhythm
and feeling that runs throughout my body, a sense of ‘being in the moment’
or ‘zone’. I have this same feeling whether it’s me operating the pencil or com-
puter, or when working with skilled craftspeople as in this case.
The ‘projection’ also happens when I compose music. I’m not a trained
musician and cannot play an instrument. I compose it in my head and hum it
to musicians. I can ‘audiolize’ exactly what I want into a material sound just
as I visualized the figures. When I edit the soundtrack with pictures, I feel the
same sense of composition.
In Line Dance, the lines and colours build, layer and overlap in much the
same way as a Jackson Pollock painting. If you look at a Pollock canvas
close-up, rather than a reproduction—Autumn Rhythm (Number 30),6 for
326 Music and Shape
sunset, water, emotion, violence, sex. It might be any of the things that are in a
film, working at lots of levels. One of the things I’m trying to do is stretch the
‘now’ over the whole. And it almost feels like a dance itself, this relationship of
the whole and being in the moment.
When I teach sound and video projects to art students, something I empha-
size is texture. The relationship between texture in sound and image can be
found in shape. The easiest way to explain this is to illustrate how sound and
music have the equivalent of smooth and rough textures, like the surfaces of
three-dimensional materials; and then to say that there are as many variations
in sound qualities as there are in colours, and that screens have diverse, visu-
ally tactile properties, as in touch. They get it instantly. It’s then a small step
to discuss the relationship with technology and how music producers like Phil
Spector,9 Hank Shocklee (from ‘Public Enemy’)10 or Lee ‘Scratch’ Perry11 are
artists working in sound, musicians whose primary instrument is technology
and the recording studio. For me, in their recordings, the key element is the
texture and movement. That can move me as much as melody and rhythm.
In music and sound, production is almost like a feeling, a total whole when
I’m working with sound and image. I can’t distinguish them: it’s like painting
with sound. In editing there’s something quite visual in the pattern of film and
sound tracks; they make shapes that change according to the proportion of
your timeline. You can look at them on a big or a small scale, though when it
gets so big that I can’t see the overall form for long, I get uncomfortable. Once
I’ve done a focused bit of editing, I’ll always return to the longer scale so that
I’ve got a sense of proportion. It has to feel proportionate to something physi-
cal in my body, and in that sense it seems to have a shape that I feel.
I’ve expanded choreogeography from the one minute of Que Pasa into the
physical world around me. It’s very much to do with how I feel myself in the
landscape. I don’t just mean the physical landscape but also the landscape inside
our heads. I walk everywhere and have made the longer films with the same
sense of travel. I would put the graphic Que Pasa and Line Dance into that cat-
egory as well as Routes and Newsreel,12 which are shot quasi-realistically. The
former don’t have any spatial perspective or consistent figuration in them. It’s
intuitively to do with the line between consciousness and reality, or the physi-
cal world and the conscious world of sense data: visually making a fluid line in
Line Dance through the ever-present original two figures, without any cuts from
the figurative to the abstract, somehow seems to connect to that space between
consciousness and the physical world.
In my current work about improvisation, cognition and movement, I’ve real-
ized the importance for scientists of the fact that none of our senses works
independently. Our perceptions are fundamentally multisensory. This rhymes
for me because I’ve always felt that shape is something that I don’t just see: it is
something that I feel. For me, music is a very physical, emotionally embodied
experience. Perhaps thinking is moving, even when we are standing still.
10
As the editors of this volume suggest in the Introduction, shape is a term more
directly allied with spatial geometry than with sound. Its referential range has
nonetheless usefully been repurposed to help share propositional and meta-
phorical understandings of musical theory and practice. These span levels
of musical form, connections of phrasing to meaning and embodied move-
ments that accompany musical performance. Movements that accompany
music naturally lead to questions about how ideas of shape in music and dance
intersect with each other. The co-occurrence of music and dance has been
universal across human history and cultures (Mithen 2005). That universal-
ity raises questions not just about multimodal perceptual integration and the
coordinated control of human action, but also about more ineffable qualities
of meaning and emotion. In addressing choreomusical1 relationships involving
shape or other properties, we must inevitably draw not just on theories of music
and dance but also on theories of mental and cultural systems.
While the idea of shape may well be ubiquitous in discourse about music,
the significance of the term inflates by an order of magnitude with its differenti-
ated use in dance. Body shapes created by dancers and observed by audiences
self-evidently lie at the heart of choreography. Posters, publicity materials and
reviews of specific choreographic works almost invariably use visual images
that draw attention to beauty or drama in the shape of a body (Figure 10.1a),
or to the configuration of several bodies shaped in space (Figure 10.1b). These
shapes may be embellished by a sense of the scenography within which the
movement is embedded. Figure 10.1c, for example, makes graphic reference
via a grid to the movement analyses of the nineteenth-century photographer
Eadweard Muybridge.2 For shapes in dance that are frozen in time, their visual
328 context inherently constrains the meaning space in which questions might be
Intersecting shapes in music and in dance 329
posed and answers sought. Accompanying titling, captions and texts can signif-
icantly embellish and expand the range and specificity of insights, but explicit
photographic images of body shape and spatial surroundings in dance clearly
assume a ubiquity for which there is no simple equivalent in music.3
Any literature search will reveal the differentiated use of shape in discourse
about form and meanings in dance, particularly in the context of education.
In one dance curriculum,4 for example, the term ‘shape’ is taken to refer to
individual body shapes (the way in which dimensional space is used by the
body) and group shapes. The curriculum states that body shapes are present in
all actions in dance and that they convey meaning. Shapes with different but
specific attributes are identified, such as those with curving or organic shapes
(Figure 10.1a); open and closed shapes; symmetrical and asymmetrical ones;
harmonious and contrasting ones; centred and off- centred shapes (Figure
10.1b); or straight lines and angles (Figure 10.1c). The text then invites the
students to think about how differently nuanced meanings might, for example,
be expressed by curved versus angular bodily shapes as well as how shape for-
mation can be realized in personal space when moving on the spot or across the
space of a studio floor.
While many might agree about some basic aspects of the use of shape in
dance discourse, the picture becomes less clear when we segue to the dynam-
ics of movement over time. Music practitioners and theorists, across genres,
appear to share frames of reference provided by notation and nomenclature.
An associated body of music theory addresses hierarchical structuring of units
that are either parts of specific forms (e.g. exposition, variation, episode, fugue
or subject) or common to all tonal forms (e.g. phrase, period, theme or motive).
Since notational form in a music score indexes required properties of form over
time in sound, the common ground of shared notation and nomenclature in
music provides a scaffolding of huge value for discussing ideas about shape and
about how it relates to expressivity and meaning. Despite the presence of move-
ment notation systems such as Laban or Benesh, an equivalent scaffolding of
widely agreed nomenclatures does not exist in dance, a condition recognized
by humanities scholars (Jordan 2000: 84). This means that it may be easier to
discuss the dynamic properties of shape in music, and how these might relate to
nuances of meaning and interpretation, than to do so with the same precision
for dance.
On the other hand, avant-garde developments in both music and dance
practice evidence the degree to which shared nomenclature gives rise to non
traditional concepts and formal innovation. For example, the idea of a phrase in
dance could be attributed to modern dance pioneer Doris Humphrey (1959: 68),
who stated that ‘the good dance should be put together with phrases, and the
phrase has to have a recognizable shape, with a beginning and an end, rises and
falls in its over-all line, and differences in length for variety’. Tracing develop-
ments in dance after Humphrey, one quickly finds choreographers rejecting the
(a)
(b)
(c)
FIGURE 10.1 Selected illustrations for the productions of (a) ATOMOS, (b) ENTITY and
(c) UNDANCE for Wayne McGregor | Random Dance (photos: Ravi Deepres)
Intersecting shapes in music and in dance 331
very notion, with Yvonne Rainer (1974) and Meg Stuart (reported in Burrows
1998) proposing either to minimize or to eliminate the use of the term ‘phrase’.
In developments in music in the twentieth century, radical experimentation
with and rejection of shared notation and nomenclature are evident in John
Cage’s 1969 book Notations, a collection of nontraditional graphic scores col-
lected from 256 individual composers. However, what Notations also makes
evident is the ubiquity of traditional historical and cultural notation and scor-
ing techniques in music, providing a clear point of departure for the musical
avant-garde and again pointing towards what appears to be lacking in dance.
Interestingly, various research projects initiated by contemporary choreo
graphers in the last fifteen years have sought to address this apparent gap in
(a)
(b)
FIGURE 10.2 (a) Still from video annotating form and flow in Forsythe’s One Flat Thing;
(b) Difference Forms in movement viewed from above (image credit: Synchronous Object Project,
The Ohio State University and The Forsythe Company)
332 Music and Shape
The term ‘shape’ itself is, of course, used literally to underline intersections
between music and dance. A recent review of Tractys: The Art of Fugue, a piece
choreographed by Wayne McGregor for the Royal Ballet to an arrangement of
Bach’s original material, provides a canonical example: ‘Osipova and Edward
Watson skate over the shape of the score as they pull themselves into extreme
poses—the audience gives a hissed intake of breath at her first long past six
o’clock arabesque’ (Anderson 2014; our italics). A reference to organization in
music is seen as driving what the dancers are doing, and what they are doing
is projected as affecting the emotional responses of the viewer linked to an
elegant but, in this case, extreme body shape. The idea of shape is admirably
flexible here, but it is difficult to differentiate, disambiguate and share system-
atically across the full range of ways in which the sense of shape is experienced
and expressed in dance.
The same applies to how shape in music might relate to more ineffable quali-
ties of meaning and emotion. While the more stable nomenclature in music
might be useful to humanities scholars, the properties of meanings in music
have been widely discussed from psychological and linguistic perspectives but
with little consensus (e.g. Barnard 2012; Cross and Tolbert 2009; Tan et al.
2010). Emotional attributes of music, in particular, have been the subject of
much analysis and debate (e.g. Juslin and Sloboda 2010). Sloboda (1985: 62),
for example, notes that the tonal system can offer analogies for how people
represent emotions in ‘some semantic space’ and that these support a partial
Intersecting shapes in music and in dance 333
mapping of tonal relations onto emotions. Upward movements away from the
tonic are seen as suitable for expressing outward emotions while movements
towards the tonic signify rest or repose. Although meanings in dance have not
attracted such extensive debate in psychology, similar observations could be
made about both the emotional expressivity of specific properties of movement
and the wider ineffability of the meaning of movements.
We all have some grasp of what Sloboda might have meant by the idea of
‘some semantic space’ because his example gives that sense form. Likewise,
the idea of dancers ‘skating over the shape of the score’ makes communica-
tive sense but tells us little about whether an attribute of the score really did
underpin the choreographer’s decision-making or what the salient attributes,
if any, might have been. What the bigger semantic space or the detail it con-
tains might be is unspecified in both cases and therefore ineffable. When
cultural influences on making and experiencing performances are factored
into our wider equation, the domain to be addressed is potentially bound-
less. In this sense, music and dance are equally challenged to provide deeper
accounts of the impact of shape on processes and expressions of artistic
practice.
The sense of shape in music and dance is perhaps more readily ‘thinkable’ in
the context of a score as described by choreographer Jonathan Burrows (2010).
In A Choreographer’s Handbook, he denotes two definitions of a score. In one,
the score holds within it the detail of what you eventually see or hear. The
classical music score, as Burrows mentions, fits this definition, and as such the
classical score may be easier to study and discuss, as has already been indicated
by reference to shared notation and nomenclature. He points towards another
kind of score in which ‘what is written or thought is a tool for information,
image and inspiration, which acts as a source for what you will see, but whose
shape may be very different from the final realization’ (ibid.: 141). The idea of
shape in either music or dance as information that is mapped into a very dif-
ferent realization is clearly important, but how we flesh out a more enriched
understanding of what is involved in the mapping requires methods for pro
bing both processes and thinking. That in turn implies that researchers need
concepts and methods to study the processes and decision-making of artists
working in situ.
