Philosophy of Music

Philosophy of Music
An Introduction

R. A. Sharpe
 R. A. Sharpe, 2004

This book is copyright under the Berne Convention.

No reproduction without permission.
All rights reserved.

First published in 2004 by Acumen

Acumen Publishing Limited

15a Lewins Yard
East Street
Chesham
Bucks HP5 1HQ
www.acumenpublishing.co.uk

ISBN: 1-84465-000-6 (hardcover)

ISBN: 1-84465-001-4 (paperback)

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available
from the British Library.

Designed and typeset in Times

by Kate Williams, Swansea.
Printed and bound by Cromwell Press, Trowbridge.
Contents

Acknowledgements vii

Introduction 1

1 Overture and beginnings 9

Early speculations 13
Eduard Hanslick 16
Edmund Gurney 22
The legacy 27

2 The work of music 29

What makes music art? 30
The work of music 54

3 Meaning 85
Music and language 85
Arousalism 96
The basis for expressive descriptions 101
The limits of musical description 108
Meaning 111
Profundity in music 118

4 Value 123
Criteria 129
The ideal observer theory 132
The consensus among connoisseurs 137
The problem of irreconcilable differences of judgement 138

v
PHILOSOPHY OF MUSIC

The crux for the realist 148

Are verdicts essentially particular? 152
The concept of interest 161
The verdict on verdicts 164

Coda 167

Notes 171
Bibliography and discography 177
Index 183

vi
Acknowledgements

Many of my fellow writers on the aesthetics of music will recognize some

of their ideas in this book. I am deeply indebted to Stephen Davies, Lydia
Goehr, Peter Kivy, Jerrold Levinson, Derek Matravers, Colin Radford and
Aaron Ridley, even where I disagree with them. Stephen Davies read the
draft manuscript and made many helpful criticisms. I have too a general
debt in matters of the philosophy of art to another old friend, Colin Lyas,
and to the late Frank Sibley. My wife, Dr Lynne Sharpe, brought her inimi-
table sense of grammatical propriety as well as her philosophical nous to
the text. My colleague, David Walford, helped me with the translation of
Hanslick.
Some parts of this book were presented in a more technical form to
audiences at the universities of Leeds, Limerick and Wales, Lampeter, at
the Bolton Institute and at the annual conference of the British Society of
Aesthetics. I thank all the audiences for their criticisms. Some of the
reflections on profundity appeared in a different form in The British Jour-
nal of Aesthetics anniversary issue 2000 and some of the arguments on
functionalism in the International Yearbook for Aesthetics 1998. Finally, I
thank an anonymous reader of this book and my publisher, Steven Gerrard,
who suggested that I write it and made its arrival as free of birth pangs as
can reasonably be expected.
R. A. Sharpe

vii
Schubert, To Music, as written in the album of Albert Sowinski
Introduction

People always think somethings all true.

(Holden Caulfield in J. D. Salinger, The Catcher in the Rye)

Since you have opened this book, let me guess at what motivated you to
do so. You may, like me, be interested in the philosophy of music because
of your life with music. You may find it puzzling that your taste should be
sometimes idiosyncratic, sometimes different even from that of people
with whom you share many enthusiasms. Music you love bores some
others. You may wonder whether you are right or wrong about it. You may
wonder on what basis some music is thought to be more significant than
other music. You might have reflected on whether or not music means
anything. You may wonder at its power to excite us and move us to tears.
You might have been bemused by some of the things you read in the
newspapers. Could putting fireworks into a piano and then setting fire to it
really count as music? I, too, have found these matters puzzling, and they
have led me to consider some of the answers that have been proposed. For
it is music and our reactions to it, rather than the prospect of another area
on which philosophical training can be practised, that sparked my interest
in the first place. The questions I raise are, predominantly, questions about
the value of music, about the individuality of our assessments and about
the way in which we prize music for its power to move us. Even with other
issues, such as what makes something a piece of music, questions of value
are in the background. It is initially hard to see how setting fire to a piano
could be of value; indeed, for a pianist, it is a crime, since pianos have
their own individual character. So we are led, inevitably, to the lengthy
discussion of artistic value that constitutes the last chapter of this book.
For the notion of art and the notion of value are, I believe, connected. This

1
PHILOSOPHY OF MUSIC

remains so even if most music is mediocre. Responding to it and assessing

it remains at the heart of our encounter with it.
Sometimes these questions can be answered, and sometimes they seem
to have no answers. Where they have no answers I try to show why. But
first some general reflections. Music is ubiquitous. Go into a shop or a
restaurant and there is background music. Since the 1930s, films have
been accompanied by a score of greater or lesser distinction. Increasingly,
announcements or continuity on radio or television are accompanied by a
beat. Think how various it is rock music, rap, country and western,
gospel music, hymns, football songs, traditional jazz in pubs, contempo-
rary jazz in jazz clubs, world music and so on. Yet philosophers have writ-
ten almost entirely about Western classical music to the exclusion of the
music that most people are familiar with. I suppose it is related to the fact
that most philosophers are either from middle-class homes or climb into
the middle class, where a passion for such music is part of what is required
for a good life, in the sense in which Aristotle understood that idea. The
educated and well-rounded individual is interested in classical music. The
assumed pre-eminence of Western classical music is, as well, hardly just a
matter of the appropriation of a certain type of music by the haute
bourgeoisie, and its consequent elevation above other forms of music is
not merely the whim of the dominant class. It is arguable, after all, that
nothing in the rest of the musical world compares with it in depth, subtlety
and its capacity to engage with the rest of our lives. This is partly a matter
of sheer scale. A symphony may last for well over an hour. If other music
goes on that long as a work song might it is likely to be repetitious.
Perhaps only in Indian classical music can we find elsewhere organized
sound at this length. At the end of The Valkyrie, Wotan bids his beloved
daughter farewell, sending her to sleep, ringed with fire. The text is not
hugely eloquent. Combined with the music, it becomes unbearably
moving. It is followed by a transcendent passage for orchestra, whose
slow-moving length is requisite for the depiction of an age-long sleep.
This is a famous passage, lauded even by those who find Richard Wagner
resistible. Let me quote another example. At the climax of Schumanns
infrequently heard Scenes from Goethes Faust, Faust sings of a moment
of beauty that the ages cannot efface, reflecting the fact that the transitory
nature of human experience does not cancel its value. Schumann
underlines this with an orchestral tutti of tremendous power. Perhaps
Schumann is an overrated composer, but he was, as this shows, capable of
greatness. Such a climax would be out of place save in a work of length.
These are both examples of music with a text. For a third and last
example, illustrating a very different feature of Western classical music,

2
 INTRODUCTION

consider a purely instrumental movement. Prokofievs Fifth Symphony is

probably his finest (although a case could be made for the Sixth). Its
second movement is a whirlwind of a scherzo full of memorable and strik-
ing ideas. Yet the Fifth Symphony is a war symphony; it was composed
rapidly in 1944 and the first movement is weighty and serious. The
scherzo can sound flippant, but its surface is, I suspect, misleading. Its
frenetic motion easily turns to hysteria; fear is only just below the surface.
It is the forced jollity of those in mortal danger. To understand this move-
ment, and there are many parallels with his great contemporary Dmitri
Shostakovich, is to understand the significance of masks. In a tyranny
these are essential, but they may also be a form of courage when some-
thing other than tyranny threatens; namely, death in wartime. Only an art
capable of profundity can elicit reflections that bear in this way on our life
outside that art.
In other cultures, music may serve other purposes. The degree of
repetition that is largely absent from classical Western music may be part
of what is required to generate a trance-like absorption in the listener.
Listened to in the appropriate cultural context, its significance may
become clear. What Western music shows, to a greater or less extent, are
analogies with the novel, with a narrative that moves to a dnouement, and
it is only in fairly recent music such as that of Steve Reich and other
minimalists that we see an exception to this.
This, perhaps is the moment to anticipate another conclusion. On the
whole philosophers have tended, although perhaps not consciously or
explicitly, to fall in with the Hegelian notion that absolute or pure
music or music alone is the pinnacle of the art towards which Western
music, in its long history, has progressed. The sonata, the symphony or the
string quartet are higher forms than song, dance or opera. Their beauties
are the beauties of form. This forms part of an ideology that is deeply
embedded in thinking about classical music. However, despite its attrac-
tions, I do not think it easy to defend what is known as formalism: the
idea that the merits of music are no more than the aesthetic appeal of
beautiful patterns. I suggest that music is surrounded not only by texts,
dances, dramatic situations, liturgies and narrative programmes but also
by associations that may not be merely personal but may form connec-
tions with the landscape and history of the composers native land, with
the Zeitgeist and with a social and political context that form an aura, a
rich backcloth, in terms of which we read the music.
If this draws music closer to the other arts that are generally representa-
tional (drama, fiction, film and painting represent people and places, real or
imaginary), it is worth reflecting on some of the differences. One striking

3
PHILOSOPHY OF MUSIC

and, I think, infrequently observed feature of the arts is that different

individuals place differing requirements upon them. I dont know how much
I differ from others in this but I listen to the same piece of music or read the
same poem far more often than I re-read a novel and I re-read a novel more
often than I see films again. I suppose some people watch a film on video as
often as I listen to a favourite piece of music or some people re-read Jane
Austens Persuasion equally often. But even if this is the case, music does
make rather special demands. In the case of most of the finest Western music,
we have an art form where the performer has to practise over and over again.
The requirements of quality are thus very stringent indeed if a player is not
to tire of the music as a result of this constant repetition. Of course, we
should not assume that the rewards the player gets from practice are strictly
comparable with those the passive listener gets from hearing. For the pianist,
the physical exercise involved in playing the piano, the intimate connection
with the instrument, the dance of the fingers across the keyboard, is very
important. This daily contact with the keyboard is a necessary part of life,
bodily gesture as well as sound, as Charles Rosen says. He concludes his
recent book, Piano Notes,1 with the remark that the physical pleasure of
playing will, if anything does, ensure the survival of piano music. Neverthe-
less, although I think that the interest in mastering a piece, in solving prob-
lems of fingering and tempo, is, to an extent, independent of the interest the
listener takes in the music, even when we allow this the main point remains
untouched. Music of stature can be played over and over again with interest
and satisfaction and the realization of what can or what should be done with
a particular movement or a particular passage may only become clear after
dozens or even hundreds of repetitions. It is true, as well, that repeated
hearings or repeated readings are required to understand much of the greatest
music and poetry. Indeed, a mark of greatness has always been that we need
to re-encounter and constantly reassess. With fiction, however, it is often
enough to remember the gist. If somebody hands me a copy of a Shakespeare
sonnet I wont dismiss it by saying that Ive read it. The only excuse for not
reading it would be that I could recite it from memory. But fiction is unlike
this. There are exceptions; the philosopher Gilbert Ryle read the Jane Austen
novels every year and my music teacher reads the Charles Dickens novels
every year, or at least a selection of them. He evidently does not regard this
as some sort of mortification of the mind. But even so we re-read and are
expected to re-read fiction far less often than poetry. So the brush-off Ive
read it is generally applicable to fiction even though it would make sense
to say Well, its worth reading again if it was an outstanding novel. This is
a non-trivial remark in this context but it usually would be otiose if it were
a poem. Likewise with music.

4
 INTRODUCTION

Now music may be changing. There is talk of disposable art and a

belief abroad that old assumptions about the eternal value of the arts are
outmoded. I wonder if younger concert-goers look for an interest that
does not go beyond the performance. They expect to be stimulated or
entertained for a short while, but they may not want to hear the music
again. It may be becoming a more ephemeral art. So, for at least some, the
point of listening to new music is not to see if there is something that may
become a significant part of their lives, a piece to which they will return
with admiration and pleasure and in which they will expect to discover
new subtleties the more carefully and frequently they listen. Listening to a
contemporary piece may not mean any more than a single reading of a
contemporary novel, something to interest you pro tem but to be then cast
aside and not read again. If so, the nature of the reception of music is
changing. Improvised music is the norm rather than traditional Western
music, where a notated work is performed over and over again.
Be that as it may, traditionally the arts do vary and the consequences
for judgement have been stark in their differences. I might go to a concert
of new music and enjoy it in a way. But if I say I do not care whether I hear
it again my judgement is somewhat damning. Normally we expect of
music, if it is of the highest quality, that it will repay many re-hearings.
And we usually do not desire to re-read a novel we have just read and
enjoyed in the way we may ache to hear a piece of music over again within
an hour or two of first hearing it. Poetry can, of course, be memorized, but
we might want very much to read again or hear again a verse that we half
remember. Painting and sculpture are rather different cases again. I
imagine that people to whom painting matters a lot are suddenly con-
vulsed with the desire to see something that is, unfortunately, half way
across the world. If this is so, life is more frustrating for them than it is for
the lover of books or music or even films, for reproductions are no substi-
tute for the real thing.
All this evidently bears on the role the arts play in our lives. The
relationship we have to art to some extent, and this I wish to explore later
in this book, resembles the relationship we have with family and friends.
In both cases the requirements vary. Some of our family we see frequently
and would be unhappy at seeing them rarely. Some we see seldom but
keep in touch, while others we see or write to occasionally. Likewise
with the arts, although, as I have said, the arts vary in the extent to which
they permit this intimacy. It is a contingency that a man who was a close
friend lives in Malta or Brisbane and that I have not seen him for years. It
is a contingency also that a favourite painting is in The Hague and that I
have only seen it once in the flesh. But for rich Westerners, books and

5
PHILOSOPHY OF MUSIC

music are rarely the victim of such contingencies. Only if they are out of
print or deleted are they inaccessible.
This book is a mixture of the general and particular. My experience in
teaching music is that the philosophical issues that arise most frequently
are, first, what distinguishes music from noise or ambient sound and,
secondly, how can the quality of music be anything else than a matter of
individual taste? It is thus difficult to write about the philosophical issues
that arise out of music without encountering problems that constitute the
philosophy of the arts more generally. Chapter 2 begins with a general
discussion of what it is for an artefact to be a work of art before going on
to consider some of the more specific questions about the nature of music.
Chapter 4 is perhaps more general than any of the others in that nearly all
the questions I raise about the nature and measurement of artistic value
arise equally for the other arts. Only when we consider performance are
the questions more narrowly focused. So some of the questions I shall
raise in this book are questions that are equally a source of puzzlement in
literature or drama or film. But most are specific to music; questions about
the status of a work of music and its relation to performance do arise with
respect to the other performing arts such as dance, but because musics
relationship to its notation is far more significant than is the relationship
between dance and its notation, the analogies with mathematics that have
so often been canvassed have a greater pertinence here. Chapter 3, in
particular, discusses a question that is peculiar to music and has always
been central to philosophical discussion of it: can music have a meaning?
If so, what can it be?
I must issue a few warnings. I discuss differing theories about the
nature of the musical work and its meaning and value. I have tried to
represent fairly both the strengths and the weaknesses of the views that
have been put forward. Very few positions in philosophy have nothing to
be said for them. But I make no pretence of impartiality. A decent
philosophical book puts forward various positions and the positions, of
course, are my own, even where I share them with other thinkers. What I
have tried to do is to say clearly when a view is my own and when it is
likely to meet with opposition. I strongly suspect that what I have to say
towards the end of the chapter on value is the most controversial part of
this book. However, these are conclusions that are the result of long
thought about verdicts on merit, and they seem interesting to me. But I
expect some readers to disagree strongly.
Secondly, some of my musical judgements may seem rather eccentric
to the reader. I have some strong likes and dislikes. I have tried to tone
down the idiosyncrasies of my judgements but they will keep toning up

6
 INTRODUCTION

again. I have therefore given up, in the conviction that the best I can do is
to warn against them. Unfortunately, as far as this issue is concerned,
many of my arguments require that I consider cases where opinion sharply
varies. It seemed, then, that the best course is to take real cases rather than
fictional cases. Were I to take an imaginary case, such as the supposition
that Beethovens middle quartets are not much good, or that Miles Davis
was a rotten musician, the consequent astonishment and irritation would
only get in the way of an appreciation of the issues. So I have taken cases
where some of us seem to be at odds with much musical opinion. I think it
was probably puzzlement over the peremptory nature of our tastes that
caused me to become interested in aesthetics in the first place. I have
always been troubled by the way that people differ in their judgements of
value. In my last book, Music and Humanism, I wrestled with this prob-
lem of diverging tastes, attributing what conformity there is to the
existence of an ideology, embodied in the canon, that directs our judge-
ments. Thus classical musicians think of the central tradition as Austro-
German and regard music from other countries as marginal or provincial
to a greater or lesser extent. Our ambivalent relationship to this official
history explains in part the difficulty we have with musicians outside this
canon and our defensiveness about them once we have recognized major
talents who do not belong to the dominant tradition. But I did not consider
the possibility of divergent judgements within a common taste. These, and
the difficulties they raise, are discussed in Chapter 4.
Thirdly, as my own philosophical life nears its end I have become
increasingly disenchanted with the tradition of philosophical analysis, not
because few analyses have been successfully completed but because,
when the concepts are interesting, they cannot be. Philosophical analysis
is the programme that attempts to provide necessary and sufficient
conditions for the application of a concept like knowledge or truth or,
relevant here, work of art. Chapter 2 gives some reasons why I think this
exercise is futile. Again, this is a judgement that will not find too much
general acceptance and it is no good pretending otherwise. For many
philosophers, conceptual analysis is their raison dtre and to challenge it
is to challenge something deeply lodged within their philosophical
ideology. On the other hand, as I said earlier, I am not in the business of
writing a book that does not state the truth as I see it.
To reject the programme of analysis is not to object to the tradition of
analytic philosophy. This book is in that tradition. I am interested in
positions that can be stated clearly and defended by argument. Ideally, I
want to know what the world must be like if a philosophical claim is to be
true. Otherwise, I do not think philosophy is an intellectual discipline and

7
PHILOSOPHY OF MUSIC

there seems no very good reason to practise it. This means that the ideas
introduced by some writers on music find no place here, writers such as
Arthur Schopenhauer and Suzanne Langer.2 I readily admit that I do not
understand what could be meant by the formers observation that music is
the image of the will itself, and nothing I have read has made this much
clearer to me, although Schopenhauer was probably important in raising
the status of music among the arts. But although I find analytic philosophy
a fascinating and worthwhile task in itself, nothing, of course, that a
philosopher can do enriches our lives in the way that the best of art does.
No philosopher, however eminent, who is realistic about his subject and
his achievements would think them worth the value of a minor but lasting
lyric, an unforgettable carol or a comic masterpiece in film like The Blues
Brothers. Towards the end of his life, Benjamin Britten grouped a
collection of settings under the title Sacred and Profane. One of the choral
settings is of a carol, Maiden in the Mor Lay, and it lasts just a minute
and a half. I have studied, taught and written about philosophy for fifty
years. I have been fortunate. The subject has always absorbed me. But I
venture that, compared with the genius displayed by Britten in this tiny
piece, the genius, undoubted genius, displayed by Kant in his Critique of
Pure Reason is minuscule. This is not even one of Brittens greatest
works, but its sheer invention takes our breath away. Such is the disparity
between the creative artist and the philosopher. Only when, and how
rarely, a philosopher like Hume in his Dialogues Concerning Natural
Religion brings to philosophy the sort of literary elegance and moral
passion displayed by Samuel Johnson in his Review of Soame Jenyns can
the claims we make for the philosopher be remotely comparable to those
that we can make for the creative artist.
What I attempt here is the next best thing. I can try to interest those of
you who have thought, with puzzlement, about this wonderful art. I hope
to clear up misunderstandings that block our appreciation of the art of
music and some of these misunderstandings are philosophical in nature.
They may distort our judgement of music. Here a philosopher should
aspire to be what Locke called an underlabourer. The only way he can
hope to enrich our lives in any way remotely comparable to that of the
creative artist is to succeed in what Horace mistakenly thought was the
task of the poet: to please and instruct.

8
Chapter 1

Overture and beginnings

Philosophical reflection on music is more than two thousand years old but
it is patchy. Beyond a handful of names, beginning, perhaps, with Plato and
Aristotle and leaping two thousand years to Eduard Hanslick and Edmund
Gurney, most of what has been written is only of interest to historians of
ideas. But the past two decades have seen an extraordinary flowering in the
aesthetics of music that has eclipsed earlier speculations. This philosophi-
cal activity has been predominantly analytic in style. It prizes and expects
clarity and detail in argument. There are other philosophical traditions but
I am not aware that there has been any sudden efflorescence in writing on
the aesthetics of music in these other provinces. The newcomer to analytic
philosophy may easily form the impression that a great deal of it is devoted
to the invention and solution of rather arcane problems requiring
considerable intelligence but rather remote from the rest of humanity and
life as we know it. There is some substance in this but it is not an entirely
just verdict. Analytic philosophy can have implications for our lives by
making us reflect upon unconsidered presuppositions; sometimes it may
lead us to reflect on our lives and our values and cause us either to value
things differently or perhaps more directly to alter our conduct. If
utilitarians persuade people that they should grade their actions solely on
the principle of what maximizes happiness, then people will change. At the
other end of the spectrum, the issue of whether the concept of truth is
evidence-transcendent, which has occupied many philosophers for much of
their time over the past three decades, does not strike me as one that is likely
to have any immediate results for the way we live our lives. It has its
fascinations, of course, but these do become rather close to the fascinations
of doing a crossword without the satisfactions afforded by finding an
answer. The distinction is not, of course, absolute. Even the most arcane

9
PHILOSOPHY OF MUSIC

issues leave eddies in their wake that may make imperative certain more
mundane decisions, and we shall see that this is true, perhaps surprisingly,
of the question of realism in music. In writing about the philosophy of
music I shall try to concentrate on issues that illuminate musical listening
and practice. Although I shall write about apparently esoteric topics such as
what a musical work is, these all eventually debouch into more practical
issues of playing and listening.
My reason for taking this approach is that philosophy is, I think,
important when it is a cultural critique or an originating force in culture.
Most often a philosopher picks up ideas that are in the air and elaborates
them into some sort of system. Individualism was in the air in the Renais-
sance, as we can see from the essays of Montaigne or from Hamlet. In
both of these, in their different ways, there appears a persona that strikes
us as utterly idiosyncratic. But it was a philosopher, Hobbes, who elabor-
ated this conception of the individual as a self-contained unit concerned
with satisfying his or her own desires, an individual who competes with
others; he considered its implications for the body politic in what has been
called his possessive individualism. Descartess picture of the self-
contained individual whose claims to knowledge need a grounding and a
justification, I suggest, equally rationalizes ideas that were in the air in
Western European culture. The greatest and most important philosophical
achievements either rationalize and render explicit what is implicit in
the culture of the times or deconstruct it, as Hume did, paradigmatically,
in Dialogues Concerning Natural Religion. In this way philosophers may
be both contributors to and critics of what is often thought of as the
ideology of a society. They contribute to or subtract from a cultural
hegemony. Many of the philosophical issues that arise about music arise
because music has been understood in terms of those ideas that dominate
in the culture, ideas whose origins do not necessarily have much to do
with purely musical issues. Thus the central feature of reflection on music
between, say, 1750 and 1950 was the idea that music is a language,
specifically a language of the emotions in which the composer expresses
his mental state and communicates it to the listener. Such an idea owes
much to Romanticism, indeed it is hardly thinkable without a general
romanticism, (recall Wordsworths famous remark about emotion recol-
lected in tranquillity1). Equally, nineteenth-century reflection in the
Hegelian tradition led to the idea that music evolves into an art that owes
nothing to a text or programme. Pure or absolute music is the pin-
nacle of the art, perhaps even the art to which all others aspire.
Pertinent to the rest of this book is the question of what happens when
philosophers introduce ideas that are rather more obliquely related to the

10
 OVERTURE AND BEGINNINGS

stock of everyday concepts. The invention of a technical concept is partly

a piece of hubris on the part of a writer; she expects her reader to expend
time in memorizing its significance and, possibly, looking back through the
pages to remind himself of the definition. Of course, sometimes the labour
is not great. Everybody can see what is meant by mentalese; it indicates
an internal mental language, a language of thought. Neologisms can be well
chosen. More significantly, in a philosopher it is a danger sign. More often
a writer takes an existing word, such as utility or person, and, quite
often unawares, delicately adjusts the meaning. Where person was both
a legal term and used to describe a role-player, philosophers use it to
describe the human beings that people our globe and any putative rational
aliens, and with that shift and replacement come problems. Sometimes their
usage adds to or modifies the ideology of their culture. If their invention or
adaptation is picked up by others, the way we think may change. An
example is Thomas Kuhns use of the word paradigm, which has become
widespread outside philosophy. It is possible that words like soul or
spirit originated in this way.
This picture of philosophy as both a cultural force and critique is very
distant from that which prevails in academia. Aestheticians have concen-
trated instead on the programme of philosophical analysis. Analysis is the
programme of giving necessary and sufficient conditions for the applica-
tion of terms like law of nature, morally good action, work of art,
truth and knowledge, to name just a few. Thus, in the middle of the
twentieth century, knowledge was defined as justified true belief,
only for counter-examples to ruin this analysis. Such is the dominance of
the model of science in our culture understandably that when doubts
arose about analysis, refuge was sought in the idea that philosophy is a
kind of science, acknowledging the ways in which great scientists have
sometimes seemed to be involved in philosophy of science. Thus many
philosophers now claim that their discipline is some sort of branch of
science, a view that I believe to be a grave mistake. The style of argument
in philosophy is different and the facts appealed to are familiar and
banal. Philosophical refutations depend not on experiments but on
reminders; very characteristically the cut and thrust of philosophical
debate depends on adjusting a thesis in the light of apparent counter-
examples to it. The characteristic philosophical error is that of overlook-
ing something that is in front of you. How else can we explain the persist-
ence of philosophical theories that present infanticide as no crime or claim
that animals do not feel pain because they do not talk about it?
The project of analysis has lingered on in the philosophy of art. Indeed,
much of the history of the subject since its coming of age as a subsection

11
PHILOSOPHY OF MUSIC

of philosophy four decades ago has been the attempt to say what a work of
art is. In parallel, we have the attempted analysis of the concept of the
work of music, a generic term not much used by musicians. What does
its advent portend? The fact that it is comparatively recent and that its use
is conspicuous in the talk of philosophers and in general talk about music
rather than in the conversation of practising musicians or music-lovers
ought to give us pause. It suggests a certain artificiality; perhaps work of
music was coined in the course of general reflection about the place of
music in our culture. There is certainly a well-argued case for regarding
the concept of a work of music as one that developed in the nineteenth
century when the status of music as one of the fine arts was still not entire-
ly secure. Works of music were posited as analogues of the works of art
found in art galleries and museums. To talk of a work of music was to
draw an implicit comparison with sculptures and paintings. The concert
hall was the museum where such items were displayed. Thus the con-
cept of a work of music plays a role that we need to consider.2 I believe
that when we do we shall feel less sanguine about either the possibility of
an analysis of the concept that will command general acceptance or about
the utility of such an analysis.
Now if it is true that work of music is a rather late arrival, a term used
by thinkers about music rather than by practising musicians or the general
public, there is a dilemma here. To the extent that the concept up for
analysis is quasi-technical (that is, coined by specialists as a term of art to
use among themselves) it is, to that degree, private property. Questions
raised in analysis may indeed have answers but those answers will arrive
directly or by few steps. It is a self-inflicted solution. Compare this with
everyday concepts used by many; these may have fluctuating senses or
delicate gradations of sense. Examples are aesthetic concepts like eleg-
ant or beautiful, or moral concepts like decent or a good man. That
different speakers apply them to different objects suggests that there may
be overlapping usages rather than an identity of meaning in such words.
These differing shades of meaning can reflect quite important differences
between people and the ways they think about music or morals or any-
thing else. For it is how our usages differ from others that reveals what is
distinctive about ourselves and, more generally, our culture. Philosophy,
then, in so far as it reflects the latter, becomes a sort of a priori cultural
anthropology. It shows us how we differ in the ways we think. But this
philosophical task, of course, is incompatible with the ideal of analysis,
for that shows the essence of a concept, such as the concept of knowledge.
Analysis possesses a generality. What I find interesting here is how analy-
sis so often privileges one sort of knowledge, such as the logical or

12
 OVERTURE AND BEGINNINGS

mathematical over others. Of course, analysis reflects the philosophers

wish to universalize. But in many areas philosophy, born from the urge to
find general patterns, has to learn that sometimes generalization is fruit-
less. For example, what is called music differs. To endorse and privilege
one sort of music, the Western classical tradition of an increasingly fully
notated music, is to simplify and distort, as I shall try to show. There are
many varieties of music. This in itself will not preclude analysis; there
would simply have to be many different analyses, one for each variety of
music. Rather, it is the loose fit of such concepts, the fuzziness of the
boundaries and the significance of such fuzziness that makes analysis
impossible.