Applied psychology offers such concepts and methods. It will typically
seek to gain some understanding of a domain of practice, and then consider
how psychological theory and data might inform, and even modestly enhance,
aspects of that practice. To reveal senses of shape requires methods and instru-
ments that keep both implicit and explicit aspects in view. Such instruments
may also assist in filling in where there is the apparent lack of a shared notation
or nomenclature, thus coming into alignment with the artist’s research initia-
tives already mentioned.
334 Music and Shape
FIGURE 10.3 Relationships and representations that bridge sources of inspiration and a finished work
in contemporary dance
Intersecting shapes in music and in dance 335
she focuses. Here is just one illustrative analysis of her noting a relationship of
parallelism between music and dance in the work of Balanchine:
There is a clear example of close parallelism to quiet, lightly humor-
ous effect in Balanchine’s Agon (1957) Pas de Deux. Here we see a series
of staccato isolated gestures when the woman is in penché with one leg
around the man’s shoulder; they release the right hand clasp, she touches
the floor with her free hand, he releases his left hand to take his arm back
high. The gestures follow the rhythm and pitch contour of three notes in
Stravinsky’s score. They stand out as three ‘special’ moments within the
Pas de Deux. (Jordan 2000: 75)
With many such examples Jordan provides evidence of a rich picture of how
choreomusical relationships of this type can be expressed and formalized.
Pursuit of those relationships is one important methodology for adding gram-
matical and semantic detail into our understanding of intersections between
music and dance. For each of the choreographers whom Jordan discusses, or
any others inspired by her choreomusical thinking, the explicit products of her
scholarly work could populate the boxes in Figure 10.3, providing depth and
breadth to the representations that can be elaborated to fill out the right-hand
box and the mechanisms of discovery and production.
The properties of scores invoked in Jordan’s scholarly analyses are clearly
rather more specific than the generic idea of skating over the shape of the
score quoted earlier from the review of Tractys: The Art of Fugue. The prop-
erties brought into focus are dependent on the type of observer, the perspec-
tive they adopt and exactly what they happen to be attending to at the time.
In our own experimental research at Wayne McGregor | Random Dance, we
found that elite dancers vary considerably in the features they attend to when
analysing the same short segments of movement material (deLahunta and
Barnard 2005). While such variation can be seen as exposing the many ways
in which latent properties of music and dance can be used, it equally well
underscores the point that intersections between music and dance identi-
fied by scholars, critics and even expert colleagues may not reflect the inten-
tions of choreographers or the actual cognitive bases of their decisions. To
find out about these, we need to probe the thinking of the choreographers
themselves.
A recent interview with McGregor on how the idea of shape in music contrib-
utes to his own choreographic practice can also be integrated into Figure 10.3.8
Because he is trained in music, a musical score can provide him with a window
on how the composer is thinking rather than with what might be thought of
as content to be described or visualized in dance. Among other things, he uses
the musical score as a discovery representation from which aspects of the music
can be assimilated into his knowledge resources and choreographic thinking
rather than as a production representation, where a score is used to support
Intersecting shapes in music and in dance 337
… I also think the body, when it’s tasked, will create something shape-
wise which has an inherent musicality to it. The actual thing that’s made
has a musicality to it and that thing that’s made on that task is different
from the musicality of the thing that’s made over here with this task, if
the tasks are working…
Taken together, implicit or explicit ideas about shape in music and in dance
intersect in several distinct ways that again touch on meaning, emotion and
the parts played by creative design processes. When viewed in the context of
creating a production, the kinds of representations and processes identified in
Figure 10.3 enable us to organize and categorize intersections between music
and dance rather than just list them. It is a descriptive framework allowing us to
characterize choreographic, and potentially choreomusical, practices.
types of mental representation and imagery are called into play alongside fre-
quent shifts in attention among images in mind, body and external world. Here
the dancers are exploring movement design, assessing and modifying shapes
and timing, as they make their own decisions. In doing so, they traverse many
points in mental as well as physical space. What they attend to is as varied as
what was attended to in our study of observers viewing and parsing move-
ment material (deLahunta and Barnard 2005). The activities surrounding and
involved in task-related creation involve many of the mental processes (analy-
sis, assimilation, evaluation and selection, assessing and modifying, contextu-
alizing and synthesizing) used within the iterative cycles of design that lie at
the heart of Figure 10.3. These mental processes are open to systematic prob-
ing, again drawing on theories in cognitive science to help expose the decisions
behind the creative choices.
Shape or form in music clearly intersects with dance in powerful but quite dis-
tinct ways. It is information rather than a prescription. The bridging framework
offered in Figure 10.3 helps us to make sense of variation in the use of informa-
tion in music by pinpointing how it can come to be used in framing choreo-
graphic approaches, thinking, decision-making and even scholarly analyses of
them. The bridging model was one of the analytic lenses we used in the collab-
oration; the second and the one that supported most of the ongoing practical
work with McGregor and the company is a macroscopic lens that looks at men-
tal architecture as a whole. As with Figure 10.3, this lens provides some core
categorical distinctions of relevance to the analysis of mental processes and
mental imagery systems occurring within creation and performance of dance.
The particular mental architecture we called on here, Interacting Cognitive
Subsystems (Barnard 1985; Teasdale and Barnard 1993), is not just about cold
analytic and rational processes. It charts the whole embodied mental landscape
and provides a vocabulary for addressing both cognitive and emotional mean-
ings including the essences of ineffable feelings and intuitions.
If you show small children from different cultures the two shapes in Figure
10.4 together with the words ‘ulumoo’ and ‘takete’, studies show they pair
them up systematically (Davis 1961). This indicates something connecting these
sounds and shapes that is deeply shared. We experience such cognitive–affective
patterns as systematic ‘senses or feelings’ that yield reliable and measurable
behaviours, yet we seem to have made this connection intuitively. The dance
curriculum quoted in our opening paragraphs invited teachers to get pupils
to think about the meanings of angular body shapes and curved ones. Just
like the children matching angular or curved shapes to words and modalities,
dance pupils can undoubtedly do something similar with the idea of different
340 Music and Shape
Shapes or
Objects
body shapes with some measure of agreement. However, like wider meanings
in music and dance that make intuitive sense, as with our earlier discussion of
Sloboda’s (1985) notion of a semantic space in which emotional meanings and
musical patterns might exist, the items across modalities listed in the left-and
right-hand columns of Figure 10.4 share an ineffable quality. What underpins
the emotional responses to these shapes and nonsense words and other related
experiences involves a recipe that has something to do with sharpness, perhaps
an element of hardness or a dose of irregularity with a pinch of abruptness in
temporal patterning thrown in for good measure. And to understand this phe-
nomenon we may not need to reference any aspect of advanced cognition at all.
Figure 10.5 illustrates the potential gains from viewing issues of shape and
meaning across modalities of human experience through the lens of mental
architecture. On the lower left is a dog, set in the kind of learning experiment
made famous by Pavlov. For the purposes of argument, let us suppose the dog
has three sensory subsystems, each with three kinds of resources. An ‘image’ in
each one contains the patterns of sensations the dog experiences as informa-
tion flows in from the eyes, the body and the ears. Over time the patterns of
what goes with what within each of these modalities are extracted and stored in
a ‘memory’. Moment-to-moment content changes are analysed by ‘processes’
that produce summaries of the information being attended to in the visual,
auditory and bodily landscapes. These are passed to the ‘multimodal’ compo-
nent where a new image is formed of what goes with what out there in the world
and within the dog’s body. This is not a sensory image, but what the dog might,
hypothetically, experience as what we humans might understand as ‘a feeling’.
Again what goes with what is extracted and stored in memory, and précis of
Intersecting shapes in music and in dance 341
MULTIMODAL
Image
Memory Processes
Som Visc
Tail
Limbs
Ears
Head
Mouth
Lips
EFFECTORS
FIGURE 10.5 A core mammalian mental architecture with four subsystems, each with three
components (image, memory and processes). The arrows here indicate how information flows—from
sensory systems through to a single multimodal subsystem that in turn sends instructions to bodily
effectors like muscles (Som = somatic; Visc = visceral.)
the deeper multimodal patterns are used to control somatic and internal bodily
responses as well as skeletal ones. Multimodal synthesis over sensory modali-
ties blends multiple ‘dimensions’ underlying patterns of information, including
emotional patterns. This enables us to understand why even such ‘simple’ sig-
nificances have an ineffable quality.
The human mind is not good at thinking about many things at the same time,
but that is what the neural networks that underpin mental architectures do well:
they pull out invariant aspects over many dimensions, like those included in
the ‘recipe’ for understanding what items in the two columns of Figure 10.4
shared, and smooth over the irrelevant variations, or ‘noise.’ In dance research,
for example, a recent study by Afanador et al. (2008) tested cross-modal per-
ception of musical tempo and speed of dance movement. The results suggested
that, through ‘aural capture’, different tempos change perceptions of (and feel-
ings in our bodies about) observed movement speed. But music might also cre-
ate an impression of greater movement impact, or heightened or sharpened
movement dynamics. The mental architecture of Figure 10.5 enables us to
342 Music and Shape
Propositional
Spatial-Praxic Image Morphonolexical
Image Image
Memory Processes
Memory Processes Memory Processes
Image
Memory Processes
Implicational
Eyes
Head
Mouth
Breath
Mouth
Lips
Breath
Arms
Arms
FIGURE 10.6 Interacting cognitive subsystems: a nine-subsystem architecture for the human mind
(Barnard 1985; Teasdale and Barnard 1993)
passing information to and fro among them. While there is much technical
detail to add, these interactions, together with that between the propositional
and implicational subsystems, collectively give rise to our more advanced cog-
nitive capabilities.
As with our first analytic lens, this specification of types of mental image
and mental processes fractionates the mind in a way that enables us to be very
specific about properties of representations ‘in mind’ and how they depend one
on another. In the case of music and dance, we can illustrate how this kind of
fractionation enables us to address the ambiguities raised earlier about shape
and meaning. Try imagining in your ‘mind’s eye’ either a very simple score for
the scale of C or a dancer spinning on the spot. The image you create would be
assigned to the image component of the spatial-praxic subsystem, and it would
have been generated from a mental image of an idea with specific properties
344 Music and Shape
to translate those into movements for performing music and dance as well as to
write down notes or musical scores.
In the context of our research on dance, we have used this analytic lens
to study how dancers direct attention to and use different forms of imagery
while they are creating innovative movement vocabularies in the studio. Our
studies used a technique called experience sampling and examined a range of
tasks used in Wayne McGregor’s practice. From this we were able to establish
a great deal about the distributions of dancers’ habits of mind during their
creative work (May et al. 2011). Off the back of this empirical knowledge we
were then able to develop strategies for enriching their imagery to break those
mental habits—choreographic thinking tools. A more elaborate discussion of
the theoretical basis of these tools is provided elsewhere (deLahunta, Clarke
and Barnard 2012).
The lesson plan traverses selected forms of imagery and puts in place numer-
ous strategies and principles for breaking mental habits, for enriching the
content of imagery, and for scaffolding the translation of image content into
movement material. Since this chapter is concerned with intersections between
music and dance, Figure 10.7 reproduces two parts of the lesson plan in Mind
and Movement specifically directed at the use of auditory and musical stimuli.
(a) Extract from Lesson 5: Exercising Imagery based upon External Acoustic Images
Imagine
· While listening to the music, select one
of the properties and create a Visual
Score on paper that follows the changes
in this property, e.g. the pitch goes from
Images A Images
high to low. Make short strokes rather specific
of idea in of idea in
than a continuous line. mind’s eye idea mind’s ear
(b) Extract from Lesson 5: Using imagery Example Principles used in this extract
developed from External Acoustic images • Select a drawn element uses the Assign/Choose principle
in movement creation Graphic: Other examples:
Select a part of an image to start from
Assign
Make a decision about the shape you
Create want to use
Pick a word you want to work with
Using the visual score to create Choose a connection between image
movement and movement
• Give it an action uses the Exemplify/Make It Specific principle
• Select one of the lines from your Visual Graphic: Other examples:
Score and give it an action, e.g. a straight
Exemplify Name the kind of object
line that crumples, a curved line that
e.g. a hat becomes a baseball cap
rotates, a zigzag line that straightens.