Early speculations

I remarked that philosophizing about music is over two millennia old.3 We

know much more about what the ancient Greeks thought about music than
about how their music sounded. The Greek word we translate as music
is more inclusive than ours, as well. It is associated with dancing and, in
particular, with poetry and physical exercise as well as with singing and
the playing of instruments. Indeed, one aspect of philosophical discussion
about music we would do well to bear in mind is the fact that, even though
we may accurately translate a word as music in the sense that it refers, in
its original language, to what we think of as music, singing and the play-
ing of instruments, it may have implications that our word music lacks.
There is good reason to think that the nuances of our word music have
shifted even over the past three centuries.
In particular the Greek conception of music was dualistic in two ways.
First, and very strikingly, the Greeks conceived of the celestial world as in
harmony. Music described the mathematical relationships that held
between heavenly bodies. Indeed, it was assumed that there was a music of
the spheres and there was discussion as to why human beings could not hear
it; was this a consequence of our being used to it from birth so that by
habituation we no longer heard it much as you may not notice the ticking
of the clock or were exceptional individuals, like Pythagoras perhaps,
able to hear it? This mathematical conception of music as abstract intervals,
perhaps best translated as harmony, a matter of harmonic relationships
expressed numerically, was to play an important part in European thought
up to the Renaissance. As late as Jean Philippe Rameau, we find the idea
that the pleasure music offers is a consequence of the way it imitates the
divinely ordered and mathematical structure of nature. Additionally and

13
PHILOSOPHY OF MUSIC

separately, for the ancient Greeks music was what was heard, sung or
played on the lute or aulos, as it still is for us. The second way in which the
Greek conception of music is dualistic has its modern progeny. Music was
seen as having very opposed aspects, one of which was ecstatic and
frenzied and the other civilizing and educational. We are familiar with
Nietzsches distinction between the art of Dionysius and the art of Apollo:
the one intoxicating and orgasmic, and the other calm, reflective, ordered
and balanced. We even see the contrast in the music of a single composer,
Stravinsky, in The Rite of Spring on the one hand and Apollon Musagte or
Agon on the other. The first aspect is what older people have found
alarming in rock music, beginning with Bill Haley and continuing with the
Rolling Stones. Finally, it is worth remarking the Greek emphasis on the
pleasure that music gives, which is particularly striking in Aristotle.
Aristotles emphasis on the fact that music involves movement is also
anticipatory. We will need to consider in more detail later on both the idea
that music brings pleasure and the idea that musics power comes from the
way it moves people. It speaks to the hearts of men. But the distinction
between music as an abstract business of tonal relationships and music as
heard is a contrast that is still evident in contemporary discussion.
These varying conceptions of music lasted well into the Middle Ages.
For medieval thinkers music was in part a theoretical branch of knowledge,
a science of numbers, and in part an aid to mystical experience. But it was
also an aural attraction: a seductive matter of the senses that presented a
temptation to indulge in those worldly pleasures that divines regarded with
considerable suspicion. All these ideas were rooted in Greek thinking about
music and were transmitted to medieval thinkers through the immensely
influential writings of Boethius. The belief, too, that the life of the mind
was of superior value to the work of the artisan implied that the dexterity
required of the performing musician was considered a poor thing compared
to theoretical understanding. Medieval writers would often scoff at the
singer ignorant of the theory that underlay the music he sang. This
prejudice certainly has not vanished. There are relics of this attitude in
modern anecdotes about the ignorance of orchestral musicians and singers.
Recently an eminent conductor who conducted the Vienna Philharmonic in
Vaughan Williamss Tallis Fantasia said that the orchestra admired the
music immensely and enquired whether Vaughan Williams had written
much else and whether Tallis was still alive; and a flautist with a leading
American orchestra who was regarded as a bit of an intellectual by his
colleagues was approached with the question Who came first Mozart or
Verdi? These may be apocryphal indeed it is hard to think otherwise
but their relevance lies in the attitude they enshrine.

14
 OVERTURE AND BEGINNINGS

With the early Renaissance, a greater emphasis was placed upon the
pleasure that music brings. We read of the sweetness of music perhaps
the main descriptive phrase to be found in late medieval writing is dulce
or sweetness and of the importance of moving the affections. But this
did not rule out a very considerable interest in the mathematical propor-
tions that underlay the stringing of instruments. A case was made for a
return to nature, a reaction against the complications of late medieval
polyphony. At the same time, improvements in the technology of key-
board instruments required great attention to the issues of tuning and
temperament. Musicians became theorists as well as composers and per-
formers and the gap between theoretical understanding and practical
competence narrowed. Theory, too, had less and less to do with the sup-
posed celestial harmonies. Rather, the Pythagorean doctrine transmuted in
the hands of later thinkers like Rameau into the idea that music, like
nature, was under the sway of laws. It was an enterprise of reason.
These seeds came to fruition in the Renaissance in two ways. First of
all the concept of what is natural became a sort of talisman. Anybody
reasonably widely read in Renaissance literature will be aware of its
ubiquity. Both the church and secular composers were uneasy at the
earlier custom of setting a single word to such cascades of notes that the
sense of the word was lost. Theorists and composers (and they were usual-
ly the same individual) began to consider what sort of music was approp-
riate for the setting of a particular text and music as accompanied monody
began to seem natural, a weasel word if ever there was one. Theorists
justified their preferences as a return to the ideas of the Greeks. In both
sacred and secular music, the music was seen as subservient to the text, an
idea that constantly resurfaced in the time of Claudio Monteverdi and in
the arguments of Raniero Calzabigi, Christoph von Glucks librettist, and,
of course, with some degree of anachronism in Richard Strausss opera
Capriccio. Music, it was thought, should move the affections, and it did so
by the choosing of music which served the emotional features of the text.
The German doctrine of the Affektenlehre developed the idea that music is
linked with the passions. Descartes wrote in Les Passions de lme that the
purpose of music is to arouse certain emotions in us, a doctrine reminis-
cent of those philosophical theories that advocate arousalism, which is
an idea we shall examine in detail later. The second legacy of Renaissance
thinking about music, then, is the notion that, as speech is the language of
the mind, so music is the language of the heart.

15
PHILOSOPHY OF MUSIC

Eduard Hanslick

There are philosophers, and Schopenhauer springs immediately to mind,

who have been thought to have influenced composers. Wagner was
certainly influenced by Schopenhauer, although whether he would have
composed differently if he hadnt read Schopenhauer is as debatable as
the question of what precisely Richard Strauss owed to Nietzsche beyond
the programme of Also Sprach Zarathustra. I am much more interested in
philosophers of music who have influenced the way music is received;
they may do this by encouraging us to pay attention to certain features of
the music, say its form as opposed to its expressive qualities of lightness
or gloom; consequently they may directly or indirectly suggest a canon of
music that causes us to privilege certain composers or certain national
schools at the expense of others. Thus a view that regards pure or
abstract instrumental music as the pinnacle of the art tends to elevate the
symphony, the sonata or the fugue at the expense of opera and song or
promote certain types of symphonic writing over others. Thus a prefer-
ence for Brahms over Tchaikovsky may be partly explained. Of course
such prejudices, if that is what they are, can also be a product of certain
national biases and they can vary. I recall an eminent Russian pianist
saying that as a young man he was led to think of the Russian school as
having a sensitivity and soul lacking in the more arid intellectual products
of western and central Europe, and particularly the first Viennese school, a
school that most Westerners are educated to see as the pinnacle of musical
achievement. These latter considerations apart, the thinker about music
who had done as much as any to influence ways of listening is, at the same
time, one of the two thinkers who have most influenced the analytic
movement in the philosophy of music: Eduard Hanslick.
Where modern philosophizing about music is concerned, the first
important figure, probably the most important figure prior to the extra-
ordinary developments in the subject in the past two decades, is Hanslick
(18251904), the Austrian music critic familiar to music lovers through
his skirmishes with Wagner. He has been thought of as the arch-formalist,
a formalist being somebody who thinks that the value of music lies in its
formal properties of design and line and not in its expressive capacity, but
he never denied that music calls into life sorrow and joy. What
formalism is we shall need to consider in a little more detail here and more
fully in Chapter 3.
Hanslick seems to have planned his On the Musically Beautiful4 as a
prolegomena to a fuller treatment of the subject; however that more expan-
sive work was never published. What we have instead is a discussion of

16
 OVERTURE AND BEGINNINGS

what he thinks of as misconceptions about the art current in his time. In

fact, he supplies in an appendix a sort of rogues gallery of quotations from
various writers, many of whom maintain forms of the thesis known as
expressionism: the thesis, first, that the purpose of music is to express,
represent or arouse feeling, and, secondly, that feelings are the content of
music and that music is the language of feeling. Both claims are, he thinks,
false; it is not the case that the purpose of music is to arouse the feelings and
feelings are not the content of music.5
Hanslicks book is, in fact, tricky. It is, despite its brevity, not very well
organized, sometimes inconsistent and, in places, pretty obscure, and this
explains the fact that most writers on Hanslick manage to miss some of
the salient points or misrepresent some of his ideas while getting others
right. (His account of imagination is difficult to comprehend and to perse-
vere with this on the assumption that there are some specific ideas here
that a temporarily inarticulate author has failed to get across is, I think,
premature.) Nevertheless the importance of the book is undeniable, firstly
because it contains some arguments that are remarkably sophisticated for
its period, and, secondly, because Hanslick expresses, mostly implicitly,
what we might call an ideology of music in which instrumental or
pure or absolute music is seen as the pinnacle of the art, a view that
has only been seriously challenged by musical thinkers in the past few
decades. This is related, as we shall see, to what has been described as
formalism: the idea that music is essentially a matter of patterns in
sound that have no meaning or reference to the world beyond music or
that, if they do, these connections are irrelevant to its value. (Part of the
remit of formalism is the encouragement of a suspicion of instrumental
colour. Music that relies on ear-catching timbres was assumed mere-
tricious. If music could not be played without loss in a piano arrangement
or reduction, as it is usually called, it was suspect. Its merit must lie in
the aesthetic appeal of its patterns of sound.) In some respects Hanslick
was a formalist, but what is meant here by formalism needs spelling out.
If formalism means that music cannot be described properly as having
expressive features such as gravity or joy, then Hanslick, despite what he
says at some points, is not a formalist. He allows that music can be expres-
sive, although he does think these descriptions are figurative. What he
argues is that this has nothing to do with its beauty.6 The constant harping
on beauty is an unfortunate feature of Hanslicks discussion. There is
great music, as there is great painting, that is ugly. If we make the discrim-
inations required, we can see that beauty is a feature of some art of quality,
but not of all. Even Mozart sometimes wrote music that it is simply a
perversion of language to describe as beautiful; the splendidly craggy

17
PHILOSOPHY OF MUSIC

Eine kleine Gigue K.574 is an example and in work by composers such

as Bach, Bartk, Stravinsky or Shostakovich we find quantities of
wonderful music that cannot accurately be described as beautiful. So let
me substitute for beauty the unsonorous artistic merit. Is Hanslick
right in the claim that expressive features are irrelevant to the artistic merit
of music? Up to a point. There is plenty of music that has expressive quali-
ties and is of no aesthetic merit. Further, possessing expressive features is
not, in general, a positive reason for value. For sometimes the expressive
features music has count against its value. It might be sentimental, for
instance, although we can easily imagine situations where sentimentality
might be a merit (for example, in an opera where a character is required to
display sentimentality). But, on the other hand, to say that music expres-
ses joy is to give a reason for valuing the music ceteris paribus. What we
can say, and Hanslick would probably deny this, is that music would be an
impoverished art if it did not have expressive features for then it would
cease to reflect or comment on our cultures in the way it can, something
that I illustrate later in this book. Part of the value that great music can
have for us lies in its expressive features and then it is not surprising that
some have thought that possession of expressive features must be a prima
facie merit. But such expressive features are neither necessary nor suffi-
cient for its merit. Such features can always be defeated by other consider-
ations. The music might be lively but poorly organized.
I remarked earlier that what is sometimes overlooked is the fact that
Hanslick allows expressive descriptions of music, if a little grudgingly.
Music can be arrogant, peevish, tender, spirited, yearning. We may use
such epithets, he says, ... provided we never lose sight of the fact that we
are using them only figuratively and take care not to say such things as
This music portrays arrogance.7 In fact most writers will agree that
when we say that music is yearning we use language metaphorically,
although, as I shall explain later, there are some difficulties here. His
reasons for denying that music can be literally yearning are generally
sound and represent the most advanced part of his case.
First, he argues that music is incapable of representing any specific
feelings because no judgements can be detected in the music.8 When I
yearn it is for something; there is, to put it in modern parlance, an object
of my yearning. We nowadays think of emotions as having intentional
objects; if I am angry, then I am angry about something and this
directedness, associated with the work of the philosopher and psycholo-
gist Franz Brentano, characterizes human mentality to a great extent.
Admittedly there are mental states that do not have any specific object and
often these are used, with less metaphorical weight, to describe music. But