Decide that a sound is percussive
Without listening to the sound, explore
Give a surface a rough texture
the action with movement.
Put lemonade in a glass
• Repeat with all the lines or shapes • Using a colour associated with a song uses the Superimpose/
on your score to create a movement Layer principle
phrase you can repeat. Graphic: Other examples
Superimpose Combine a sound image with a visual
• Think back to the colour you image
associated with your imagined song Connect something in the mind to
and use this to layer a quality onto something ‘out there’ in the world
your created movement. Bring things that were apart together
Colour in a shape
FIGURE 10.7 Extracts from the Mind and Movement educational resource that illustrate (a) the
development of imagery based upon musical stimuli and (b) the translation of that imagery into
innovative movement material (content and graphics credit jointly to Wayne McGregor | Random
Dance and Magpie Studios, London)
Intersecting shapes in music and in dance 347
The upper panel of this figure reproduces a segment dealing with the devel-
opment of imagery in response to a musical stimulus, and this section occurs
in the plan after students have been asked to identify a number of properties
of the music. From that point of departure they are asked to develop a visual
score and then layer in additional properties. Attributes of sounds are trans-
lated into shapes and emotionally enriched by meaning (through the ‘person-
ally important’ probe) and colour.
The upper panel of the figure also includes a graphic representation, taken
from the teaching resource, that is directly derived from the mental architecture
of Figure 10.6. It shows three mental processing loops—a representation of
the reciprocal arrows from the earlier, more detailed diagrammatic form. One
loop involves relating intuitions and feelings to specific ideas, a second relates
specific ideas to content in the mind’s eye, and the third relates specific ideas
to content in the mind’s ear. The imagery task is all about strategies for differ-
entiating and enriching image content—a key element of breaking habits and
generating innovations in their choreographic thinking. While this particular
example actually invites the students to generate a visual score, the primary
task is to be conducted within the mind alone. It is helpful to construe the
imagine task as an ‘attentional score’ that directs attention to move among the
three imagery loops and to delve into, and differentiate, the content of specific
images.
The lower panel of Figure 10.7 contains extracts from the same lesson and
illustrates some of the many mental steps involved in taking properties of men-
tal images, translating them into innovative movements and exploring how they
feel. The right-hand side of the lower panel provides examples of the twelve
general principles employed in this mapping, and additional examples of how
the same principles can be called into play across a number of modalities of
imagery. It is also noteworthy that the graphics, used to illustrate and index
each principle, were carefully designed to enhance senses of meaning through
visual shapes. Superimposing one property over another is a principle that can
be applied to any class of information or action. Looking back to our earlier
discussion about classes of intersection between music and dance within chor
eographic thinking, we can think of the visual scores derived from properties
of music as media for realizing information to underpin decision-making in
choreography or to constrain performance improvisations.
Mind and Movement is designed to be usable by teachers who are not spe-
cialists in theories from cognitive psychology. Therefore the model of mental
architecture that we have used throughout this collaborative research has been
carefully integrated into the structure of the lesson plans, the progression of the
exercises, and the questions that encourage the students to reflect on their exper
ience. The creative products of engaging with Mind and Movement clearly have
artistic potential, but these products may have wider impact for others curious
about collaborative research of the kind summarized here.
348 Music and Shape
Conclusion
basic research. The utility and validity of the research outcomes have been
grounded on long-term, cumulative and intense access to source material about
practice and the embedding of research within that practice. A key feature of
probing how information about shape or other properties is factored into artis-
tic decisions about movement and music has been to understand choreographic
thinking. In order to identify the patterns reported in this chapter, long-term
research access to a particular choreographer and his company was essential,
providing opportunities to document and analyse creative processes in direct
collaboration with the artists.
References
on the audience, some of this musical experimentation in turn affected the way
movies came to be shot and edited, and thus the shape of a new breed of cinema
was affected by new music.
As an alternative to the modus operandi of most composers who only ever
allow the shape of their music to be dictated by film, I prefer, when given the
opportunity, to compose and produce music in response to a script. So before
a single frame is shot and the production has got under way, I can, in discus-
sion with the director, influence the shape of a film in terms of how it is both
shot and edited. There are directors I’ve developed relationships with who will
take music I’ve prepared for them and construct scenes in which they move the
camera in response to my music, making the notion of shaping a two-way rela-
tionship. The shape of music in cinema doesn’t always have to be governed by
the shape of the film.
In sum, for me the notion of shape is embraced in my composition process,
which involves the creation of music that is sculpted and moulded around film.
A movie score forms the emotional shape around the film’s narrative, through
which, using the power that music has to shape human emotion, my music aims
to help its audience understand both the basic and the sometimes sophisticated
underlying themes expressed in the storytelling.
As I’ve tried to suggest, music and film have a mutually beneficial rela-
tionship in shaping each other in the minds of those who watch and listen,
and when that really works well it gives composer and film-makers together a
uniquely satisfying creative experience.
PART 5
Shapes felt
Reflection
Julia Holter, singer and composer
Most of the time I find that once a piece of music I have made has a ‘shape’—
or as I say it, has a ‘form’—it is finished, regardless of what shape it is. But it’s
hard to say how I know at what point it has a shape—it’s obviously a subjective
thing. I think I have in my mind a kind of closed rounded figure whose shape
changes continuously, like an amoeba or something. But it can (and always
will) stretch and morph into something new with every experience of listening
to the piece; all the parts inside are alive and will move around and change. It’s
just important that it is closed. That closure and the fact that things within it
can change but always remain within is what makes it a piece.
Again, it is hard to know how to explain the moment at which I decide
something I’ve written ‘has a shape’ and thus is finished, but a lot of times
what I notice the most is a consistent timbral blend. Once the individual sound
sources—the different instruments as well as ambient noises—start intertwin-
ing and feeling ‘like one’, that is like the closing of the shape. It is the point
at which I no longer can distinguish one part from another, because, even if
I think I can, that part that was recorded will never actually be what it was in
the moment it was recorded. For instance, maybe it will be clear that the bass
is the bass, but once the recording is finished, what was once ‘the bass’ will be
forming endless numbers of relationships within the song with other sounds,
every time I listen. Of course, those interactions would happen the moment the
bass was introduced into the recording, whether the piece was finished or not,
but when out of the chaos of all these relationships I get a feeling of oneness—
that’s when I know it’s done. I think this feeling of oneness comes when the
relationships between sounds grow so strong that they start forming offspring
sounds (artefacts, noise, interesting harmonics) and the connections between
all the sounds become denser, so that they are no longer independent.
Today I had an experience where I started hearing things that were never
recorded in a song I was working on—phantom sounds. I think I even heard
357
358 Music and Shape
people talking, and there were no actual people ever recorded talking. It was
because there was so much going on in the song during a particular climactic
moment that there was a lot of distortion and there were complicated interac-
tions between frequencies. It created noise and these artefacts, and it is what
made me love it even more. Everything was blending together and, in so doing,
seeming to emphasize new sounds. I think almost all pieces of music I like do
this to me, so I was glad it was happening.
I’m not a mathematician, but I am pretty sure there will always be an infinite
number of sounds that possibly could be heard while listening to a recorded
piece of music, because every time you hear it you might hear different fre-
quencies brought out in the harmonic spectrum. So when I say that the sounds
might ‘blend’, I don’t mean that when they blend and become ‘like one’ there
are fewer sounds or that they become just ‘one sound’. But that ‘feeling’ I get
when I hear outcomes of relationships between sounds—whether it’s people
talking or noise or whatever—is like an acknowledgement that the sounds have
become self-sufficient creatures, breathing on their own and reproducing, but
all within the world of a particular piece, a particular amoeba.
11
Prior (2010) and, in this volume, Prior (Chapter 7), Greasley and Prior
(Chapter 8) and the many Reflections offered by practitioners all show how
widely, easily and variously the concept of shape is used by musicians. And yet
what do sound and shape have to do with each other? I argue in this chapter
that, more than everything else it affords, shape functions as a quasi-submodal
concept, common to all the senses, that readily links musical sounds to the
feeling responses of listeners. Music invokes feeling states through modelling
their dynamic properties, and in turn those dynamic properties are used by
musicians to help them give lifelike qualities to music. In speaking of shape,
musicians are indicating, by a highly efficient means, the character of, or the
need for, dynamic patterns in sound that can model states of movement and
feeling, or indeed of anything that changes over time. To make the case, I draw
on recent work in philosophy, psychology and neuroscience in search of mech-
anisms capable of affording the apparent interconnectedness of shape and
musical sound. The chapter builds through a survey of relevant recent work,
climaxes in the middle with a view of the late work of Daniel Stern, and winds
down through thought about underlying mechanisms.
Synaesthetes who see colours on hearing sounds sometimes report that the
colours appear in specific shapes, including mobile shapes (Ward, Chapter 9
in this volume). Nonsynaesthetes, presented with animations representing syn-
aesthetes’ experiences in original and altered forms, tend to prefer the originals,
suggesting that the mappings, while highly personal, are not arbitrary (Ward
et al. 2008) and recruit some of the same mechanisms as normal perception
(Ward, Huckstep and Tsakanikos 2006). That a music–shape relationship is 359
360 Music and Shape
used so widely by musicians (Prior 2010, 2012) suggests that the mapping is
very easy to make, and therefore that the capacity is to some extent structural
or at any rate very thoroughly learned.
Walker et al. (2010) showed that infants as young as three or four months
prefer matchings of visual and pitch direction (both moving up, or both mov-
ing down) and pitch and sharpness (rising and pointed, or falling and rounded)
that correspond to the mappings reported in the extensive literature on adult
cross-modal mapping and synaesthetes, implying that infant perception might
be synaesthetic at birth though, in most people, later unlearned (Maurer and
Mondloch, 2006). But how fully unlearned?
Ramachandran and Hubbard famously demonstrated (2001; see also
Maurer, Pathman and Mondloch 2006) that almost everyone agrees that
rounded and pointed shapes map onto the nonsense words ‘bouba’ and ‘kiki’
rather than ‘kiki’ and ‘bouba’, confirming the findings of a similar experiment
by Köhler (1929) with ‘baluma’ and ‘takete’ (see Spence 2011 for a review of
these and other studies, including cross-cultural ones). Exceptions have been
found among children with autism spectrum disorder and people with damage
to certain brain areas, suggesting ‘that crossmodal correspondences, at least
those involving sound symbolism, can occur at quite a high level’ (ibid.: 974).
On the other hand, Dolscheid et al. (2013, discussed from a similar perspec-
tive in Eitan 2013) provided good evidence that the metaphors for pitch posi-
tion available in one’s native language influence one’s sense of pitch. For Dutch
speakers, high and low visual stimuli influenced the pitches they sang, while
thin and thick did not, and for Farsi speakers, for whom pitches are thin and
thick rather than high and low, it was the thickness of the visual stimuli that
had the effect, to approximately the same degree. In this respect, at least, lang
uage use apparently feeds back into music cognition. Training Dutch speakers
in the use of the Farsi terms produced a similar result, but training them in the
opposite terms (thick for high and thin for low) did not. It seems, then, that
language has an effect only in the direction already established in pre-linguistic
infant synaesthesia: it can reinforce but not fully determine our responses
(Eitan 2013). Similarly, the linguistic terms (high, low, thick, thin) appear to
have been adopted by languages because they have pre-linguistic origins.
Many of the synonyms for musical shape offered by participants in the
research leading to Prior (2012) can be classified linguistically (though not
necessarily according to participants’ intentions) as drawing on images of
either intensity/quantity or trajectory/direction or both (Table 11.1). Images
suggesting simple mapping onto two-dimensional space—shapes as visualized
forms—are less frequent than those suggesting motion with change, which is
perhaps not surprising given the nature of music, but it nonetheless suggests
that there is much more going on here than simply the pitch content of music
being imagined as height in relation to time. (More sophisticated uses are exam-
ined by Leech-Wilkinson and Prior 2014.)