18
 OVERTURE AND BEGINNINGS

Hanslick is right to claim that the specificity of emotions depends largely

upon the objects they have. Judgements enter the picture because of the
significance of beliefs in this equation. I have beliefs about the object that
are highly pertinent. That is, if I fear something or somebody, setting aside
cases where we speak of irrational or mistaken fear, then it is because I
correctly judge that that person or object threatens me. The difference
between fear and anger has more to do with the judgements about the
objects of these respective emotions and less to do with the feeling-tone
of these psychological states. It may not feel very different to be fright-
ened rather than angry but the judgements about the objects in view are
very different. In one case we are typically threatened and in the other
case characteristically injured in some way or other.
You may have noticed the use of the word represent here. Hanslick is
not particularly careful; sometimes he speaks of representing and some-
times of expressing.9 Music can represent, although its powers to do so are
limited. Haydns Creation begins with a famous representation of chaos,
largely through some highly unorthodox harmonic shifts. But we do not
naturally say that sad music represents sadness. Our misgivings about
the viability of this locution are partly due to the fact that x represents y
is usually a move in a discursive process: Let this salt cellar represent
Wellingtons army and this pepper pot Napoleons, then .... Undoubtedly
paintings represent their objects and there is no discursive process there,
save in interpreting the painting, which, as we shall see later, involves
further issues. But music is a temporal art and the analogies that are
relevant to any attempt to introduce the concept of representation into
music are discursive. So if we are told that a violin passage represents the
wife of the composer, we are inclined to say And then what? But since
music does not make statements, the point of representation is somewhat
oblique. All in all, we are better off with the vocabulary of expression.
A second argument offered by Hanslick is that some music seems to
have no expressive content. Bachs Forty-Eight Preludes and Fugues is
the example he offers.10 You might be a little uncertain as to whether this is
so, but no doubt examples can be found of music that it is very hard to
characterize in expressive terms. You might consider some parts of Bachs
The Art of Fugue to be music for which it is hard to find a suitable
expressive description. Or you might find examples in Pierre Boulezs Pli
selon pli.
Thirdly, he argues that there is music on whose merit we are agreed,
although we may differ on its expressive content. His example is the
music of Mozart, which is characterized differently by different critics;11
indeed, he might have cited the fact that the expressive qualities of the

19
PHILOSOPHY OF MUSIC

great Symphony in G minor are described in ways that are not mutually
compatible by a single author, Schumann, on different occasions. There
are other examples. Haydns Military Symphony might have sounded
ferocious and threatening to its contemporaries; to us it is inclined to
sound jolly. Of course, it will be said that this does not preclude one of
these descriptions being correct and the other incorrect, but that reply does
not cut the mustard. Where we have, as we do, a difference of judgement
about the expressive features of a piece of music by two equally attentive
and well-equipped critics, we have grounds for accepting both. Yet they
cannot both hold. In Chapter 4 we shall look in some detail at difficulties
of this kind.
Finally, the same music can express different feelings in different
contexts. The borrowings of Bach and Handel12 offer examples of secular
music being re-used to set sacred texts. The expressive nature of the text
differs radically but the music was deemed equally suitable for an erotic
or a religious text. This, of course, might not be an objection to the idea
that music is expressive. For if both of these texts express a more generic
state, say, contentment, then the music would be suitable for both texts
and its use in the two contexts unobjectionable. The argument does,
however, have purchase upon those who claim that music is very precise
in its expressive features. Mendelssohn famously said that what music
expresses is too precise for words and, to this thesis, Hanslicks argu-
ment is an effective counter.13 If erotic satisfaction and religious peace are
different emotions, then they cannot both be true descriptions of the
expressive features of the music. Now Bach and Handel, who after all
should know, thought the music appropriate for different texts. If they
were right the character of the music is only appropriate because of its
more general features such as contentment.
Of these four arguments, it is the first that has the most weight. In
Chapter 3 we shall consider how it may be answered.
The expression of feeling is not, Hanslick concludes, that which gives
music its merit, contrary to what many previous thinkers claimed. At most
music can depict the motion that attends an affect like love.14 The
beauty of music is a beauty of tones. Tones are its content and music is
peculiar in that content and form cannot be separated.15
So far we have been considering Hanslicks reservations about the idea
that music can represent feelings. But what about the thesis that musics
purpose is to arouse feelings? It is, after all, possible that music might
arouse feelings while not representing them or having them as content.
Now this cannot, in Hanslicks view, bear on the quality of the music. We
need to distinguish the quality of the music from the feelings that it

20
 OVERTURE AND BEGINNINGS

produces. In fact, he thinks of the appreciation of music in very different

terms from the mere occurrence of feelings stimulated by it. We grasp the
music when we follow it. In a justly famous description, Hanslick refers to
the mental satisfaction which the listener finds in continuously following
and anticipating the composers designs, here to be confirmed in his
expectations, there to be agreeably led astray.16 This is what is involved
in hearing the music for its own sake; it is not a matter of being in a warm
bath of sensations and feelings stimulated by the music. What Hanslick
scorns is an ecstatic wallowing in rather than a following of the music. If it
was only the sensations that mattered (and here Hanslick anticipates a
famous observation by Wittgenstein), then they might be obtained in some
other way, perhaps by a fine cigar.17 Curiously, he does in fact think that
the performer may lose herself in the act of performing so that the clich
of the artist as possessed by a mental state that she cannot but articulate
in the form of a work of art, holds of the pianist, say.18 In fact, such a
picture is equally silly as a description of the performer; she may not feel
the emotion the music expresses; indeed, an over-indulgence in the
sentiments of the music may distract; she, too, has to think about what she
is doing, work out how a piece should be performed and consider the
technical implications for fingering, etc. So Hanslick brusquely rebuts the
view that musics purpose is to arouse feelings; he remarks that beauty has
no purpose; it is mere form and remains beauty even if no feelings are
aroused. It is here that his position seems most resolutely formalist.
Hanslick is right to stress the difference between the properties of the
music and our response to it and the distinction certainly matters when we
come to those questions of value that are his main concern. There is a
difference between the value of the music and the value of the experience
we get out of it, for a start. We might think, and a surprising number of
people do, that the main purpose of composing music is to give others
pleasure, yet any thoughtful individual will recognize that there is music
that we admire very much but from which we dont get pleasure, unless
the word pleasure is to be stretched or substantially redefined. It seems
hard to deny that The Merry Widow offers its audience more in the way of
pleasure than the Goldberg Variations yet it would be absurd to rate the
first higher than the second. Equally, there isnt much doubt that Kind of
Blue is a superior work of art to Everybody Needs Somebody to Love as
sung by the Blues Brothers, but it seems pretty likely that the latter offers
the greater pleasure. Of course, this simply starts a debate. Perhaps there
are different sorts of pleasure. You will go on listening to Kind of Blue for
the rest of your life at regular intervals; you think about it, about Miles
Daviss use of modes, and you keep noticing new things about it.

21
PHILOSOPHY OF MUSIC

Everybody Needs Somebody to Love does not offer rewards of that

sort. These questions arise again even if I take two pieces of music that are
more comparable in stature. How do I compare the pleasure I take at the
subtle harmonic structure of Ravels Trio in A minor as against the intense
and intoxicating effect of Bachs Mass in B minor? (I return to some of
these issues in Chapter 4.)
Hanslick hoped to begin an exploration of music that would use the
methods of the natural sciences.19 Now the word science is often used in
a very general way; we speak of the human sciences and may not want
to commit ourselves to the idea that they follow the same procedures and
methods as the natural sciences such as physics, biology and chemistry.
But Hanslick unquestionably has the natural sciences as a paradigm; he
thinks of physiology and chemistry as models. If you think, as he does, of
a movement in music as developing from a melodic germ, then the model
of a crystal developing is perhaps a half-decent analogy. Hanslick empha-
sizes the fundamental importance of melody. All of this sits rather uneas-
ily with his remark that it is the spontaneity of art that makes it different
from physical experiment.
I suspect, although these things are difficult to demonstrate, that
Hanslick was an influence on the formalism of such musical thinkers as
Heinrich Schenker. Intellectual relationships are often mysterious and we
can be influenced indirectly through intermediaries even when we are
familiar with the origins of an idea. Schenker tried to demonstrate that the
unity of a complex movement can be displayed by showing it to be the
elaboration of a basic cadence, something he called the Urlinie; he did
this by stripping a movement down to its essentials so that a complex
movement such as the first movement of Beethovens Fifth Symphony can
be shown to be the elaboration of a basic triad of the home key. Hanslick
believed in the autonomy of music; his privileging of the tonal combina-
tions as the locus of artistic merit is what makes us want to call him a
formalist. He opposes any idea of music as a mixed medium and views
the music as what is important in the setting of words.

Edmund Gurney

The other major figure, prior to the developments of the very recent past,
who rewards study is Edmund Gurney (18471887). His The Power of
Sound is not read as much as it should be, partly, I suspect, because of its
sheer length. It is also repetitive and not well organized. Readers who
have managed the first couple of hundred pages may not be encouraged by

22
 OVERTURE AND BEGINNINGS

the observation it seems better to encounter the charge of tediousness

than of obscurity.20 Gurney was a Fellow of Classics at Trinity College,
Cambridge for a time and tried his hand unsuccessfully at careers in
music, medicine and law before becoming interested in the paranormal.
He spent the years before his premature death working on ways of testing
theories about telepathy, spiritualism and the like, although music seems
to have been a steady passion. This experimental bias made Gurney an
acute observer, a rather neglected merit in a philosopher. Remarks like his
observation on melody show this: a common (though very far from
universal) feature of a simple melody which is strongly approved on first
acquaintance is that, while it seems perfectly natural and easy to catch,
one cannot quite get hold of it afterwards ....21 Try playing by ear one
of Prokofievs more catchy melodies; even if you are skilled in this
particular art, it is pretty difficult to land in the right key. Consider, too,
the remark that when we remember a tune we dont usually recall it in its
instrumental colour.22 His prescient observation, following Hermann
Helmholtz, that the modern predominance of dissonant chords, and of
perpetual modulational change, threatens to destroy the all-important
feeling for tonality23 anticipates the musical developments of three
decades later. His scientific interests also surface in the way he so often
alludes to evolutionary explanations, albeit of a speculative form. He was,
I think, puzzled by the development of the ability to discriminate pitches
to a degree that is not required in the natural environment, although put-
ting this down to the pleasure it gives hardly qualifies as a Darwinian
explanation.
Like Hanslick, his discussion tends to make the unquestioned assump-
tion that what matters is beauty and that beauty is a matter of form. But to
a greater degree than Hanslick he anticipated contemporary interests in
the philosophy of the arts. For example, his discussion of what a work of
art is is very modern sounding. He proposes that a work of art is some-
thing designed by man with pleasure as its end.24 This leads him to discuss
what are nowadays called negative emotions, such as horror or dread of
death, whose stimulation by the arts seems a counter-example to the claim
that pleasure is what art is designed for. He is not very successful in handl-
ing these issues, but he is hardly alone in this: What is partially true of all
the arts is wholly true of this one that it must be judged by us directly in
relation to pleasure, and that pleasure is the only criterion by which we
can measure the relative worth of different specimens of it.25 He is scep-
tical about the possibility of getting a consensus on musical value; all
there is, it seems, is the pleasure of the individual. (I have already sug-
gested, and will elaborate later, the view that such emphasis on pleasure is

23
PHILOSOPHY OF MUSIC

a mistake; but his emphasis on the primacy of the individual judgement is

well founded.) Eventually he offers what we would now call a functional
theory of art by defining art in terms of the intention of its creators to
bring about particular states in its audience. He writes perceptively on the
importance of originality and on the creative imagination. The work of art
is in the most literal sense a creation and it possesses an organic unity.
He observes, like Hanslick, how different the arts are from one another;
indeed, he regularly compares and contrasts music and architecture.
Gurney is at pains to point out dissimilarities and certainly is sceptical of
the close parallels suggested by the adage that architecture is frozen
music.26
Musical analysis in the twentieth century was dominated, it is fair to
say, by the ideas of Schenker and his followers. Of course, Gurney knows
nothing of these later developments but it is easy to see that he would
disapprove of them. Melody is central to Gurneys conception of music
and, in this respect, he differs from Schenker. For harmony, thinks
Gurney, is subsidiary, whereas in Schenker the harmonic cadence is basic.
(Gurney is closer to the theoretician Rudolph Rti on this matter.) Our
interest, he says, is in the surface of the music. What matters is cogency
of sequence at each point,27 the enjoyment of the parts moment after
moment. Overall structure, musical form and design are less significant in
the pleasure of music. We learn, he thinks, very little from being told that
a piece is in sonata form or that this, that or the other is the second subject
other than having our attention drawn to noticeable points that may
strike us. This extreme view is not very plausible; not only does it deny
what seems to be the case, that our understanding of music is often
conducive to our appreciation and that holds even if we no longer
admire it as much as we did before the flaws were pointed out but it
leaves Gurney with a reduced role for the judgement that we often make,
that a piece has cogency and a satisfying overall structure. Consider the
opening movement of Haydns Quartet in D minor op. 76 no. 2, which is
an exploration of what can be done with a bare fifth: two notes five tones
apart. The answer is a very great deal indeed, in the hands of a master. But
if you are not aware of the way the music develops from this tiny germ,
much will pass you by. And what you lose is not merely an intellectual
grasp; this awareness enters into your very excitement and interest in the
progress of the music. Conversely, the overall effect of a piece may be
disjointed and to see this, and to make that judgement, is important.
Furthermore, ones judgement of the structural integrity of a piece may
involve noticing thematic, harmonic or rhythmic relationships between
movements as well as within a single movement.

24
 OVERTURE AND BEGINNINGS

Gurney introduces a rather unhelpful technical phrase to describe the

way a succession of notes (or tones or pitches, if you prefer) forms a unity
in which the parts are felt as indispensable, such that one has a sense of
movement from the beginning of the melody to its end. He calls this
musical process Ideal Motion.28 In this, the sense of movement and
rhythm is essential. Indeed some continuous groundwork of satisfied
expectation is a necessity. This aspect of music is obviously central and
has, as we shall see, been discussed by later thinkers. By contrast, har-
monic progression makes sense in the absence of rhythm.
Perhaps the best known and, probably, the most widely read section of
The Power of Sound is the chapter entitled Music as Impressive and
Music as Expressive. It is worth reading even if you find the rest of the
tome indigestible, although rather typically it is disorganized, the crucial
arguments appearing in comparatively terse prose at the end. Although
Gurney read widely, I have found no evidence that he knew Hanslicks
work, to which his own views show some resemblances. Like Hanslick, he
did not think that the beauty of music was in general a product of its
expressive features. Music may impress us while lacking expressive
features. In a footnote (like Edward Gibbon in The Decline and Fall of the
Roman Empire, the best of Gurney is often in his footnotes), he observes
that the way we characterize music may be distinct from the effect it has
on us, something on which I shall elaborate later in this book. Elsewhere
he remarks perceptively, mournful music, if sufficiently beautiful, may
make us happy.29
Gurney attacks the idea that music is an art of expression, as he puts it,
the reproduction in tone language of the feeling in the mind of the
composer, reproducing these feelings in the listener. It is not, in the
familiar phrase, a language of the emotions. He offers a battery of argu-
ments that counter the idea that the beauty of music comes from its
expressive features, several of which are familiar from Hanslick.