Musical shape and feeling 361
TABLE 11.1 Some of the synonyms for ‘shape’ collected for Prior (2010)
Loudness
Crescendo Coming closer, acceleration, increasing energy (due to external force in slow
tempos only), running motion
Diminuendo Moving away, deceleration (at slower tempos only), falling pitch, falling or
sliding motion
Pitch contour
Rise Acceleration, spatial ascent, moving away (small effect), higher energy, running
or walking
Fall Deceleration, spatial descent, lower energy, leftwards motion, falling
Speed
Acceleration Descent
Deceleration Descent, moving away
Articulation
Legato changing Moving away, slowing (at slow speeds)
to staccato
Stern’s final work is concerned with The Present Moment (Stern 2004) and with
dynamic experience or Forms of Vitality (Stern 2010). Phenomenology has long
had related concerns, but Stern’s particular focus is on the experience of ‘now’,
the ever-present moment. Stern sees experience as characterized by a sequence
of present moments, each no more than a few seconds in length, which are
shaped by feeling responses to incoming perceptions, and which group together
to form dynamically shaped mini-dramas, sensed as a gestalt, through which
one lives. Stern records examples for analysis by asking participants to describe
episodes in their everyday lives. Here, a participant describes breakfast.
Present Moment 4
I am holding a slice of bread, not yet spread with honey. But it is a differ-
ent kind of bread than I normally buy. It feels strange and I am surprised
by it. I think, ‘What do I do with this bread?’ A mild negative feeling
arises.
Musical shape and feeling 365
This moment took about three seconds. She then spreads honey on the bread
without paying conscious attention to the act. A new moment begins adjacent
to the previous one.
Present Moment 5
I am then aware of biting into the honeyed bread. I like the texture and
think, ‘It’s not so bad.’ And with that a sense of feeling better builds up.
I then become conscious of the radio interview again. (Stern 2004: 13–14)
So too music, when one focuses one’s attention on it, can be experienced as
wholly engrossing, ‘unbelievably rich’. It occupies the subjective now. It cre-
ates a sense of stylistic and psychological wholeness (the one leading to the
other). Its phrases and sections have definition and gestalt-like qualities of self-
containedness within a greater whole; and at the local level they can be quite
short, short enough to be experienced within a psychological ‘now’. We identify
and understand music in terms of these experienced ‘moments’ which give it its
meaning for us. The feelings it generates ‘trace a time-shape of analogic risings
and fallings. In other words, they are carried on vitality affects (dynamic time-
shapes) that contour the experience temporally’. We sense musical continuity as
having narrative qualities, although it usefully lacks the precision of the ‘lived
stories’ which Stern finds make up everyday life. They ‘capture a sense of the
[music]’s style’ and analogously behave as if they were people, indeed as if they
were us. Thus musical units are indeed ‘psychodynamically relevant’.
Similarly, the curves Stern uses to represent these feeling shapes could
equally well be mapping aspects of a phrase, or melody, or loudnesses, texture,
rhythm, or also qualities like expectedness, complexity, mood, character, edgi-
ness, tension, all adding up in complex ways to give a sense of shape. This itself
emphasizes how directly the musical features (contour, loudness, speed) model
the qualitative (mood, tension and so on).
Stern’s ‘present moments’ bring together things one experiences as ‘now’ in
time spans that are only as long as the 2–5 seconds our perceptual systems allow
(Fraisse 1984). Stern here draws on Husserl’s phenomenology of time: the past
of the present moment, not remembered but rather still experienced because
still fading from the present; the present of the present moment; and the antici-
pated, expected, immediate future. This ties in usefully with existing music per-
ception and analytical theory, which argues that expectation is essential in the
creation of musical meaning (Huron 2006; Stern’s particular reference point is
Narmour 1990). But Stern’s point is that all three stages are taken together as
‘now’. This offers a more ecologically plausible notion of what it is like to hear
and to perform music, both performer and listener aiming always to treat what
comes next as a good continuation of what just happened. It follows, both for
Stern describing life and for us thinking of music, that the present moment is
not always the most intense; it may be happening in the shadow of what hap-
pened a moment ago, but what happened a moment ago is also something that
Musical shape and feeling 367
can define ‘now’ and give it shape. Indeed, ‘now’ can be relatively trivial so long
as one understands it in relation to the past and the anticipated future. And
Stern goes on to make precisely this analogy with music, noting how ‘much of
the richness of music lies in the fact that each subsequent phrase recontextual-
izes the previous one’, but at the same time, ‘A coherent experience was grasped
during the present moment, even though that experience may have multiple
fates’ (2004: 30). There is thus a constant ‘trialogue’ between past, present and
future characteristic of music and life (ibid.: 31).
But there are important differences. Music has hierarchical levels, for one
thing. It has an orderliness, in other words, that we do not find in life. In music
each event leads coherently into the next, or contrasts with it in a way that will
later be reconciled or resolved. There are no loose ends in music, or if there are
we tend to fault it, saying it is badly composed. Music is lifelike in a utopian
fantasy world where everything that happens makes sense and where, whatever
conflicts we may experience along the way, everything turns out for the best.
The entire tradition of music analysis is directed at proving this. (On music as
utopia, see also Levitas 2010.)
In Stern’s view, present moments characterized by ‘temporally contoured
feelings’ (2004: 36) are best thought of as ‘vitality affects’: ‘these temporal
contours of stimulations … are transposed into contours of feelings in us’
(ibid.: 64). Thus ‘temporal contours’ are the objective changes in intensity or
quality of the stimulations; ‘vitality affects’ are the subjectively experienced
shifts in internal feeling states that accompany the temporal contour, their
‘vitality’ being that sense in which they are the most characteristic aspect of
the affective experience of living. Most stimulation of the nervous system,
Stern says, whether it comes from within or without, ‘has a temporal shape
or contour that consists of analogic shifts in the intensity, rhythm, or form
of the stimulus’ (ibid.: 62). And this, of course, is exactly what happens in the
performance of a (western classical) score. Through the score, a roadmap is
already provided by the composer—the events of a life are determined—but
what the performer does is precisely to provide a temporal shape through
moment-to-moment adjustments in intensity, rhythm and pitch (or which-
ever of these dimensions is available through the instrument in use) that
provide triggers for the feelings that seem best (most satisfyingly) to arise
from those events. One can think of these as ‘expressive gestures’ (Leech-
Wilkinson 2006, 2009a). Expressive gestures generate vitality affects: it is
through expressive gestures that musical scores, which without them are so
dull as barely to be music at all,1 come to life. And what musicians over-
whelmingly mean when they talk about ‘shaping’ a phrase given in a score is,
as we have seen, that process of enlivening that is achieved through the use
of expressive gestures.
For Stern, though, the key idea is that the dynamic qualities of vitality
affects may be linked across modalities (Stern 2004: 37, 64–5). Vitality affects
368 Music and Shape
occur in many modes and can map easily from one mode to another. This
is certainly what has been repeatedly suggested for expressive gestures, which
index other things, or behave like them, and take meaning from the likeness
(Leech-Wilkinson 2006). As Kim (2013) points out (in her essay for the special
issues of Empirical Musicology Review on music and shape), for both Stern and
his predecessors, Hausegger (1887) and Truslit (1938), the experience of expres-
sive shape (vitality affect) is ‘understood in a broader sense than emotions’
(Kim 2013: 165). The emphasis here is on dynamics, not states, and in that
sense (shaped) forms of vitality may offer a more appropriate way of thinking
about the experience of music than seeing music as a sequence of inductions
or representations of emotional states (ibid.). A shape-focused view of music,
therefore, might offer a more ecologically valid way of understanding feeling
responses to music than do attempts to see music as expressive of particular
emotional states.
A final point of Stern’s (2004), though he intended it as a key to psychoana-
lytic practice, ties in with Kim’s (2013) focus on the aesthetics of empathy and is
especially relevant to our understanding of musical performance and response.
For Stern (2004: 22, 172–3) present moments shared, where analyst and analy-
sand seem to understand each other’s feelings with unmediated clarity, can be
especially intense and life-changing. Stern describes these as ‘a shared feeling
voyage’, when ‘two people traverse together a feeling-landscape as it unfolds in
real time.’ It seems possible that something like this takes place between musi-
cians performing well together and is imagined as happening also for listeners
attending to a performer with exceptional concentration and sympathy. In the
latter case, the listener, and perhaps the performer, feels that they understand
the other and become one with their music.2 It may be an illusion, but there is
intersubjectivity concentrated in these moments, as music happens, of a sort
that may not be unrelated (though this needs focused research) to song as sex-
ual attractor (Miller 2000).
Stern (2010) takes the notion of vitality form (a more abstract conception of
the form underlying a vitality affect) and sees it now as the most fundamental
percept characteristic of life: ‘Subjectively, a thought can rush onto the mental
stage and swell, or it can quietly just appear and then fade. It has a beginning,
middle, and ending… Mental movement, while it is happening, traces a profile
of its rising and falling strength as it is contoured in time. This is its dynamic
form of vitality’ (ibid.: 21). At the same time,
Vitality forms are hard to grasp because we experience them in almost all
waking activities. They are obscured by the felt quality of emotions as it
accompanies them. They are absorbed into the explicit meaning as the
vitality form accompanies a train of thought, so we do not pay attention
to the feel of the emergence of the thought, but only to its contents. It [the
vitality form] slips through our fingers. (ibid.: 10)
Musical shape and feeling 369
increasingly hazy tail) and an implied immediate future which may yet sur-
prise us. Shape conceptualizes all these aspects of music and feeling. More than
that, it carries them along in us. As Stern emphasizes, his work has significant
commonalities with phenomenology, sharing its prioritizing of experience over
structure or language as the dominant reference domain for cross-modal sig-
nification. What seems to be implied by the prioritizing of lived experience is
that knowledge of music acquired through our bodies and their preconscious,
unreflective motor and limbic responses underlies cognitive responses, provid-
ing a frame within which they operate and in relation to which they are selected
and form as thoughts about music. Underlying Stern’s work, therefore, and also
the work on embodiment that we consider next, is the wealth of recent research
on music’s interaction with the brain’s limbic and motor systems (Koelsch
2010; Altenmüller, Wiesendanger and Kesselring 2006). A full understanding
of music and shape, in so far as present research allows, would need to take
detailed account of that work.
Embodiment
Underlying principles
Stern and Johnson are in many ways offering similar views, albeit with differ-
ing foci.3 Stern is concerned with the dynamics of feelings as they are experi-
enced phenomenologically, Johnson with the process by which those feelings
have meaning. Together they offer us the beginnings of a coherent explanation
for the meaningful interaction of music and shape. They also suggest how it
is that shape seems to represent so many aspects of our experience of music
both as sound and as meaning—melodic contour, harmonic tension, loudness
envelopes, pitch inflections, textural change, performance actions, affective (in
the right social context, physical) response—most of which can function simul-
taneously on many structural levels, shapes nested within shapes within shapes
and so on from a single note to a whole piece. It may be possible to take our
understanding of this phenomenon one step further, however, by going back to
some of the research on which Stern and Johnson draw, and supplementing it
with findings from more recent work.
Two research themes are of particular interest: suprasensory modalities
and multisensory perception. Both offer ways of understanding the mechan-
ics of cross-domain mapping. First, though, we need to consider the level on
which mapping takes place. As Stern points out, Lawrence Marks in his classic
study of the theories then relevant to synaesthesia, The Unity of the Senses:
Interrelations among the Modalities (1978), traced ideas about suprasensory
attributes as far back as Democritus (c. 460–370 BCE) and Aristotle (384–22
BCE). For Aristotle a sixth, common sense, integrated the outputs of all the
others, allowing them to be perceived in terms of general attributes: ‘motion,
rest, number, form, magnitude, and unity’ (Marks 1978: 4; Doğantan-Dack
2013 points out a predecessor in Ehrenfels). Marks proposed that a level on
which (perhaps due to a common phylogenetic heritage) the senses overlap, and
where they could find features in common, could explain the ease with which
we make analogies between different sensory experiences (ibid.: 182–5). To
clarify what kinds of common features these might be, consider the process of
analysing auditory data into perceptual features—including melody, contour,
rhythm—demonstrated by the results of selective brain injury (Peretz 2003).
This necessarily involves the process, already well documented in studies of
auditory perception (summarized in Rees and Palmer 2010), by which the audi-
tory cortex analyses incoming sound. The question is what happens between
Musical shape and feeling 373
that initial analysis and our perception of melody, contour, rhythm and so on,
synthesized out of them.
It is easy to see that contour, for example, combines aspects of frequency,
relative quantity, speed and time, and is already quite a complex phenomenon.
Frequency, quantity, speed and time are not entirely straightforward either,
perhaps involving signals registered in a simpler form. So in constructing a
sense of contour the brain has to analyse out of the incoming sound data, and
then combine into a contour gestalt, components that are very much simpler.
Alternatively, if the gestalt is perceived first (which is also possible) it may then
be deconstructed into its components in order to search for them in gestalts
experienced before, identifying them and giving them meaning through previ-
ous experience. Either way, in their simplest, lowest-level form, we have no per-
ceptual access to these simpler components. We perceive them only within more
complex phenomena. However they are registered, each has a role in other per-
cepts too: frequency and quantity are necessary components of loudness, for
example, since without frequencies there is nothing to be heard, and without a
variable response to quantity there is no way of loudness being a percept. These
simpler components would not be consciously perceptible on their own, but
precisely because they are so simple, not yet combined into specific percepts,
they could be simple enough to be shared across several (or all) sensory modali-
ties. All that is then required for cross-domain mapping is a mechanism in the
brain that habitually compares one data stream with, or that has a response
mechanism sensitive to, a memory of others.
Both the simpler components and the mapping mechanism are implied in
studies by, respectively, Näätänen and Winkler (1999) and McLachlan and col-
leagues (2010, 2011). Näätänen and Winkler propose ‘sensory feature traces’
assembled during a pre-representational phase of sensory information pro-
cessing which then, in the following representational phase, are mapped onto
time and compared with the contents of long-term memory. It would be at this
second stage that features such as frequency, relative quantity, speed and time
become perceptible. First-stage features, Näätänen and Winkler suggest, are
simpler and inaccessible to conscious perception, which is exactly what we need
to explain cross-modal commonalities. Refining Näätänen and Winkler’s work
in a major survey of recent research on auditory processing, McLachlan and
Wilson (2010) propose that incoming auditory information is stored in a multi-
dimensional array in short-term memory, enabling multimodal similarities with
data in long-term memory to be identified and to contribute to sound source
identification and association (further developed in McLachlan et al. 2011)
This process, or something like it, allows us to make sense of the very high
rates of agreement that Eitan and Timmers (2010) found among participants
faced with having to choose which of two terms was most like what in the West
we call ‘high’ pitch. Table 11.3 shows the terms that produced most agreement.
I have added a column proposing an association learned through everyday
374 Music and Shape
TABLE 11.3 Highest-scoring results from Eitan and Timmers (2010; Table 1), with a
proposed environmental cause for the participants’ preference
experience which may underlie the participants’ preferences. Thus, old people
have lower voices than young, so high pitch is female or a granddaughter, while
low is male or a grandmother and so on. But other examples have no such
everyday explanation. What has pitch to do with brightness in our experience
of the world (an association we may share with chimpanzees: Ludwig, Adachi
and Matsuzawa 2011, though see also Spence and Deroy 2012)? In this case, it
seems necessary to hypothesize that an aspect of pitch, too low-level for us to
perceive, is being found to correspond to (or even be identical to) an aspect of
light. And the analysis of, for example, sound and light into very basic com-
ponents seems necessary for the construction of the gestalts evidenced by the
selective results of brain injuries (Peretz 2003) and for the process suggested by
Näätänen and Winkler (1999).
The same mechanism seems likely to underlie Marks’ (1978) suprasensory
attributes: ‘Suprasensory attributes are those categories or dimensions of expe-
rience that … apply to most or to all modalities. Intensity is a classic exam-
ple, to which duration must also be added. Size (extension), brightness, and
hedonic tone are other candidates, though perhaps not universally applicable’
(ibid.: 5). As we have seen in discussing automatic and learned responses, inten-
sity comes closest to the generalized quality that seems necessary to link pitch
and brightness, for example. Nonetheless, it may already be too complex to be
fundamental to cross-domain mapping. Shape–sound relations seem to depend
less on intensity as a basic category than on a lower-level aspect of changing
quantity that can behave identically in awareness of physical space (allowing
sound to be sensed as having height or proximity) and of intensity of feeling.
Musical shape and feeling 375
Bueti and Walsh (2009) offer a mechanism by which a sense of magnitude may
be acquired from infancy, building on Walsh’s (2003) work on quantity. But
all these could be synthesized out of Näätänen and Winkler’s (1999) ‘sensory
feature traces’.
Martino and Marks (2000) introduce key additional points. First, relation-
ships between modalities depend on an awareness of the relative position of a
stimulus on a scale defined by experience. We can say that a sound is bright only
because we can compare it to other sounds we have heard that are darker. No
sound can seem one or the other without that experience being accessible to us.
Here is where synaesthesia differs from cross-modal experience, however, since
for synaesthetes these relationships are automatic and invariant. This relativity
of attributes for most listeners is entirely compatible with the view I have sug-
gested: comparison is made between incoming data and things already known
(whether known through inheritance or, much more usually, through embodi-
ment, enculturation or learning). There is no need to assume measurement
against a baseline.
Secondly, Martino and Marks point out that a post-perceptual represen-
tation of stimuli (analysis into constituent parts and cross-domain mapping
after perception of domain-specific gestalts) would explain recent findings that
semantic relationships between stimuli can also produce congruence effects
(cross-domain mappings). McLachlan and Wilson (2010: 181) propose a simi-
lar role for verbal labels at an early stage in sound identification. Again, this
allows (indeed requires) embodiment, enculturation or learning to play a pow-
erful role in establishing mappings: experience creates rapid matching of stim-
ulus and meaning involving whichever domains have repeatedly been found
relevant (Yu 2008). Thus, music getting quieter maps to increasing distance
because of repeated experience in the real world (embodiment); music increas-
ing in frequency maps to increasing height because of repeated linguistic expe-
rience of ‘high’ and ‘low’ as descriptors of musical sounds (enculturation), or
more specifically for western classical musicians because of repeated experience
of notation (learning). Or to take a different kind of example, violent music is
embodied, martial music is encultured, cigar music is learned (see www.you-
tube.com/watch?v=NIckHmwZAeI).4
The other very relevant body of work that we need to consider (as Stern 2010:
49 suggests) concerns the possible coupling of motion and thought within the
sensorimotor system. Gallese and Lakoff (2005), building on Gallese’s earlier
work on mirror neurons and Lakoff’s on embodied concepts, offer an explana-
tion for the strength of couplings that bring together emotional response with
thought and motion. Musical shape, which involves all three, would provide an
outstanding example. They argue that ‘sensory modalities like vision, touch,
hearing and so on are actually integrated with each other and with motor con-
trol and planning’ (ibid.: 459). The same areas that control action also con-
struct ‘an integrated representation of (1) actions together with (2) objects
376 Music and Shape
acted on and (3) locations toward which actions are directed’ (ibid.: 460). In
other words, they argue that action, imagined action and understanding are all
tied together through using the same neural systems. The same might very well
be true for musical performance, listening to musical performance and under-
standing it. A very similar point is argued by Molnar-Szakacs and Overy (2006:
236): ‘according to the simulation mechanism implemented by the human mir-
ror neuron system, a similar or equivalent motor network is engaged by some-
one listening to singing/drumming as the motor network engaged by the actual
singer/drummer; from the large-scale movements of different notes to the tiny,
subtle movements of different timbres.’
Recent work on multisensory perception offers further ways in which sound
might generate a sense of shape. A review by Stein and Stanford (2008) con-
cludes that many or even most neural systems may be multimodal. One of their
sources, Ghazanfar and Schroeder (2006: 284), seems especially pertinent:
Traditionally, it has been assumed that the integration of such disparate
information at the cortical level was the task of specialized, higher-order
association areas of the neocortex. In stark contrast to this assumption,
the neurobiological data reviewed here suggest that much, if not all, of
neocortex is multisensory… The world is [a]barrage of sensory inputs,
our perception is a unified representation of it, and the neocortex is
organized in a manner to make the underlying processes as efficient as
possible.
In conclusion
The kinds of processes that are being revealed by research in the multisensory
brain go a long way towards explaining how sound could so easily be mapped
onto shape and why it might seem so ‘natural’ to musicians (and especially
western musicians) to think of musical performance as giving shape to the
notes in the score. It is an immensely flexible concept, very easy, because it
involves changing quantity and intensity, to apply in any aspect of sound and
experience where it can do some useful work. In performance, as in other kinds
of experience, shape can apply to anything that changes. In shaping a score,
performers can select one or more of many available dimensions in sound. And
this multi-applicability of shape to sound enables them to be responsive to con-
text, shaping different dimensions from moment to moment. It enables them to
define a personal approach to shaping as part of their own performance style,
identifying them, communicating their way of understanding music. It enables
them to work personally within a constrained period style, itself defined by
particular ways of shaping sound. But this same flexibility also affords the pos-
sibility of massive change in period style over time (Leech-Wilkinson 2009b).
This is exactly what we find now that we have well over one hundred years of
recorded performance.
The fundamental level on which a sense of shape can be passed around the
brain could be a crucial factor in allowing music to be made expressive by per-
formers in a shifting variety of complex and interesting ways, and it explains
how the concept of shape can be used to think about and act within all of them.
Shape gets us about as close as we can get to suprasensory modalities. It works
in every dimension that can change over time. And this is why it is so useful to
performers in thinking and talking about how to make music. It is one of the
most powerful ways we have of making sense of the experience of music with-
out having to be too specific.
A performer in a recent interview for Prior (2011) said: ‘And that’s the thing
about music, if you use imagery it makes your muscles do all kinds of things
that you don’t necessarily have to describe in a minutely physical way.’ This
is a crucial point. Instead of having to say, ‘I want you to make the A in bar
3 slightly softer and maybe 40 milliseconds longer than the previous G, and
then the B semiquaver a little bit longer, maybe another 20 ms, and about 20
dB quieter than you’d expect, or alternatively you could make the A slightly
louder and shorter and the G very short, or …’ and so on, you can simply say,
‘I’d like you to shape that phrase a little more’, and then the performer does
378 Music and Shape
whatever s/he feels works. Shape is great value for performers, communicating
the effect that is required, leaving them free to produce it through feeling, not
analysis (Leech-Wilkinson and Prior 2014; perhaps assisted by auditory imag-
ery: Keller, Dalla Bella and Koch 2010), and thus, free to use their experience,
judgement and taste, to express their musicianship. And we as listeners align
ourselves to the shaped template that the musician supplies (DeNora 2004).
Performed shapes, arranged in a persuasively and movingly managed sequence,
become our felt experience (Johnson 2007: 238). I suggested above that music is
lifelike in a utopian fantasy world where everything that happens makes sense
and where, whatever conflicts we may experience along the way, everything
turns out for the best. I was really speaking of composition then. But we can
say something very similar about performance and the way it uses the notion
of shape: to speak of shape in performance is to speak of the way music, given
a highly skilled performer, enacts an idealized image of the feeling experience
of our everyday lives.
In sum, we have seen that shape is a highly flexible concept widely used
by (especially, but not only) western musicians (Prior, Chapter 7) to talk
about the expressive qualities of a performance. It relates closely to other
concepts involving real or imagined motion through space (including gesture
and trajectory) or across terrain (landscape, contour). At a more general
level it conceptualizes change over time. But fundamentally, in all this dis-
course, shape is modelling changing feelings, and it is that mapping between
the dynamics of musical sound and the dynamics of feelings that allows
shape to function so effectively as a way of thinking and speaking about
musical expressivity. The dynamics of musical sound are easily analysed and
visualized with sound visualization and mapping software (such as Sonic
Visualiser) which gives some access to the shaped nature of performance and
the expressive work it does for the listener. But the underlying mechanisms
need to be teased out through other kinds of research. Stern’s ‘forms of
vitality’ offer a powerful means of thinking further about the shape of feel-
ing, while Johnson’s work on embodiment and image schemas helps to show
how the relationship between shape and feeling is grounded in bodily experi-
ence. Research suggesting the existence of modes of suprasensory perception
can be linked to work on the neural mechanisms of sensory perception that
arrives at similar conclusions. Research on multimodal perception offers a
neural mechanism by which a sense of shape may be generated simultane-
ously in sound, vision and motion, and sheds additional light on the ease
with which music is described using metaphor and in terms of its likeness
to other things, the means by which it acquires so many of the meanings
attributed to it.