Much beautiful music has no definable expressive character. In

general, the character of music is extremely indefinite.30 It cannot
express pity, let alone the sort of pity we feel for Antigone as opposed
to the kind we feel for Hamlet.
Where it does have a definite character, it may have no effect upon us.
People differ over the expressive character of music. The opening of
Schuberts Piano Trio No. 1 in B flat is variously described by differ-
ent critics as, on the one hand, energetic and passionate and, on the
other, as girlish and confiding, although they may concur that it is
enjoyable.

25
PHILOSOPHY OF MUSIC

If music produces in us a definite emotion and if this stirring of the

emotion is the whole business of [it],31 what is to be gained by
listening to other music with the same character?
If the stirring of emotion is the point, why not by some other person
than by music? If we want to experience an emotion, there are all
sorts of other, more foolproof ways of experiencing it.
There is no reason to suppose that the character of a piece of music
depends on the mental state of its composer. A musician may
compose a funeral march without losing his good spirits and the
prolonged labour involved means that it is implausible to suppose
that he is in that state for the entire time of composition.32

How, then, do we describe, perfectly aptly, music as stormy, tragic and

the like? Gurney suggests that the connection between music and its
expressiveness is most easily seen in the matter of pace. A slow pace is
often solemn; on the other hand, confidence is expressed by a strong
rhythmical outline at a quickish pace. Pace is so characteristic of emo-
tional speech that the corresponding features in music owe some of their
quality to it.33 The higher registers of the voice are more penetrating and
so have been used to attract attention or give force to an utterance.34 But
Gurney does not think that such expressive features are that important.
Our interest lies elsewhere and we have to force ourselves to consider
those aspects of the music that are expressive. Indeed, although we can
describe music in such terms, the excitement that music creates in us is
something sui generis. For how could mournfulness or exultation explain
or guarantee the beauty of a tune?
This conviction that music is a non-representational art existing in
comparative detachment from the others is what is most striking about
Gurneys position.35 He is an autonomist. The musical mood is unique;
about the most we can say of it is that it is simply that unique one which
we can but describe as the mood of musical exaltation.36 He frequently
remarks on the isolation and uniqueness of our experience of music and
his reservations about the value of musical expressiveness underline this.
If Hanslick is a formalist, Gurney is one too.
We shall return to these issues later. For now let me say that I think
music is more deeply involved in the culture from which it emerges than
autonomism suggests. Certainly Victorian music doesnt tell us about
Victorian life and culture in the way that the novels of Dickens or George
Eliot do. You wont find any helpful diagnosis of the limits of utilitarian-
ism in Edward Elgars early music. But the range of expressive descrip-
tions is constrained by the culture of origin; and remember that neither

26
 OVERTURE AND BEGINNINGS

Gurney nor Hanslick deny the aptness of such descriptions. It matters to

us that we can object to the sanctimoniousness of Elgars early religious
music and it is significant that this is a description that we could not
properly apply to religious music of the sixteenth century.

The legacy

Hanslick and Gurney bequeathed to us what nowadays is called, rather

pretentiously, a problematic and this sets the scene for Chapter 3. His-
torically the central question in the philosophy of music over the past two
and a half centuries has been this: is music a language and, more specific-
ally, is it a language of the emotions or is it an art of form alone and its
merits merely those of artfully composed patterns? In attendance come a
series of subsidiary questions. How does the beauty of music relate to its
expressive power? How can we explain its capacity to express? If it is not
by expressing the mind of the composer or producing correlative effects in
the listener, then what entitles us to call it grave or spirited? But the
most fundamental difference lies between those who think of music as an
autonomous, abstract art form, and those who think that it draws its
importance for us from connections, no matter how tenuous, with our
lives outside the world of music. On the whole I side with the latter.

27
Chapter 2

The work of music

Beethoven was more important than whoever was playing him.

Phil Spector

For the philosopher, two questions about what it is for something to be a

work of music spring to mind. First, music is described as one of the fine
arts. So what is it that makes music or, to be precise, some pieces of
music works of art or examples of fine art? This is our first question. The
second is what is the work of music per se? Is it just what we hear, the
performance, or is it something over and above that? Is it something we
invent or something we discover? These questions occupy the second part
of this chapter.
Now you might ask why we should need to answer the question: what
is a work of art? This is, after all, a book about the philosophy of a specific
art, music. Work of art and fine art are used in quite broad and not
very well-differentiated senses. Music is sometimes described as one of
the fine arts, but the fine arts also cover Faberg eggs, Staffordshire
pottery and the sort of things found in the Victoria and Albert Museum. If
we describe the latter as works of art then we are likely to be grading
them not only as exceptional examples of their type but as something
more. For works of art have expressive features such as sadness, optim-
ism, joy and such like. Works of art may be loosely said to have mean-
ing. Paradigmatically, they are about something in the sense in which,
for example, Jane Eyre is about the moral growth and liberation of a
young woman. This makes works of art something far more than decor-
ative. They have implications for our lives and the way we judge others.
We learn morality in part from works of fiction, plays and films. If music
is to be included in the family of works of art, then we need to think

29
PHILOSOPHY OF MUSIC

about what work of art means and whether it can be extended to an art
that seems not to have moral content. We need to distinguish music that is
art music from hymns, advertising jingles, work songs and football
songs. Once we use the term work of art to describe something, the
range of descriptions that are appropriate is immediately expanded. A
jingle can be jolly, but once we describe it as ironic we are already in the
foothills of art.

What makes music art?

Hard cases

To begin at the beginning, why do we think of some pieces of music as

works of art? The question is quite general. What, after all, makes a paint-
ing or a sculpture a work of art? The question is very familiar in recent
writing on the philosophy of art; it has been the centre of attention over
the past four decades. Yet it might strike you as strange, largely because
the question as put is contextless (as philosophical or, perhaps more accur-
ately, metaphysical questions characteristically are): in some contexts, at
least, asking such a question would be absurd. Certain sorts of contexts
are surely required for the question to be sensible. We certainly cannot
imagine somebody au fait with the English language and who is not a
philosopher asking of a Gainsborough What reason have we for calling
this a painting? any more than she is likely to ask What reason have we
for calling this blue? It would be equally odd to ask the question What
reason have we for calling this a work of art? of an exhibit in the National
Gallery, although it might be less puzzling if we were looking at some-
thing in the Museum of Modern Art.
Why has the question What is it to be a work of art? taken centre
stage? I believe that it is because of what I shall call the problem of the
hard cases. It is the hard cases concept art, aleatoric music (music
involving the selection of sounds by chance) and the like (notably cases in
the visual arts and music rather than examples from literature) that have
caused a massive academic industry to be employed in trying to analyse the
concept work of art, one consequence of which is that philosophy of art
has rather edged out general aesthetics in the past few decades. There has
been comparatively little written on the aesthetics of natural beauty or
fashion, or on the decorative arts and crafts, compared to the concentration
on the notion of art. The difficult, or what I shall continue to call the hard
cases, begin with the use of unfamiliar materials. And here the problems

30
 THE WORK OF MUSIC

are, I think, less striking. If you make sculptures out of elephant dung
instead of marble, there seems less reason to hesitate about calling the
result art than there is when you recycle bicycle parts. One reason is
connected with a deep-rooted assumption that skill is a requisite in the
production of art; the amount of skill required in the latter seems minimal
whereas I presume that moulding elephant dung requires technique,
although I suppose the fact that elephant dung is more malleable than
marble makes the technique less exacting. Certainly a mistake is more
easily corrected, although the same is true of some other forms of art;
corrections are easy in writing poetry or music and paintings have
pentimenti.
But the standard examples that take pride of place in modern discus-
sions seem to require no technical ability at all. Thus Marcel Duchampss
Fountain, an inverted urinal signed R. Mutt, does not require the
technical ability of a Rodin or Bernini. Tracey Emins My Bed and Andy
Warhols Brillo Boxes are other examples, as, indeed, are most of the
shortlist for the Turner Prize. Lets turn to music. Takehisa Kosugis Micro
I requires paper to be wrapped around a microphone. The crackles created
as it unfolds constitute the music. But the standard example in music is
John Cages 4'33", the celebrated silent music, which has come again to
public attention because his estate made some legal enquiries about a
silent track on a disc by another musician:

In June, after the British musical group the Planets introduced a 60-
second piece of complete silence on its latest album, representatives
of the estate of composer John Cage, who once wrote 4'33" (273
seconds of silence), threatened to sue the group for ripping off Cage
(but failed, said the group, to specify which 60 of the 273 seconds it
thought had been pilfered). Said Mike Batt of the Planets; Mine is a
much better silent piece. I (am) able to say in one minute what (took
Cage) four minutes and 33 seconds. (Independent, 21 June 2002)

In Cages composition, it seems that the music is supposed to be

whatever ambient sounds occur during this interval. Perhaps this is a new
kind of music that somehow replaces and at the same time challenges
traditional music. Cages general justification of his position suggests that
he might have benefited from the intellectual rigour that a first-year
philosophy course can offer.

I said that since the sounds were just sounds, this gave people hearing
them the chance to be people, centred within themselves where they

31
PHILOSOPHY OF MUSIC

actually are, not off artificially in the distance as they are accustomed
to be, trying to figure out what is being said by some artist by means
of sounds. Finally I said that the purpose of this purposeless music
would be achieved if people learned to listen; that when they listened
they might discover that they preferred the sound of everyday life to
the ones they would presently hear in the musical program; that that
was alright as far as I was concerned.1

I suppose this is hardly worth arguing with (although connoisseurs of

fashionable egocentric twaddle will love the idea that hearing sounds
gave people hearing them the chance to be people, centred within them-
selves where they actually are). But, of the last few sentences, where
Cage deviates into sense, I am inclined to say that few would travel miles
and pay good money to hear everyday sounds and for very good reason.
Of everyday sounds, a few most of us like, many we dislike very much
and most we dont care about. I like the sound of birdsong in my garden.
But, like other people who are comfortable living in the country, there is
much I dont like: the sound of crows and jays, the bleating of sheep and
the roar of low-flying aircraft. I dont like the sound of vomiting either.
Cage does not give us any reason for preferring other sounds to music and,
indeed, we should be surprised were he able. He wants us to direct the
concentrated attention normally required of music to these everyday
sounds but, although it is conceivable that people might come to prefer
these sounds, he gives no grounds for supposing they would nor any for
supposing this is not a dereliction of taste rather than any sort of advance.
The problems of silent music hardly stop here. Could it be performed?
Each performance would probably sound very different from the previ-
ous one, so that whereas I can easily recognize two different performances
at different tempi of Beethovens Eroica Symphony as performances of
the same work and not of Beethovens Fourth Symphony, I certainly
would not be able to distinguish a performance of 4'33" from the perform-
ance of another piece of silent music that lasts 4 minutes 34 seconds
unless I knew when it started and stopped and was equipped with an
accurate chronometer. (Could the performer of silent music opt for a
slower tempo?)
In any case, the aesthetic appreciation of ambient sounds is closer to the
appreciation of natural beauty than the appreciation of art. We might be
struck by the beauty of one oak tree among a group of less interesting trees
just as we might be struck by the beauty of one man or one woman in a
crowd of less attractive people. But in all sorts of ways this diverges from
our attraction to art. Where are the roles of education and the development

32
 THE WORK OF MUSIC

of taste? How does our preference here connect with other preferences
through our lives? Our taste in opera, say, or drama or painting, will
probably show some consistencies with our moral values and religious
opinions, for example. A love of Ravel or of Thelonius Monk says much
more about somebody than her love of gherkins and rather more than her
love of the Yorkshire Dales as opposed to the Scottish Highlands.
A potent objection to conceptual art might very well be that these
artists just cannot do the technical stuff, one sign, perhaps, of the problem
of the fraudulent that Stanley Cavell thinks, is both so significant and so
elusive when we are considering the contemporary arts.2 Even David
Hockney is sufficiently mystified by the draughtsmanship of Holbein and
others to suggest they must have used the camera obscura to produce such
verisimilitude. We know that Picasso, for example, was more than compe-
tent in the traditional skills and that makes us ready to accept what strikes
the uninitiated as the apparent crudities of some of his work. I have heard
it argued that a sculptor getting somebody else to do the modelling that he
cannot do himself is no different from the fact that, in all probability,
much of Michaelangelos work was produced by an army of apprentices
working in his atelier. (Hogarth objected to the creation of artistic factor-
ies and took a pride in doing everything himself, including priming the
canvas.) No doubt, but there is also no reason to think that Michaelangelo
could not also do the job had he time and every reason to suppose that he
went around touching up and correcting the work of his employees.
Unless he could and did, we might as well credit Pope Julius with the
Sistine Chapel ceiling. Composition technique is something that com-
posers need; harmony and counterpoint have to be learnt. Of course, it is
true that they cannot play all the instruments they write for Mozart was a
brilliant pianist and good violinist but he did not play the horn, as far as I
am aware but orchestral technique requires knowing the possibilities
and limitations of the instruments written for. It is difficult to think of a
significant composer who was not competent in this and only a few come
to mind who were not decent keyboard players. For a start, all major
composers have been able performers. Certainly, there is music that seems
to require no skill at all in composition such as aleatory music or silent
music (if you count this as music), but most of the music found difficult
by the contemporary non-specialist listener is music that requires a great
deal of technical mastery to compose, such as the music of Elliott Carter,
Pierre Boulez, Karlheinz Stockhausen or Harrison Birtwistle. It is music
that is difficult because we find it hard or impossible to follow in the
way in which we follow earlier music. Even minimalist music usually
turns out to be a more complex business than it at first appears. It certainly

33
PHILOSOPHY OF MUSIC

can be difficult to perform since it requires enormous rhythmical

precision. Most contemporary music does not constitute hard cases in the
sense in which I have used the term. On the whole, anxiety as to the
technical competence of the creator is less a consideration in the case of
contemporary music.
It is a familiar point that the term work of art is used in a special way
in philosophical discussion. Outside philosophy, we would use it of
painting, sculpture and things like Greek vases. We also use it to describe
what we find in the Metropolitan Museum of Art or the Wallace Collec-
tion. We dont normally use it to refer to individual pieces of music and
poetry. (It is also worth noting that German lieder, French mlodies, or the
English school of song writing is described as art-song, presumably in
distinction from folk song.) But the philosophical usage is justifiable.
We need some generic term to cover the group of man-made objects
whose fuller appreciation requires special preparation and which we value
above most other things that man makes or does. That the use of work of
art as a generic term by philosophers is not an entirely arbitrary or idio-
syncratic usage is shown by what is covered in the arts pages of the broad-
sheet newspapers, and by the existence of various organizations for the
arts.
So this usage is not purely down to philosophers. But neither is it
wholly down to people who write about and finance the arts rather than
create art. Although Cage was prone to speak of the music he and others
created, he also spoke of art and himself as an artist, significantly in
the context of his silent music or in one of a set of variations where the
audience is told to wander away and listen to ambient sounds. Robert
Rauschenberg spoke of his erasure of a drawing by Willem de Kooning as
a work of art.3 So, and perhaps this is significant, work of art is often
used to describe the hard cases by those who produced them. The salient
fact is that both the extension and the intension of work of art differ
from the extension of pieces of music, poem, painting and so on. I
shall suggest later that the decision to count music as art has pretty
profound implications for the way we treat it.
I said that only some music would count as art. Nobody would
quarrel with the suggestion that Mozarts Don Giovanni or Bachs Forty-
Eight Preludes and Fugues count as what we might call art music. So
does Indian classical music, Miles Daviss Kind of Blue, the music of
Thelonius Monk or the Modern Jazz Quartet (MJQ). What about the
Beatles? Well, probably most of Sergeant Pepper counts: certainly A
Day in the Life. What about the operettas of Gilbert and Sullivan? Do
we include West Side Story but exclude South Pacific? Porgy and Bess

34
 THE WORK OF MUSIC

gets in. Muzak certainly does not. We wont include footballs World
Cup songs, advertising jingles, most hymns and nursery rhymes. Do we
include music by Dire Straits or folk songs? That is somewhat moot.

Art music and philosophical analysis

In the next part of this chapter I am going to examine some general

accounts of why something counts as a work of art to see whether the
question of when a piece of music counts as art music can be clarified
thereby. But perhaps I should say in advance that I dont think this enquiry
will, in the end, help us to decide on the hard cases. I am going to discuss
four approaches: procedural, functional, historical and recognitional. My
conclusion will be that each of them explains some cases but that none of
them work for all cases and that problems remain with the hard cases.
All of these are attempts at analysis. The goal is to produce a set of
necessary and sufficient conditions by reference to which we can establish
when something is a work of art and when not. This programme of analysis
has been part and parcel of the tradition of philosophizing in the twentieth
century; for instance, there have been attempts at analysing the concept of
knowledge so that we can see how it differs from belief. (The initial sug-
gestion was that knowledge is belief that is true and justified. That thesis
has been shown to fail.) Now there are some reasons to be wary of attempts
at analysis. In the case of art, we have to say that despite years of activity
there is no agreed analysis. This does not show that analysis is impossible,
of course somebody may always turn up with the goods but it suggests
that we should be cautious about the prospects of success. It may be that art
is an essentially contested concept,4 one that is always going to be the
locus of debate in part because its function is to serve as a battleground for
disputing parties. There is, for example, always likely to be conflict
between conservatives and the Young Turks as to whether a certain piece of
avant-garde art counts as art at all. We can see this process in action in the
debates over the hard cases I mentioned earlier. Indeed, I suspect that
understanding the concept of art is as much like understanding the sorts of
arguments admissible in a court of law as it is knowing what the verdict is.
That is, you understand the concept of art when you understand what
considerations lead people to call something art.
Secondly, there are cases where it cannot be clear whether the concept
applies or not. Usage gives us no idea how to classify these cases so the
concept is indeterminate at this point. It is pretty evident that there are
many hard cases where we are not sure whether we are dealing with art or

35
PHILOSOPHY OF MUSIC

not. Nor can we be quite sure where the boundaries lie that distinguish art
from entertainment, kitsch or pornography. I am not sure whether the grif-
fin handles that are all that remain of a seventh century BCE Greek pot
count as art or craft. We might give reasons either way but these reasons
will culminate in a decision if they culminate in anything at all. We might
even make an arbitrary decision. But in both of these latter cases it will be
a decision and not an analysis and that rather defeats the point of the
enquiry. Wittgenstein puts it well: It is only in normal cases that the use
of a word is clearly prescribed; we know, are in no doubt, what to say in
this or that case. The more abnormal the case, the more doubtful it
becomes what we are to say.5

Proceduralism
A distinction is currently made between what are called procedural and
functionalist accounts of art.6 The distinction roughly is between those
theories that make arthood a status that is conferred upon artefacts by those
qualified to do so (which may include the creator or bona fide critics) and
those theories that view art as a group of artefacts that have a special point
or function (they do something for us). The two may not be independent.
For example, we may classify something as art because it is already seen to
have a function of the sort common in art or it may be that the act of
classifying it as art leads us to look for these sorts of functions or even that,
as a result of classifying it so, it takes on some functions (it becomes an
object whose meaning and significance we debate, for example). For the
time being I want to concentrate on proceduralism.
What is called proceduralism differs little from what was previously
called institutionalism. Roughly speaking, and a rough characterization
is all that is needed for the moment, the idea is that art is art because the
art-world or, in the limiting case, a member of it, decides so. A work of art
is an artefact presented to the public for their appreciation by a repre-
sentative of the art-world or an agent for it. This account has been vari-
ously modified and has aroused a great deal of comment. I shall not repeat
the excellent discussions available elsewhere. What I am more interested
in here is the way the weaknesses of institutionalism reflect more general
problems in the programme of attempting to secure an analysis of the
concept of art.
The two difficulties with the institutional theory that are most often
aired are: (i) the problem of circularity; and (ii) the problem of what is
called first art. As far as (i) is concerned, it is argued that it is hard to see
how we can get the notion of the art-world off the ground without a
reference to art. Since art is then defined in terms that refer to the art-

36
 THE WORK OF MUSIC

world the circularity looks both inevitable and dangerous. To some, the
question is not important, especially for those who reflect on the parallel
difficulties of defining moral or religious or spiritual in terms that
are entirely independent of the original notions. After all, perhaps we can
build up a concept of the art-world via examples and instances that may be
sufficient for us then to be able to specify the concept of art well enough. I
want to spend a little more time on the question of first art because it is
both a more interesting question and somewhat germane to subsequent
discussion.
First art covers those artefacts that were created before anybody
possessed anything like our concept of art. The Altamira cave paintings (in
France) are standard examples. They were painted 11,00019,000 years
ago. It is unlikely that those communities had our concept of art. We count
these as art although their creators might have counted them as having
magical powers and created them for that purpose or as votive offerings.
There is certainly no reason to suppose that they thought of them as works
of art, although it is highly likely that their contemporaries were as struck
by their grace and beauty as we are. At some point in the history of the
West, probably in the Renaissance, the modern concept of art began to
emerge and, from the outset, gathered under it those artefacts, paintings,
sculptures, poems and so on that were created shortly before and during its
emergence, together with a selection of classical sculptures and plays. The
notion of music as a fine art seems to have developed in the eighteenth
century as a response to the development of what used to be called pure
or absolute music. The problem for the proceduralist is that we want to
say that a Mass by Josquin des Pres dating from the early sixteenth century
counts as art music even though there was then no concept of art as we
understand it and, a fortiori, no concept of art music. How, then, can its
status as art music depend upon its being so christened?

Functionalism
Functionalism7 seems to avoid the difficulty of first art. What the
functionalist will say is that the artefact is intended to produce a specified
result; and such an account neither needs reference to previous works of
art nor seems to require the concept of art to get off the ground. Function-
alism may also have a condition of success built into the rubric; then
works of art are works of art because of a certain function that they fulfill.
However, I shall concentrate on versions that build in the notion of
intention. We need a concept of art that allows that some works of art
are failures. The most obvious candidate for this has seemed to be pleas-
ure. The old phrase to please and instruct gives the basis for a functional

37
PHILOSOPHY OF MUSIC

definition of art. Concentrating on the first part of this rubric, music

counts as art if it was intended to please. Another function it might serve is
that of being intended to generate aesthetic experiences. But then, of
course, we must define aesthetic experience before this analysis has any
value. So the threat of regression is not immediately banished. (In as much
as institutional theories speak of appreciation, they introduce a func-
tionalist element. They say the work of art is intended to be appreciated;
indeed, the relative thinness of the characterization of appreciation in
institutionalist theories has aroused some adverse comment.)
So far, the discussion has been very general; as I have described it,
functionalism is a theory that applies to any of the arts. But in talking
about functionalism, we have focused on an aspect of particular interest in
music and for this reason I propose to devote more space to it in this book.
More than in any other art, in my experience, the value of music has been
seen as a matter of the pleasure it gives. Pleasure has thus been taken to be
central to this art. But the supposition that pleasure is basic to assessment
certainly raises problems in the other arts. Recent work for television by
Rob Brydon (Marion and Geoff), Ricky Gervais (The Office), Steve
Coogan (Im Alan Partridge) and others is rightly admired but exceed-
ingly uncomfortable to watch; we squirm with embarrassment. We watch
these dramas for their brilliant characterization and timing, but not, I
think, because they give pleasure. We admire the achievement but find it
difficult to watch all the way through. Because the other arts have tradi-
tionally been thought to have, inter alia, a moral purpose and may be good
for us, that we do not enjoy a play or a novel seems to present no particular
difficulty. But this consideration has never been very plausible in music,
which cannot, even indirectly, embody a homily. Perhaps partly as a
consequence, music has, in my view wrongly, been thought to aim at and
be justified by the pleasure it gives. In Chapter 4 I shall argue that we need
a contrast between taking pleasure in and taking an interest in music.
As far as functionalism is concerned, we can look outside the philoso-
phy of art for clear guidelines as to how to approach this theory. For the
paradigms of function are to be found in biology and in man-made mach-
ines. The function of the heart is to pump blood around the body. The
organ has this function within a larger unity: the body of the creature. If it
fails the creature dies; if it fails to function properly, then the creatures
capacities will be much impaired. It may find that it cannot move quickly
enough, for example. In the car, the function of the carburettor is to mix
petrol and air so as to form an explosive mixture. If it fails to function
properly, the car will not run evenly or, in the last event, not run at all.
Importantly the concept of function has been extended to the social

38
 THE WORK OF MUSIC

sciences with the claim that the way to understand social practices and
institutions is to understand how they operate in keeping society in being.
There are some analogies in the arts, although these analogies do not
play any role in the present debate about functionalism. W. H. Auden used
to teach schoolchildren about poetry by replacing a word in a line with a
near synonym: the most effective way of showing the work that a
particular word does. Change damages. The perceptive critic of music will
show me how a particular harmonic shift operates in the course of a work
as a whole. Its absence may damage the work or render a passage
incomprehensible. But, as I said, these are not considerations that the
functionalist in philosophy of art is concerned with. For he is concerned
with the function of the work of art as a whole. (In fact he removes
functionalism from the area in which it might have some insights to offer
into an area where it does not.) Given this, the most plausible reading of
functionalism as a theory of art is to conclude that individual works of art
are analogous to tools. Their specific function has been variously
described as being to reward disinterested attention, or to produce
aesthetic experience or aesthetic pleasure.
Now for the objections. If we define the function in terms of either
aesthetic experience, disinterested pleasure or aesthetic pleasure we have
an unduly narrow view of what the creator might have intended his work
to do. According to some recent writers, the intention of Wagner in The
Mastersingers of Nuremberg was to call German people to an appreci-
ation of their community. Certainly Dickens and mile Zola had social
reform in view. Some of Bachs church music was, we are told, intended
to harrow the souls of the congregation.
But even if we could find a more plausible general account than one in
terms of pleasure or aesthetic experience, we face a major obstacle. A tool
has the proper function its maker intended. But the thesis that a work of art
has the function its maker intended is too close to the thesis that it has the
meaning that its creator intended. We shall point out that on many
occasions a work of art comes to function in a way or have a meaning that
the creator might not have known about and so could not have intended.
Bach would have been surprised to find audiences in concert halls listen-
ing to his cantatas. The problems are even more obvious in literature.
Lewis Carroll would have had second thoughts about publishing had he
known the Freudian interpretations placed upon the Alice stories (and
these interpretations can certainly be defended). If Stanley Fish is right,
Paradise Lost was intended to recall backsliders to a firmer faith in Christ;
it was intended to have a didactic function. But it is not read that way now.
Should Brittens The Turn of the Screw be seen as a warning against

39
PHILOSOPHY OF MUSIC

submitting to the sexual attractions of children? If it is, it wont have a

very wide application.
Of course, I can use a tool in a way the inventor did not intend; a tyre
lever can be used as an offensive weapon. These are, of course, improper
uses. There are also improper uses for art. To play The Rite of Spring very
loudly to annoy the neighbours, or to pick a Howard Hodgkin to complete
the decor or to use the full score of Der Rosenkavalier to stop a table
wobbling are all improper uses. But these are peripheral cases and the
admonition that the work of art is being used improperly is presumably
jocular. But it is not an improper use of a work of art to interpret it in a
way that its creator did not intend, even if the interpretation is
indefensible. The problem, then, with the analogy between a work of art
and a tool is that whereas it makes perfect sense to say of any use of a tool
that was not intended by the creator or manufacturer that it is an improper
use of that tool, if the work of art was intended, like Bachs church
cantatas or Miltons verse, to have a certain function, the setting aside of
that function is not regarded as an improper use of the work of art. (An
improper use of a tool, of course, might well be justified if nothing else is
available and the matter is urgent.) In saying all this, I am not down-
grading the liturgical or didactic use of art; the point is that qua work of
art, considerations other than utility enter.
Still, where an artefact does not belong to the central forms of the arts,
the decision to classify it as a work of art may be a consequence of the fact
that it works in the same way as standard works of art. When we decide
to include jazz and some world music or some musicals (a very few) as
works of art, we do so because we find that we respond in the same way to
them as we do to standard works of art. Thus a basically procedural
account may employ elements of functionalism. In some of these cases we
may be too ignorant of the culture in which they originate to be able to talk
intelligently about their meaning. For instance, we may know very little
about the significance of some African music and its place in society.
Critical debates over its interpretation cannot even begin. So it is, as a
matter of fact, very likely to be our sense of the aesthetic qualities of bal-
ance and design, melody and shape that matter here. We then naturally
infer that the creator intended to appeal to that sense and, in very many
cases, we are surely right to do so. So functionalism certainly covers some
of the cases where works of art become accepted as art, but only some, I
think.
But perhaps the most striking problem is this. Good art, and especially
great art, is valuable precisely to the degree its qualities are non-generic. It
is a criticism of, say, poetry, that its effects are generalized. What we look