Bringing this work together under the umbrella of shape brings us closer to
understanding both how a sense of shape is generated by musical sound and
how and why it is so effective as a tool for musicians. For the end user, thought
Musical shape and feeling 379
about and talk of shape enables the defining and sharing of ideas about an
activity (expressive performance) largely carried out within the domain of
feeling and intuitive response, a domain otherwise inaccessible to analysis or
discussion. Shape functions, then, on at least two levels. As a way of talking
generally about the dynamics of performance, it affords efficient and commu-
nicative teaching and rehearsing. As a model of the dynamics of performance,
it encapsulates the changing magnitudes of sound during performance, afford-
ing a sense of contoured trajectory through which feeling and sound can be
aligned.
References
Aksnes, H., 2001: ‘Music and its resonating body’, Danish Yearbook of Musicology 29:
81–100.
Altenmüller, E., M. Wiesendanger and J. Kesselring, 2006: Music, Motor Control and the
Brain (Oxford: Oxford University Press).
Antovic, M., 2009: ‘Musical metaphors in Serbian and Romani children: an empirical
study’, Metaphor and Symbol 24/3: 184–202.
Athanasopoulos, G. and N. Moran, 2013: ‘Cross- cultural representations of musical
shape’, Empirical Musicology Review 8/3–4: 185–99.
Bueti, D. and V. Walsh, 2009: ‘The parietal cortex and the representation of time, space,
number and other magnitudes’, Philosophical Transactions of the Royal Society B
364: 1831–40.
Currie, G., 2011: ‘Empathy for objects’, in A. Coplan and P. Goldie, eds., Empathy:
Philosophical and Psychological Perspectives (Oxford: Oxford University Press),
pp. 82–97.
DeNora, T., 2004: ‘Historical perspectives in music sociology’, Poetics 32: 211–21.
Doğantan-Dack, M., 2013: ‘Tonality: the shape of affect’, Empirical Musicology Review
8/3–4: 208–18.
Dolscheid, S., S. Shayan, A. Majid and D. Casasanto, 2013: ‘The thickness of musical pitch:
psychophysical evidence for linguistic relativity’, Psychological Science 24/5: 613–21.
Eitan, Z., 2013: ‘Musical objects, cross-domain correspondences, and cultural choice: com-
mentary on “Cross-cultural representations of musical shape” by George Athanasopoulos
and Nikki Moran’, Empirical Musicology Review 8/3–4: 204–7.
Eitan, Z. and R. Y. Granot, 2006: ‘How music moves: musical parameters and listeners’
images of motion’, Music Perception 23/3: 221–47.
Eitan, Z. and R. Y. Granot, 2007: ‘Intensity changes and perceived similarity: interpara-
metric analogies’, Musicae Scientiae, Discussion Forum 4A: 39–75.
Eitan, Z. and R. Timmers, 2010: ‘Beethoven’s last piano sonata and those who follow
crocodiles: cross-domain mappings of auditory pitch in a musical context’, Cognition
114: 405–22.
Fraisse, P., 1984: ‘Perception and estimation of time’, Annual Review of Psychology 35: 1–36.
Gallese, V. and G. Lakoff, 2005: ‘The brain’s concepts: the role of the sensory-motor system
in conceptual knowledge’, Cognitive Neuropsychology 22/3: 455–79.
380 Music and Shape
Levitas, R., 2010: ‘In eine bess’re Welt entrückt: reflections on music and utopia’, Utopia
Studies 21/2: 215–21.
Lipps, T., 1903: ‘Einfühlung, innere Nachahmung, und Organempfindungen’, Archiv für die
gesamte Psychologie 1: 185–204.
Ludwig, V. U., I. Adachi and T. Matsuzawa, 2011: ‘Visuoauditory mappings between high
luminance and high pitch are shared by chimpanzees (Pan troglodytes) and humans’,
Proceedings of the National Academy of Sciences 108/51: 20661–5.
Marks, L. E., 1978: The Unity of the Senses: Interrelations among the Modalities (New York:
Academic Press).
Martino, G. and L. E. Marks, 2000: ‘Cross-modal interaction between vision and touch: the
role of synesthetic correspondence’, Perception 29: 745–54.
Maurer, D. and C. J. Mondloch, 2006: ‘The infant as synesthete’, Attention and Performance
21: 449–71.
Maurer, D., T. Pathman and C. J. Mondloch, 2006: ‘The shape of boubas: sound–shape
correspondences in toddlers and adults’, Developmental Science 9/3: 316–22.
McLachlan, N. and S. Wilson, 2010: ‘The central role of recognition in auditory percep-
tion: a neurobiological model’, Psychological Review 117/1: 175–96.
McLachlan, N. M., L. J. Greco, E. C. Toner and S. J. Wilson, 2011: ‘Using spatial manip-
ulation to examine interactions between visual and auditory encoding of pitch and
time’, Frontiers in Psychology 1/233, doi 10.3389/fpsyg.2010.00233 (accessed 9 April
2017).
Miller, G., 2000: ‘Evolution of human music through sexual selection’, in N. L. Wallin,
B. Merker and S. Brown, eds., The Origins of Music (Cambridge, MA: MIT Press),
pp. 329–60.
Molnar-Szakacs, I. and K. Overy, 2006: ‘Music and mirror neurons: from motion to
“e”motion’, Social Cognitive and Affective Neuroscience 1: 235–41.
Murray, M., 2003: ‘Narrative psychology’, in J. A. Smith, ed., Qualitative Psychology:
A Practical Guide to Research Methods (London: Sage), pp. 111–31.
Näätänen, R. and I. Winkler, 1999: ‘The concept of auditory stimulus representation in
cognitive neuroscience’, Psychological Bulletin 125: 826–59.
Nakamura, J. and M. Csikszentmihalyi, 2009: ‘Flow theory and research’, in C. R.
Snyder and S. J. Lopez, eds., Oxford Handbook of Positive Psychology (Oxford: Oxford
University Press), pp. 195–206.
Narmour, E., 1990: The Analysis and Cognition of Basic Melodic Structures: The Implication-
Realization Model (Chicago: University of Chicago Press).
Peretz, I., 2003: ‘Brain specialization for music: new evidence from congenital amusia’, in
I. Peretz and R. Zatorre, eds., The Cognitive Neuroscience of Music (Oxford: Oxford
University Press), pp. 192–203.
Prince, J. B., M. A. Schmuckler and W. F. Thompson, 2009: ‘Cross-modal melodic contour
similarity’, Canadian Acoustics 37/1: 35–49.
Prior, H. M., 2010: ‘Links between music and shape: style-specific; language-specific;
or universal?’, paper presented at ‘Topics in Musical Universals: 1st International
Colloquium’, Aix-en-Provence, France, December 2010.
Prior, H. M., 2011: ‘Exploring the experience of shaping music in performance’, paper
presented at the Performance Studies Network First International Conference,
Cambridge, UK, 14–17 July 2011.
382 Music and Shape
Prior, H. M., 2012: ‘Shaping music in performance: report for questionnaire participants
(revised August 2012)’, http://www.cmpcp.ac.uk/wp-content/uploads/2015/09/Prior_
Report.pdf (accessed 9 April 2017).
Ramachandran, V. S. and E. M. Hubbard, 2001: ‘Synaesthesia: a window into perception,
thought and language’, Journal of Consciousness Studies 8/12: 3–34.
Rees, A. and A. R. Palmer, eds., 2010: The Oxford Handbook of Auditory Science: The
Auditory Brain (Oxford: Oxford University Press).
Spence, C., 2011 ‘Crossmodal correspondences: a tutorial review’, Attention, Perception,
and Psychophysics 73/4: 971–95.
Spence, C. and O. Deroy, 2012: ‘Crossmodal correspondences: innate or learned?’,
i-Perception 3: 316–18.
Stein, B. E. and T. R. Stanford, 2008: ‘Multisensory integration: current issues from
the perspective of the single neuron’, Nature Reviews: Neuroscience 9: 255–66 and
corrigendum.
Stern, D., 2004: The Present Moment in Psychotherapy and Everyday Life (New York: Norton).
Stern, D., 2010: Forms of Vitality: Exploring Dynamic Experience in Psychology, the Arts,
Psychotherapy, and Development (Oxford: Oxford University Press).
Stevens, C. J., E. Schubert, R. Haszard Morris, M. Frear, J. Chen, S. Healey, C. Schoknecht
and S. Hansen, 2009: ‘Cognition and the temporal arts: investigating audience response
to dance using PDAs that record continuous data during live performance’, International
Journal of Human-Computer Studies 67: 800–13.
Tarasti, E., 2004: ‘Music as a narrative art’, in M.-L. Ryan, ed., Narrative across Media: The
Languages of Storytelling (Lincoln: University of Nebraska Press), pp. 283–304.
Truslit, A., 1938: Gestaltung und Bewegung in der Musik (Berlin: Chr. Friedrich Vieweg).
Walker, P., J. G. Bremner, U. Mason, J. Spring, K. Mattock, A. Slater and S. P. Johnson,
2010: ‘Preverbal infants’ sensitivity to synaesthetic cross-modality correspondences’,
Psychological Science 21: 21–5.
Walsh, V., 2003: ‘A theory of magnitude: common cortical metrics of time, space and quan-
tity’, Trends in Cognitive Sciences 7: 483–8.
Ward, J., B. Huckstep and E. Tsakanikos, 2006: ‘Sound-colour synaesthesia: to what extent
does it use cross-modal mechanisms common to us all?’, Cortex 42: 264–80.
Ward, J., S. Moore, D. Thompson-Lake, S. Salih and B. Beck, 2008: ‘The aesthetic appeal
of auditory–visual synaesthetic perceptions in people without synaesthesia’, Perception
37: 1285–96.
Wingstedt, J., S. Brändström and J. Berg, 2010: ‘Narrative music, visuals and meaning in
film’, Visual Communication 9: 193–210.
Yu, N., 2008: ‘Metaphor from body and culture’, in R. W. Gibbs, ed., The Cambridge
Handbook of Metaphor and Thought (Cambridge: Cambridge University Press),
pp. 247–61.
Reflection
David Amram, composer, conductor,
jazz French horn player
Music already tells a story. Any good teacher always talks about the shape of
the phrase and how that individual moment relates to the whole picture. There’s
an arc to a piece of music, just like in classical theatre where there’s a begin-
ning, a middle and an end, and within those three essential areas of any form
of expression there’s a presentation of themes and variations, recapitulation,
dénouement and conclusion. Basic structure and symmetry are essential in both
art and life. I think most people, even if they don’t know there is such a thing,
have a much better experience when there is a structure, and then of course
within that structure you can do just about anything. Thinking in terms of sym-
metry and construction and shape and form, you’re able to deal with any kind
of music and understand as a performer, a composer or a conductor that what
you’re here to do is supposed to help the listener paint a picture themselves;
or when composing in the silence of your own space, make your composition
like a perfect building to inhabit, so that during the time that the piece is being
performed it gives the listener a chance to create some order out of the chaos
of their own life by following what it is you have given to them in the musical
journey you have created to be shared.
Music is visual as well as aural. Until modern times, people saw as well as
heard the musicians performing. Charlie Parker’s famous song Now’s the Time
expressed the whole idea of what music has always been about: a celebration
of the moment and the sanctity of what is transpiring at that moment which
may never happen again. Real-life experiences are of course as visual as they
are aural. That is why everyone involved in making music has to be concerned
with shape, form, movement and overall structure.