40
 THE WORK OF MUSIC

for in art is the individual and idiosyncratic. Even where we think of a

painter like Raphael or a composer like Mozart as displaying the qualities
of the quotidian style in a canonic way we still note their individuality.
Classicism does not exclude it. Mozarts best works are undoubtedly
Mozarts. Nobody who really knows about classical music would mistake
a piece by Mozart for one by Haydn; even where Haydn, in tribute, seems
to imitate Mozarts style, we still know it is the authentic Haydn manner
we hear. So it is a criticism of a work that it merely shares the general
features with others of its genre. What we look for in art is the individual
and those writers who stress the effect of art upon us rightly emphasize
how important it is that the response mirrors the uniqueness to be found in
the work. Indeed, if you are going to embrace any form of functionalism,
you must emphasize the idiosyncrasy of the work and of its correlative
effects. These are insurmountable obstacles in the path of a general account
of the function of art, especially when one is dealing with what is out-
standing and in the philosophy of art it is what is outstanding that is of
overall importance in determining the nature of art and of our response
to it.
There is a parallel with familiar thoughts elsewhere in the philosophy
of art. We shall see later that the notion of judging the value of a work of
art with reference to criteria fails because, if the critical standards are to be
of general applicability, they are too vague to be of help. The more precise
we make them, the easier it is to find counter-examples, works of agreed
accomplishment that infringe these standards. Equally, any plausible
account of the function of art will be too generalized to be informative;
make it more exact and many successful works of art will be omitted by
the formula.
Turning, then, to our experience of art, I remarked that the functionalist
has the problem of describing the experience in a way independent of the
work of which it is an experience. This is something quite widely
acknowledged. Wittgenstein famously observed, If I admire a minuet I
cant say: Take another. It does the same thing.8 This is a counsel of
perfection though. A celebrated musician once suggested replacing some
of the weaker movements in early Haydn with some of the stronger
movements in the same set of quartets. Whether anybody would have the
nerve to do so in the current climate I rather doubt. However, Wittgen-
steins point was not about what liberties we may or may not take in
classical music. It has been taken as suggesting that our experience is an
experience of one particular and individual piece. The strong individuality
of artistic experience comes out in the way that the experience that the
work produces is thought to be unique to that work. Now that view is not

41
PHILOSOPHY OF MUSIC

quite accurate. It is not that our experience is different on every occasion;

it isnt. The sense of exhilaration produced by one Haydn finale may not
be much different from that produced by another. Wittgensteins aphorism
should be taken differently; it is that what should matter to us is the work
of art itself and our attention should be focused upon it and not upon our
experience. The thesis should be understood as a covert thesis about the
attention we pay and the attention we pay should, ideally, enable us to
recognize the individuality of the work. And we wont know what the
peculiarity of the work is until we see, read or listen to it. Now, we select a
tool for a certain job and any other tool will do as well providing it does
the same job. But art is not like that. Indeed, what we find here reflects the
thesis associated with the philosopher R. G. Collingwood. He argued that
the creator does not know what he accomplishes until the work of creation
is complete, and that this distinguishes art from other activities like craft.
It is the conceiving of the work of art that is the real creative act. My
suggestion is that, in parallel, you dont know what the character of a
work is until you experience it. Having it described to you is, as we all
know, no substitute. This holds whether or not the phenomenology of
listening to the Minuet of one of Haydns Paris Symphonies is identical
with the phenomenology of listening to the Minuet from another. The
quality of our conscious experience may not vary. (The study of phenom-
enology is the study of the nature of consciousness in abstraction from
what it is consciousness of.) And with great works, it does not stop there;
you might think that when we read or hear or see it again we know what to
expect, but this is not necessarily so; when we encounter it again we may
see different things in it. What I have described is, of course, a feature of
outstanding music. Plenty of mediocre music sounds like other uninspir-
ing music. Perhaps unfairly, Vivaldi is said to have composed the same
concerto six hundred times. Whatever is the case here, there is competent
music, like that of Telemann, that shows no strong individuality. Haydn,
Bach, Mozart and the like are different, and it is the individuality of their
voices that we prize. While we continue to prize art for this, our concept of
art will not be one that can be satisfactorily defined in terms of function.
For the best we can say is that the function of this piece of music is to be
just what this piece of music happens to be.
I have no overall purpose in settling down to listen to some new music.
Indeed, it is the work itself that calls for my attention, not what I
experience when hearing it. This is just what it is to be interested in music.
Certainly, on occasion, you may look forward to a performance, anticipat-
ing pleasure, and be disappointed if you are bored, but not all encounters
are like this. It is not to be doubted that an individual piece of music or an

42
 THE WORK OF MUSIC

individual poem may be designed to please. It may be produced with that

function in mind. The point is that this is implausible as a general thesis.
Not only do most major works of art not set out to please, but the creator
nowadays will probably expect divergent reactions from the public. In any
case, whether or not he or she expects it, that is going to be the case, and
the greater the stature of the work the more likely it will be. If it pleases
and is good, there is usually more to be said. Even Die Fledermaus has
an edge.
Indeed, setting aside the composition of music merely to entertain, it is
often a grave charge against a work of art that it has a function. When
artists have such an intention then we are alarmed. What we resent is
propaganda. The ideologically or religiously loaded work can offend
although much depends upon its social and cultural background. I dont
object to Mariolatry in a fifteenth-century English composer but find the
religious sub-text of Elgars The Dream of Gerontius or the work of
Olivier Messiaen or Arvo Prt, John Taverner and other holy minimal-
ists somewhat rebarbative.

An historical account
Consider now a third account. Suppose an artefact counts as a work of art
in as much as it was intended by its creator to belong to a tradition of
artefacts. We can describe this as an historical account.9 This could even
incorporate a functionalist element so that the artefact is expected to have
the same sort of function as its ancestors did. This theory accounts for art
in terms of the intentions of the creator, intentions that themselves are
controlled by an historical sequence or tradition of works of art; a work of
art is something made for regard or treatment as previous artworks have
been regarded or treated. An immediate objection is rather obvious. The
account looks either regressive or circular for it seems to be a fault in the
historical account of art that it, too, cannot cope with first art. The first
work of art cannot aspire to the status of its predecessors for there are
none. But if all that we require is that when an artist intends to produce an
object that has the functions that previous works of art have, he thinks of
his work as belonging to the tradition to which these sculptures, painting
and poems belong, then there is no especial problem. It is not required that
the artist has the concept of art as opposed to some concept of the genre in
which he works. And he may very well even have a concept of the genre
without fully articulating it. So somebody fashions a flute from a bone
and finds that covering holes in it allows them to alter the pitch. At this
stage we have something that could be called music but we dont have
anything that contemporaries call a work of art or art music. They

43
PHILOSOPHY OF MUSIC

dont have those concepts. These are concepts that evolve as other factors
begin to intercede. Players practise and the art becomes more complex.
(Imagine how revolutionary it must have been when somebody found that
practice makes perfect!) Masters take on pupils. A tradition develops.
Now one important advantage of this account is that it connects closely
with the way creative artists work, something certainly lacking in the first
account I discussed and, I think, in the second as well. For it is a fact about
art that artists add to a tradition of art; they learn from, modify and add to
the achievements of their predecessors. The Hegelian idea of aufheben,
frequently translated as the neologism sublating, catches something of
this.
What an historical account wont do is explain the addition of new
genres once the concept of art is launched. How can photography, film
and even music become accepted? They come under the banner of art or
fine art once these concepts are already understood as applying to
painting, sculpture and literature. Additionally, we need to take into
account the situation where what was once dismissed as entertainment or
craft becomes accepted as art. Thus world music, primitive art and jazz
joined the lists in the twentieth century. But this cannot be explained in
terms of the accretion of individual works of art to a tradition consisting of
other works that are more or less similar. This is a matter of new genres or,
at least, sub-genres, being accorded the status of art. Here, something like
a functional account appears to be a plausible description of why these
classes were given the status of art. Jazz had many of the functions of
classical music. It was noticed that it moved and excited people and was
discussed and criticized in terms that were familiar. The move to include it
under the heading of the arts thus seemed a natural and defensible one.

A recognitional account
So far I have mentioned three accounts of what it is for something to be a
work of art. I add to these a fourth account of art, the recognitional.
(The neologism is ugly but has the important virtue of being transparent.)
This has not been much discussed in the literature, although there are
strong hints of it in a paper by Terry Diffey, albeit embedded within a form
of institutionalism.10 It has much to recommend it, even though I shall
eventually reject it as a satisfactory general account. Its distinctive feature
is that it emphasizes what would now be called the reception of art: how
it is appreciated, understood and, above all, valued by its public. An
artefact may become art because of some perceived quality seen in it by
qualified people. Work of art is a category that marks out, inter alia,
some of mankinds greatest achievements. The point about arthood is that

44
 THE WORK OF MUSIC

in order to understand the concept we need to understand this feature and

we really need no further elaboration of why we should think recogni-
tion is important. As Wittgenstein said, man is a ceremonious animal; we
need no further functional account of funeral rites over and above their
procedural description nor of the exchange of rings beyond what is ritual.
No more need we a justification of the business of recognition, even
though we can imagine, at least in the case of art, how such a justification
might run. We might say that the greatest of human achievements are
objects that repay study and attention and that we find we can return to on
many occasions, often with added rewards. Then recognition would
have a functional justification. Thus you may feel impelled by the sheer
power of a piece of music to re-describe it as a work of art. Hearing Ray
Charles singing Yesterday or Dire Straits performing Love over Gold
or Sultans of Swing may induce this reaction.
So what a recognitional account will offer is something like the
following. There are certain objects that human beings make that we
describe in terms like expressing the human spirit. This sounds preten-
tious but is, I think, defensible. We talk about what these works mean and
we describe them in expressive terms largely borrowed from our vocabu-
lary for describing people. We speak of what Hamlet is about and we can
describe a late painting by Rembrandt as expressing love or resignation. I
think Kenneth Clark was exact when he described The Jewish Bride as a
picture of grown-up love, a marvellous amalgam of richness, tenderness
and trust.11 As we shall see, it is important that music, too, can express
tenderness, resignation or opulence. It is this, it is worth repeating, and the
capacity of music to be something that can be followed, that make it a fine
art and not merely an art of ornament. There is, of course, also technical
mastery. In celebrating all this we acknowledge that the human race has
managed to produce Bach and Proust as well as Stalin and Napoleon.
However vain and difficult creative artists can be, they can have a devotion
to what they make and what they make can be of the greatest intrinsic
value, and that can give us solace in our darker moments. At least, cultivat-
ing the arts can help to give us a balanced outlook on our species.
You might ask how a recognitional account differs from others.
Obviously it contains elements of functionalism and proceduralism. I
have discussed the effect that some music may have such that it persuades
us to count it as art music. I have referred also to the role of the expert in
welcoming it in; he would then be operating like an agent of the art world.
But, standardly, procedural accounts claim to be value-free. They are an
account of what it is for something to be art, not for it to be art of distinc-
tion. The recognitional account, however, claims that what is involved in

45
PHILOSOPHY OF MUSIC

making something a work of art depends upon a judgement of value. The

value may or may not be narrowly aesthetic. Aesthetic features are not
first and foremost in the fiction of George Eliot, for example; here it is the
psychological insight that matters. Music may be ugly but fascinating. I
mentioned Mozarts Eine kleine Gigue K. 574 earlier; Charles-Valentin
Alkans Gigue op. 24 shares its character; both offer us une belle laide.
Music can also be repulsive and brilliant and certainly not to be avoided
because of the former quality; Bartks The Miraculous Mandarin is a
good example. Secondly, unlike a functional account, the recognitional
account does not demand that a work of art is or was designed to produce
a specific effect in the audience. No such intention may be present. The
recognitional account differs from the historical, too; for although an
artefact may well stand in a relation to a tradition of music-making or
painting, that relationship alone does not make it a work of art. Indeed this
account has the merit of explaining how we may decide that a work
outside a tradition counts as a work of art; it has to have outstanding qual-
ity. Recognition is one more element in a concept that, I am more and
more convinced, is very convoluted indeed.
Some writers argue that our experience of music owes little or nothing
to any technical knowledge we might have. But, again, a recognitional
account puts this into perspective. Pleasure seems a little anaemic for
the ecstatic effect that music such as Talliss Spem in Alium can have on
the listener. The knowledge, as well, that it is written in forty independent
parts and the astonishment that music so spontaneous and free sounding
could be written to such constraints makes it seem a miracle. Hardly
anybody who hears it will be unaware that it is written in forty parts,
although they might be unaware of just how difficult an accomplishment
this is without a constant din of dissonance, let alone the production of
music of such consistent melodic and harmonic interest. Music is to be
admired for all sorts of reasons, and the fact that it is the result of extra-
ordinary gifts and accomplishments ranks very high among them. Such
achievements are astounding in a mere mammal. The principal reason
why silent music is such an odd case is because nobody could admire the
skill, the technique, the years of honed accomplishment that go into this;
there isnt any. So the account I have dubbed recognitional has this
advantage over its rivals: it registers something intrinsic to the way we
value music.
What are the problems with a recognitional account? There is an
obvious one; there is bad art. It cannot be the case that what counts as art is
only what is greeted as of outstanding quality nor even (a weaker thesis
that brings it close to proceduralism) that it is thought by some agent of

46
 THE WORK OF MUSIC

the art-world that it is of outstanding merit. Even in the art-world there are
sharp operators.