I shall always remember a campfire in the Menominee Indian reservation out-
side of Green Bay, Wisconsin. I was playing with the great Lakota Sioux singer
383
384 Music and Shape
Floyd Red Crow Westerman. After we were done with the concert, we went
off to this big bonfire and all the young Indian men were sitting there around
this huge campfire, probably as their great-great-great-great-grandparents had
done, and they were singing. They said, ‘David, how does this make you feel?’
I said, ‘Well, I just feel like I’m here thousands of years ago with your ancestors
and they’re here with us right now.’ And they said, ‘Well that’s why we do it.
This music is being sent by us directly up to the Great Creator.’ Then several
of them made a gesture with their hands over the fire as if they were fanning
both the smoke from the fire and the song itself upwards towards the sky. At
that moment I understood why and how this music should be played and how
it should be listened to and appreciated. The men at the fireplace had painted
me a picture of the shape and the form and the direction of what this music
was all about.
Musicians in my experience play differently when they watch dancers,
because just as the musicians inspire the dancers to dare to travel into the
unknown, playing with and for dancers enables you to bring out new things
in the music. I was at the Marlboro Music Festival in Vermont as composer in
residence in the summer of 1961. They had a famous flute player named Marcel
Moyse; he was legendary among flute players. He would get these phenomenal
classical players and he would take out a sheet of music for some old folk song,
or an old cowboy song or a German drinking song or a little French folk song,
and he would say, ‘Play this’. As the musician began to play, Marcel Moyse
would get up and start dancing round the room like a ballet dancer, singing
the melody of the song and waving his arms around like a crazed conductor.
With his unusual croaking style of singing and bizarre dance moves, it became
obvious that he was painting a picture for us of what the music was supposed
to do to those listening so that they could feel the music and the shape of the
piece being played. It’s just like what happens in the New Orleans Second Line
marches when you are as lucky as I was to be invited to march with the bands
who are playing the music that reflects their community. The singing, dancing,
playing and group collective feeling has its own unique sense of shape and
tempo.
The participatory nature of the Second Line does not promote the idea of
invading someone else’s turf. Quite the opposite: it shows that if you’re a seri-
ous musician and a respectful person, you don’t have to be terminally incarcer-
ated in your assigned slot for the rest of your life and never venture out of your
comfort zone. Once you see that music is a gateway to achieving a higher level
of understanding which creates a desire to learn more about all that transpires
in the rest of the world, musicians can begin to think of themselves as being
like a person painting a picture, who while creating is often obliged to start and
stop time. The silence between the stops and starts makes a sacred space for all
of us to be in, giving us the chance to store this magic moment in our memory
bank and in our hearts.
Reflection: David Amram 385
When we realize that music opens the doors to feelings, forms and history
that connects us to every other person on the planet, we become much more
comfortable when spending time with people from every walk of life, from street
people to architects, visual artists, brain surgeons and lawyers, astronauts, bar-
tenders, postal workers and athletes. Music shows us about all the things we
have in common and provides access to knowing how to ask others about all
the things we don’t know much about. And when we can put all this knowledge
in some kind of subjective storage space, we can become tourist guides for the
shapes and forms we have come to understand and, while sharing this informa-
tion, continue to learn more ourselves.
Reflection
Antony Pitts, composer and producer
Towards an outline …
Today I’m struggling with a piece that should have taken an afternoon to write
down. It appeared in the mist when summoned, almost on cue and apparently
fully formed, but it has taken another few months to grasp once more the
geometry of its form, the ratios and rationality of its quixotic light and shade.
The piece is a gift-cum-commission for Edward Higginbottom, at the end of
his long tenure at New College, Oxford. It’s a short setting of George Herbert’s
‘Love bade me welcome’ for unaccompanied choir, and from the moment
I started working on it, it was clear in my mind that this piece existed—com-
plete, perfect and (to me at least) unutterably beautiful and heart-rending.
A murmur, a hue, a shadow, an outline, a pang (what C. S. Lewis might well
term ‘Desire’)—these are the beginnings of creativity that I’m aware of: but
then I’m generally not aware of the real genesis of a piece of music, only
sometimes; often the beginnings are as impenetrable as forgotten dreams.
When I do know I’m thinking about a new piece, I sometimes feel sure that
I’m seeing or feeling it rather than hearing it. It—whatever it is—is amodal or
multimodal, an Ur-expression of some deeper confluence of ideas or tangling
of neurons.
In fact I wrote down the outline of ‘Love bade me welcome’ in an afternoon,
but in the weeks since I have struggled to agree with myself on its final form
(I’ve been seriously tempted to produce a folk-rock version). It’s the writing
that both destroys and captures the original idea in pinning it to the manu-
script: much of the struggle has been to reconcile what I consider the aesthetic
perfection of the original apparition, itself the ghostly flesh on the exquisitely
proportioned skeleton of Herbert’s poem, with the necessity of making it sing-
able while avoiding the safe danger of quantizing its chaotic edges down to an
excessively crystalline beauty.
386
Reflection: Antony Pitts 387
Reference
Gallo, F. A., [1977] 1985: Music of the Middle Ages II [Storia della Musica: Il Medioevo II],
trans. K. Eales (Cambridge: Cambridge University Press).
NOTES
Preface
1. Perhaps this explains why the marking criteria for the Associated Board of the Royal
Schools of Music state: ‘Candidates will be marked under five categories: pitch, time, tone,
shape and performance.’ Associated Board examiners, at any rate, need read no further.
http://gb.abrsm.org/en/our-exams/information-and-regulations/graded-music-exam-mark-
ing-criteria/.
2. As well as those mentioned below, from the previous essays by Doğantan-Dack (2013)
and Kim (2013), precursors with somewhat related ideas are well surveyed in Rothfarb
(2001) on ‘energetics’.
3. As a metaphor, ‘shape’ is not merely an interesting linguistic feature: as Gibbs states,
‘Metaphors are not simply an ornamental aspect of language, but a fundamental scheme
by which people conceptualize the world and their own activities’ (2008: 3). Even research-
ers who are unwilling to accept the theory of conceptual metaphor agree that the use of
metaphors provides information about thought processes (Cameron 2010).
4. http://www.cmpcp.ac.uk (accessed 9 April 2017).
5. http://www.charm.kcl.ac.uk (accessed 9 April 2017).
Chapter 1
1. Such ‘octave compression’ can be easily done and listened to with a modulo 12 opera-
tion on MIDI note data, effectively compressing everything into the confines of one octave.
Also, other shape manipulations can easily be done on MIDI note data and listened to
in view of shape features, such as so-called modus quaternion variants of pitch contours
(i.e. mirrored, retrograde and retrograde-mirrored in addition to the original).
2. In the original: ‘le premier objectif consiste à caractériser un phénomène en tant que
forme, forme “spatiale”. Comprendre signifie donc avant tout géométriser.’
3. Sound examples together with an overview of these typology and morphology prin-
ciples are available on audio CDs in Schaeffer (1998).
Chapter 2
1. Claxton (1980: 13) summarizes this aptly:
… [cognitive psychology] does not, after all, deal with whole people, but with a
very special and bizarre—almost Frankensteinian—preparation, which consists of a
brain attached to two eyes, two ears, and two index fingers. This preparation is only
to be found inside small, gloomy cubicles, outside which red lights burn to warn
ordinary people away. It stares fixedly at a small screen, and its fingers rest lightly
389
390 Notes
Chapter 3
1. Notably, while the first stanza takes place at dusk, the third stanza depicts sunrise.
Assuming temporal continuity, this implies that the protagonist has spent the entire night
rowing in front of the darkened town. This darkened gap separates the second and third
stanzas (see also Youens 2007).
2. Note that none of the action and motion described in the second and third stanzas is the
protagonist’s: the acting or moving forces are external (wind, water, oarsman, sun). Even the act
of seeing is passive and forced: the sun ‘shows me’ (zeigt mir) the town. The only act the protago-
nist is able to perform is that of losing (verlor), whose disclosure terminates the poem.
3. We are grateful for the advice of Clemens Wöllner, who suggested this and the follow-
ing German expressions.
4. As several analyses of ‘Die Stadt’ have noted (e.g. Morgan 1976; Schwartz 1986), the
modified melody of stanza 3 presents ‘cover tones’ above the structural melodic line. The
latter, directly reflected in the vocal line of stanza 1, is maintained in the upper line of the
Notes 391
piano accompaniment (e.g. E♭–D–C, bars 34–35). Notably, the modified, disjunct vocal line
at the stanza’s highpoints (bars 29–31, 33–35) is aligned with the bass.
5. Note that in comparing the performances, the profile of intensity variation is of par-
ticular interest. The absolute values (e.g. actual maximum or minimum intensity) strongly
depend on the recording and recording conditions.
6. Morgan (1976) links the unresolved ending of ‘Die Stadt’ with the opening of the next
song in the published Schwanengesang, ‘Am Meer’. Note that though this ordering is based
on Schubert’s autograph, several scholars have suggested that it does not represent Schubert’s
original intention (presumed to follow Heine’s ordering), but a revision made at the pub-
lisher’s request. For a discussion of this issue, see Litterick (1996) and Reed (1997: 258–61).
7. http://www.sonicvisualiser.org/(accessed 9 April 2017).
Chapter 4
1. See in particular Juslin and Sloboda (2010), Robinson (2005), Huron (2006) and
Nussbaum (2007).
2. Private communication.
3. Produced on Sonic Visualiser. The darker line maps tempo, the lighter, energy.
4. For a compelling exception to this apparent norm, see Leech-Wilkinson’s (2013: 50) dis-
cussion of Alfred Cortot’s 1920 recording of Chopin’s Berceuse, where the pianist matches
the rising and falling of the melody with slowing and speeding up of the beats.
5. The first group comprised first-year undergraduate music students at the University
of Liverpool. The second group were musicologists, including three renowned Bach schol-
ars. Another striking finding was that, when asked to select an emotion for the movements
(out of the set sadness, anger, tenderness, fear) nonexpert listeners (undergraduates) identi-
fied the Fuga with fear, while the experts identified it with anger, perhaps reflecting more
advanced structural listening.
Chapter 5
1. An adjectival form of the term ‘perspect’, a contraction of ‘perceived aspect’, used in
contradistinction to the term ‘parameter’, which in zygonic theory is reserved for the physi-
cal correlates of perceptual domains (Ockelford 2005: 10).
2. From the Greek word ‘zygon’, meaning ‘yoke’, and implying a union of two similar
things.
3. Although two dots sharing the same location can be distinguished functionally, as is
the case in which two lines of dots both converge on the same point, for example.
4. Observe that to depict the physical production of sounds visually requires the use of
symbolic representation (see Figure 5.16).
Reflection: Alice Eldridge
1. Improvisers were recruited via UK and European free improvisation organizations
and forums.
2. http://toplap.org/wiki/ManifestoDraft (accessed 9 April 2017).
3. Personal communication.
392 Notes
Chapter 6
1. For the sake of space, examples in this section were not reprinted. They can be accessed
online through the public domain IMSLP/Petrucci Library at http://imslp.org/wiki/Violin_
Concerto_in_D_major,_Op.61_(Beethoven,_Ludwig_van) (accessed 9 April 2017).
Chapter 7
1. Violinists were asked to play François Devienne’s (1759–1803) Sonata for Clarinet
in B♭ and Pianoforte No. 2, I: bars 1–12; harpsichordists were asked to play Thomas
Roseingrave’s (1690/ 91–
1766) ‘Sarabande’ from Complete Keyboard Music, Musica
Britannica, Vol. LXXXIV, ed. Johnstone and Platt, p. 60. None of the participants was
familiar with the piece given to them.
2. References for quotations from participants link to the complete model available on
the companion website . This quote can be found under Musical level: Phrase, L5.10.
3. L2.4.
4. H12.11.
5. L2.19.
6. L6.1.
7. Tina, T1.4.
8. Bridget, T1.1; Nathaniel, T1.8.
9. T1.3.
10. T1.5.
11. T2.4–7.
12. T2.8.
13. Bridget, T2.1; Tina, T2.9.
14. Jane, T2.10; Julian, T2.13; Yoshi, T2.16.
15. T2.2.
16. Jane, T3.4.
17. Bridget, T3.2.
18. Elsie, T3.3.
19. Bridget, T4.1.
20. Victor, T4.14.
21. Elsie, T4.3.
22. Julian, T4.26.
23. See Table 14.T4, online.
24. Darragh, T5.1; Elsie, T5.3; Tina, T5.4; Katharine, T5.16; Yoshi, T5.20–1.
25. Jane, T5.5– 8; Julian, T5.9–12; Katharine, T5.13– 14; Nathaniel, T5.18; Yoshi,
T5.19.
26. Bridget, T6.1; Darragh, T6.2–4; Elsie, T6.6; Tina, T6.7–8; Victor, T6.11; Yoshi,
T6.23.
27. Katharine, T6.16; Nathaniel, T6.18; Yoshi, T6.22.
28. Jane, T6.12–14; Julian, T6.15; Katharine, T6.17; Nathaniel, T6.19–21.
29. Jane, T7.8.
30. Julian, T7.12–13 and T7.15; Katharine, T7.18; Yoshi, T7.22–3.
31. Elsie, T7.1.
32. Victor, T7.4.
Notes 393
33. Victor, T7.5–7; Jane, T7.9– 11; Julian, T7.14; Katharine, T7.16– 17 and T7.19;
Nathaniel, T7.20–21.
34. T7.3.
35. See Table 7.T8, online.
36. T9.6.
37. See Table 7.T10, online.
38. See Tables 7.M1 and 7.M2, online.
39. Julian, M3.3; Katharine, M3.4.
40. Julian, M3.2; Nathaniel, M3.7.
41. See Tables 7.M4 and 7.M5, online.
42. See Table 7.M6, online.
43. Tina, M1.26; Yoshi, S6.53.
44. H11.18.
45. H4.11.
46. As this is a situational factor, it is not included in the online tables. The quote is
taken from approximately 00:55:30–00:56:30 in the interview.
47. S7.27.
48. This is also not included in the online tables. The quote is taken from approximately
01:01:00–01:02:00 in the interview.
49. L1.2, H1.17, S1.1 and S2.2.
50. L2.1, L3.1, L5.4, H4.2 and H12.2.
51. L2.5, T2.5, H1.4, H5.5 and H13.3.
52. L2.6, T2.6 and H1.5.
53. H5.22.
54. T4.38, H5.26, H14.30 and S6.45.
55. L5.20, T4.23, T10.8, H6.8, H7.19, H9.7, H12.13, M5.5, S5.15 and S7.37.
56. L5.18, T4.16, T6.10, T7.7 and H4.12.
57. L5.13, T4.10, T6.7, T9.4, H1.8, H4.8, H7.14, H10.3, S6.7 and S7.19.
58. L5.8, T5.2, T6.5, H2.2, H3.1, H12.4, M1.10, M2.3, S5.3, S6.4, S7.11 and S8.1.
59. L6.4, T4.8, T7.2, H1.7, H5.9, H9.3, H12.9, M1.18, M2.7, S5.4, S6.6, S7.16 and S8.4.
60. L6.16, H6.15, H7.25, H9.20, M4.11, S5.34 and S7.50.
Reflection: Simon Desbruslais
1. I recommend that the reader listens to a trumpet masterclass at a university or music
college to experience first-hand how technically accomplished trumpeters can create diver-
gent sound characters.
Reflection: Malcolm Bilson
1. Excerpts from Knowing the Score and a second video, Performing the Score, can be
seen at http://www.malcolmbilson.com (accessed 9 April 2017).
2. These three subjects, rhythmic notational conventions, tempo fluctuations and tempo
rubato, are covered in detail in the DVD Knowing the Score. Excerpts can be seen at
http://www.malcolmbilson.com. The Prokofiev and Haydn examples in this Reflection are
excerpts from that video.
394 Notes
Chapter 8
1. For a discussion of the ontology of recorded music, see Kania (2008).
2. Participants were asked to indicate which of twenty-nine categories of music they per-
formed in; the categories were derived from normative classifications of music and refined
through a pilot study (Prior 2012b).
3. Théberge (2012: 90) argues that ‘the rise of the Internet and what has been called, in
more general terms, “digital culture” has posed challenges and opportunities to contempo-
rary musicians, engineers and producers; indeed the widespread dissemination of software-
based tools for recording music challenges the very idea of who can lay claim to those roles
in contemporary culture.’
Reflection: Mark Applebaum
1. Robert Arnold’s documentary film about the project, There Is No Sound in My Head,
appears on Vimeo.com and the Mark Applebaum DVD The Metaphysics of Notation
(Innova 787; 2010).
2. A three-dimensional score would afford countless other possibilities.
3. A more exhaustive discussion might consider everything from music in Kandinsky’s
art to a survey of today’s young visual artists for whom the employment of sound in their
work is more typical than atypical.
4. They even gather for scholarly conferences on the topic, such as ‘Time Stands
Still: Notation in Music Practice’ at Wesleyan University, April 2013.
5. This is a particular pet peeve of mine: Why must nonstandard notation always stimu-
late an improvised response? Can’t a performer work out a matching of notational signs to
musical sounds and actions in advance?
Reflection: Alex Reuben
1. See for example Black on Maroon, Tate Gallery, http://www.tate.org.uk/whats-on/
exhibition/rothko/room-guide/room-3-seagram-murals (accessed 9 April 2017).
2. 1:00, DV, UK 2001, http://www.alexreuben.com/home/que-pasa (accessed 9 April
2017).
3. 5:00, 35 mm, DKTV, UK 2003, http://www.alexreuben.com/home/big-hair (accessed
9 April 2017).
4. My current cognitive research is supported by awards from the Wellcome Trust and
Arts Council England for a movie project, Cinderella (RockaFela).
5. 3:00, Digi., Channel 4 TV/ACE/MJW, UK 2004, http://www.alexreuben.com/home/
line-dance (accessed 9 April 2017).
6. http://www.metmuseum.org/toah/works-of-art/57.92 (accessed 9 April 2017).
7. http://www.youtube.com/watch?v=7bICqvmKL5s (accessed 9 April 2017).
8. 48:00, HD, ACE, UK 2008.
9. http://www.youtube.com/watch?v=LRmRBrnQq8o (accessed 9 April 2017).
10. http://www.youtube.com/watch?v=pbFvQo6Ao6o (accessed 9 April 2017).
11. http://www.youtube.com/watch?v=y651C7aNXRc (accessed 9 April 2017).
12. 64:00, HD, Sadler’s Wells/ACE, UK 2008-11. Excerpt: Rosemary Lee’s Meltdown,
5:00, http://www.alexreuben.com/home/newsreel-two (accessed 9 April 2017).
Notes 395
Chapter 10
1. A term first used by Paul Hodgins (1992).
2. http://www.eadweardmuybridge.co.uk (accessed 9 April 2017).
3. The closest thing in music would be the graphic score, but reading the score requires
translating the marks on the page to spatial properties in sound.
4. https://web.archive.org/web/20160119165847; http://www.curriculumsupport.educa-
tion.nsw.gov.au/primary/creativearts/dance/elements (accessed 9 April 2017).
5. http://synchronousobjects.osu.edu/media/inside.php?p=gallery#projectgallery (accessed
9 April 2017).
6. Wayne McGregor | Random Dance is now known as Studio Wayne McGregor and
the dance company is referred to as Company Wayne McGregor, both with the web address
http://waynemcgregor.com (accessed 9 April 2017).
7. https://blogs.montclair.edu/creativeresearch/2011/04/19/creativity-and-bridging-by-
philip-barnard-and-scott-delahunta/ (accessed 9 April 2017).
8. McGregor was interviewed by Philip Barnard in London on 22 August 2013; the
transcript is in the archives at Studio Wayne McGregor.
9. http://www.ted.com/talks/wayne_mcgregor_a_choreographer_s_creative_process_
in_real_time (accessed 9 April 2017).
10. Available at http://waynemcgregor.com/learning/resources/ (accessed 9 April 2017).
11. An extensive self-analysis of the creation of four of her seminal works made between
1981 and 1986.
Chapter 11
1. The easiest way to experience expression-free performance today is to listen to the
output of a plain MIDI encoding of just pitches and durations. The qualities that a ‘musi-
cal’ or ‘expressive’ performance brings are then very obvious.
2. Kim (2013: 166–7) brings together Lipps (1903), the mirror neuron literature, and the
philosopher Gregory Currie (2011) in support of a similar point. The literature on flow is
also relevant (for a recent survey, see Nakamura and Csikszentmihalyi 2009).
3. Aksnes (2001) first brought together the earlier work of Stern and Johnson. For this
reference I am grateful to Alessandro Miani, commenting on a draft of this chapter.
4. Should this web address change, a search on ‘Hamlet cigar ad’ is likely to work for
some time to come.
INDEX
indigenous Canadian ethnic groups, 152 Kremer, G., 109, 113, 114, 115–16
Infant-Directed Speech (IDS), 117–18 Kubrick, S., 353
infant synaesthesia, 308, 311, 360 Kupka, F., 304
inspiration, 334 Küssner, M. B., 41, 42–4, 152
intensity, 363, 374
loudness and, 144 Laban Movement Analysis, 43, 45–6
maximum, 65–6, 67 Lakoff, G., 130, 371, 375–6
mean, 65–6, 67 Lamont, A., 254–5
in musical notation, 144 Lang, L., xxvi
in performance, 144 Langer, S., 102, 364
phrase intensity and forte indication, 77–8 language, 12, 390n3
pitch and, 75–6 cognition relating to, 152–3
in ‘Die Stadt’, 73–6, 78–9 cross-modal mapping and, 58–9, 60
interdisciplinary collaboration, 334–8 emotion and, 60
interperspective relationships live coding languages, 167
diagram of, 135 music software languages, 167
directionality of, 136 perception relating to, 152–3
in zygonic theory, 131–6 pitch and, 152–3, 360
interperspective values, 130–1, 139 Larcombe, G. K., 310
Interpretative Phenomenological Analysis large-scale form, 90–5
(IPA), 220 shaping of, 102
inversion, 293–4 large-scale formal connectivity, 292–8
IPA. See Interpretative Phenomenological Layden, T. B., 322
Analysis learning
isolated tones, 47–8 children with learning difficulties, 135
practice and, 165–6
James, L., 324 shape, 165–7
Japan, 41–2 Leech-Wilkinson, D., xxxi, 41, 42–3, 152
Jarre, J.-M., 129, 158–160 on motherese and IDS, 117–18
jazz, 326 Leeds International Pianoforte
conventional jazz improvisation repertoire, Competition, 250
171, 175–81, 201 Légende (Enescu), 243
parameters in, 179 Leman, M., xxxi, 45, 50, 51
pedagogy of, 170, 175–81, 201 Léonard, H., 193–201
rhythmic patterns in, 179, 249 Levinson, J., 102, 104, 107
and synaesthesia, 257–8, 309 Ligeti, G., 25, 26, 89
Johnson, M., 370–2 light, 68–70, 78–9
Jordan, S., 335–6 limitation and variation of musical
Josquin, 300 topics approach, 175–81
Juslin, P., 102, 103, 238 Line Dance (2004) (Reuben), 325, 327
link schema, 130
Kanach, S., 191 Lipps, T., 395n2
Karwoski, T. F., 310 listener, 368
Katz, M., 252, 269 listening, 3
Kelly, M. E., 42, 48 active, 48–9
key, 89 musical gestures and, 100–2
key-postures, 5, 22–3 passive, 48–9
Kim, J. H., xxx, 368, 395n2 live coding languages, 167
kinaesthetic responses, 39–40 live performance
Kivy, P., 102 acoustics in, 258
Klavierschule (Türk), 249–50 audience in, 262
Klee, P., 303 body in, 262
Knowing the Score (Bilson), 248, 393n1 by DJs, 266–7
Kohn, D., 43 microphones in, 261
Kozak, M., 44 performer’s role in, 255–62
404 Index