Conclusion
My conclusion is not difficult to guess. Of these four accounts, as in Lewis
Carrolls caucus race, all have won, but by the same token they have all
lost. That is, I suggest, there are works of art whose place in the canon can
be accounted for by one or other of these theories, possibly in combina-
tion, but that none of them account for all cases. They may be best
regarded as describing, inter alia, the various routes whereby something
becomes art rather than as offering an analysis of art. Some of the
familiar difficulties that have attended the various accounts are straight-
forward consequences of ignoring the ragged nature of the concept of art.
This, of course, provides further reason for being sceptical about the
possibility of an analysis. So the various philosophical theories have been
regarded as competing when in fact they are dealing with different cases.
Most composers are uneven; they sometimes produce poor work. The
acceptance of inferior stuff as art music can be accounted for in procedur-
alist terms. I am inclined to think, although you may disagree with me,
that Michael Tippetts work, both early and late, was uneven. Yet there is
no question that the music he composed counts as art music. By the end
of his life, Tippett was himself a grand old man of music and his status
meant that he was an agent of the art-world and therefore in a position
to give the imprimatur to his own work. A procedural account also fits the
situation when a new genre is recognized, such as photography or film or
latterly, perhaps, pop videos. To recapitulate a little, the sort of con-
siderations that come into play might be the following. Thoughtful people
found cinema to be interesting and discussable and realized that it raised
issues of the sort that arise when we talk about paintings or about theatre.
Gradually a body of critical opinion developed and it became increasingly
clear that we were dealing with works of art. Questions of meaning and
value became important and then a canon began to form. In this way,
although critics do not make the artefacts, they certainly make the
artefacts count as art and this, I believe, is one truth behind the procedural
account. It is, however, noticeable that functional elements seem to be
playing a role. Perhaps we are prompted to call film art because we find
we react in the same way as we do to theatre or photography. We are
moved. However, such functional elements may well recede into the
background as the form of art develops. As film matures, idiosyncratic
aspects of the form, such as montage and so on, draw our attention.
Presumably something like this happened when music became recognized

47
PHILOSOPHY OF MUSIC

as a fine art on its own account rather than merely as an adjunct to dance,
drama, a text or liturgy.
But how could a proceduralist theory explain the admission of jazz? It
probably was not enlisted by classical musicians, although it interested
both Ravel and Stravinsky, among many other composers. Rather, it
excited and interested people in much the same way as some classical
music had done and, with the development of a tradition of influences and
critical commentary, the basic requirements for the status of art music
were in place. A mixed functional and historical account seems to explain
the emergence of jazz as art music. But, as should be obvious, these
historical explanations are fairly speculative. It is difficult to disentangle
the various elements and the actual causes of or reasons for a genre
becoming art probably involve a melange of different factors.
In the case of silent music we find we are pulled in two directions. On
the one hand, because Cage had composed proper music we are inclined
to follow the usual principle and say that anything a figure of some stature
composed must be a work of art, although we may have to add that this
particular work is silly, pompous, boring and a waste of time. But then
even the prodigiously talented nod. On the other hand, we are inclined to
say that this is so different from anything that counts as music that it
simply is not music. It has to be at least organized sound to stand a chance
of counting as a musical work of art.
What emerges from all this is that there is no single route by which
something becomes art. There are many different ways: procedural, func-
tional, historical and recognitional, and, perhaps, others I have not identi-
fied. They may function together and it may be very hard to say which is
the more important factor in any given case. Some art, too, is art from its
very inception. Does it follow from this that no general justification of
why an artefact is a work of art is possible? For centuries, philosophers
were exercised over the problem of induction. What justifies us in
concluding that because all the sheep we have observed are herbivorous,
all sheep are herbivorous? The conclusion could be false while the
premises are true. After all, generalizations like All swans are white,
which once looked pretty well-founded, turned out to be false. The widely
accepted view nowadays is that there is no general justification for
induction. What there are are particular justifications in particular cases. I
can show why this is a sensible induction given context or subject-matter.
In our case, we are considering the application of arthood to something or
other and, again, I have yet to be persuaded that a general justification can
be given. We might argue that this, that or another item ought to be
counted as art. And the reasons will be various.

48
 THE WORK OF MUSIC

Understanding and analysis

I began by saying that what philosophers of art have tried to do, on the
whole, is provide an analysis; they have attempted to find necessary and
sufficient conditions for something being a work of art. Such a program-
me, if successful, will enable us to rule the hard cases either in or out. I
want now to shift the attention from the idea of analysis to the notion of
understanding. For if we think not so much about what it is to be a work
of art but what is involved in understanding what a work of art is or what
it is to have a more or less adequate notion of the concept of art, the
question of hard cases can be looked at anew. For to understand the
concept of art is to see that cases like silent music are just those cases on
the edge of the concept over which no final motivated decision can be
made. Our concept of art does not stretch that far. We have grounds for
thinking it art and also grounds for refusing that title. In no-mans-land the
boundaries are not marked. To mark them entails, as I have said, a deci-
sion. Decisions can be well-grounded but even then they are decisions.
And it may well be the case that no grounds decisively favour one
alternative. The other advantage about thinking in terms of understanding
rather than analysis is that understanding can be a greater or lesser, a fuller
or more exiguous business. I may have a better grasp of the concept of art
than a teenager and my better grasp shows itself, in part, in my realization
that the concept has a vague penumbra. As I use the word concept it is a
product of language use. Our concept of atom is acquired as we pick up
the context in which the word is used. You dont have the concept unless
you have a decent grasp of how the word is used but grasp is not an all-or-
nothing affair. I have a better grasp of the concept of an atom than a child
but a physicist has a better grasp than me. Likewise, people who are better
informed about the arts are likely to have a better grasp of the concept of a
work of art than others and, pari passu, of the concept of fine art. What
remains, even among such experts, is the difficulty of the hard cases.
Some philosophers think that the job of philosophy is to investigate the
world at its most general. Metaphysicians, they say, should inter alia be
studying what art is and that might not be the same thing as the objects
that we think of as constituting the range of fine arts. But then the problem
is how to delineate the object of study. Unless they mean by art what we
mean by art, then, to a greater or lesser extent, we dont know what they
are talking about. Since art is, in some respects, an unclear concept, we
need to become clear about what is central and what is peripheral, what is
incontestable and what is controversial about its range, before we say any
more. We need first to decide what art means, and here lie all the

49
PHILOSOPHY OF MUSIC

uncertainties. Metaphysicians can be insouciant about this. They will tell

you that what they study is not subject to the little incoherencies and
frayed edges of what they call folk concepts. But then, of course, we ask
for the point of this undertaking. Why tempt the raising of questions
whose answers are but mere corollaries of the way they define their con-
cepts? One thing is obvious. Unless you deal with demotic concepts you
do not deal directly with the concepts which reflect and characterize our
culture and our values. The results of this are hugely important for the
consequence is that philosophy as cultural criticism is then emasculated.
We should be interested in the ideas that are endemic in our culture and
whose use reveals what is presumed to be significant; and, if there are
inconsistencies within such concepts, what this tells us about the most
general features of our culture matters. After all, the problems with the
concept of art arise specifically from the hard cases mentioned earlier. Is
silent music really music? These cases puzzle the woman in the street and
the man on the Clapham omnibus and they should be able to look to the
philosopher for help. They wont get it if he insists that how he uses art
is not subject to the imprecisions that leave us with problem cases.
Look elsewhere for a moment. Concepts of God are notoriously
problematic. Thinkers have characterized God in a way that cannot easily
be made consistent. He is simultaneously omnipotent, omniscient and
benevolent. Yet there is evil in the world. Of course, there are countless
attempts at making the notion of a deity compatible with the existence of
evil. Even if you think, as I do not, that these might be successful, you
must allow that it is a possibility that concepts in this arena might not be
self-consistent. Even if you deny this, believing that, with the grace of
God, a consistent theodicy can be found, you may think that classical
Marxism is not so favoured. It cannot be a coherent intellectual enterprise
because of, for example, the difficulties in reconciling human freedom
and economic determinism. My point is that we allow that broad systems
of ideas, ideas that are central in a culture or sub-culture, may not be
internally consistent and if we allow this why should we not allow the
possibility that the same holds with respect to our ideas about the art of
music?
My rejection of metaphysics as a discipline prior to conceptual enquiry
is motivated by a feeling that metaphysics has fallen between two stools.
On the one hand, it aims at consistency. On the other hand, it also attempts
to describe the concepts we have. It cannot be utterly detached from the
way we use terms like fine art. But failure to distinguish these leads to
an assumption that ordinary concepts will be consistent and well-defined
in the tidy way a philosopher requires.

50
 THE WORK OF MUSIC

Those readers familiar with modern philosophy will identify my answer

to the question that opened this chapter, What is art?, as Wittgensteinian,
and I will not complain if it is taken that way. But to the Wittgensteinian
account there is a traditional objection. Unless you specify the similarity
that links different forms of art together, you beg the question. It is no good
talking about a loose pattern of disparate arts unless you are prepared to be
explicit about the resemblances that lead us to call them all the arts. Now,
I specify some routes whereby things become art but I have not said what
general similarities unite the strands. My argument is open to the question:
why, on this view, should these diverse phenomena be called art if they have
nothing at all in common with each other?
We have said something in the preceding pages. Works of art do have
some features in common; indeed the concept is paradigmatically what
Wittgenstein thought of as a family concept; just as there may be no one
thing that all games have in common, neither winners nor losers, nor a
competitive element, so there may be no one thing that all works of art
share. Rather, there are various strands woven into the concept. I have
identified four; there may be more. But it is unlikely that we can find any
necessary conditions and it seems unlikely that we can find sufficient
conditions either. What I would say is that, in addition to what I remarked
earlier, to declare that something is a work of art is to suggest that it has
two important aspects: expressive content and interpretability. The first
implies that it has, in a sense, meaning; works of art are the sorts of
things that can be ebullient or reserved, as people can. Secondly, they can
also comment on our culture and on the times we live in as well as on the
more eternal themes of human nature; teasing out these implications
is what we do when we interpret art. What is very much of concern to us in
this book is that this expressive content (in the sense of having meaning)
is a feature of music that seems permanently controversial, as might have
been concluded from the brief discussion of Hanslick in Chapter 1. In
Chapter 3 we shall see how vigorously claims about meaning have been
debated. One important preliminary finding is this. You can make a strong
case for saying that some music counts as art because it has an expressive
face, because it can be followed and because it can be variously inter-
preted. But a fuller discussion of what is a complex issue will follow; in as
much as there is a gap in my discussion of what it is for something to be a
specifically musical work of art, it will be filled retrospectively.
There is, then, a feature that is central to the fine arts such as fiction,
poetry, drawing, painting, sculpture and so on, and that is found in music
but is either absent or etiolated in the decorative arts such as embroidery
or carpet-making. We think of a work of fine art as having expressive

51
PHILOSOPHY OF MUSIC

features. We sometimes ask what a work means, although the locution is

perhaps not le mot juste. We may describe a work as sunny or tragic,
for example. The opening movement, the Allemande, of Bachs French
Suite No. 5, seems to me to have a frank and open character that needs to
be reflected in the way it is played. It is not reticent, nor does it hint at
concealed issues. That works of art tend to have expressive characters in
this way is, I believe, central to what distinguishes works of art from other
things that human beings make and do.
But there are difficulties even for this cautious suggestion. These
conditions are neither necessary nor sufficient. There are works of art that
present us with a blank face or where we know so little about the back-
ground that we cannot read them. Interpretation is not a possibility.
However, the decision to count something as a work of art means that we
search for these features. There may be art for which the search is futile;
there is no possibility of interpretation. Much contemporary music seems
to lack an expressive character. The music of Iannis Xenakis or Boulez
seems to me very hard to describe in expressive terms. (Hanslick thought
that some music of the eighteenth century lacked expressive nuances.)
Furthermore, we certainly cannot follow the music of Anton Webern in
the way we can that of Brahms or Mahler.
When I wrote Contemporary Aesthetics12 over 20 years ago, I made two
proposals. The first is that art is something of a ragbag of various sorts of
artefact and the second is that the concept has some distinctive social and
political functions. The first point suggests that there is something of
contingency about the way that those artefacts that we regard as art count as
art and, indeed, I think this is so. It does not seem to me implausible to
suppose that embroidery and tapestry, crochet, carved walking sticks, love-
spoons, decoupage or scrimshaw may be moved from the category of craft
to the category of art. Were this to happen, perhaps we would begin to
concern ourselves with meaning, expression and interpretation and the
creation of a canon and these would be a matter of critical debate. We would
also look for originality. All of this can be a matter of degree, of course. It
was brought home to me recently how a skilled carver of love-spoons varies
traditional patterns while keeping some of the symbolism.13 Perhaps, as
well, arguments might break out as to whether the unknown Master of some
ship had intended a particular piece of scrimshaw, evidently a late work, to
have the spiritual transcendence that some commentators have seen in it
or whether in fact it expresses a cosmological despair, a sort of oceanic
ennui. But although these considerations are signs of arthood, they are
not necessary. Primitive art may be created in a context where these
considerations are absent and only later may it be classed as art. There is no

52
 THE WORK OF MUSIC

canon for Benin bronzes and nothing much in the way of interpretative
debate, beyond showing how they express certain prevalent ideas in their
societies. Arthood is indeed a matter of a melange of cultural, political and
historical factors. Among other things, it also marks out the important
artefacts whose appreciation requires hard work and preparation and whose
understanding demonstrates sophistication, education and cultivation.
Once given the status of art, expressive and interpretative features may
come to the fore. The function of calling this piece of music a work of
art is to celebrate its distinction and to cue certain approaches and
responses from us. Calling something art may then be as much a cue as
a description.
Of course, there are works of art that are not worth our consideration.
They have an expressive face, perhaps, but they may be sentimental or
mawkish. Does their poor quality mean they cannot be interpreted? I think
it is more the case that they are not worth interpreting, for to interpret them
means spending time in their company and that is hardly worth the effort.
Even those who find Wordsworths poetry moving or thought-provoking
will agree that much of his verse is dreadful and does not merit further
consideration. Our conception of art is evaluatively centred in the ideal
case. But most works of art lie outside this charmed circle. Such bad works
of art count as works of art, sometimes by being the noddings of the
otherwise great.
Now consider the second point. Understanding the arts is a sort of
badge. Some argue that it is a mark that the middle classes place upon
what belongs to them and that its function is to mark them rather than
what they appreciate, a view which was widely advocated in the 1970s.
Art simply is an indication of social status, a mark of sophistication. Now
there is undoubtedly something in this, regrettable though it is. It was
predictable that when the tabloid newspaper the Sun sent its music critic
to Glyndebourne, he reported that it was an expensive noise. For Sun
readers to like opera would be nothing less than class betrayal (or
possibly betrayal of proprietor Mr Murdochs tastes or betrayal of his
judgement as to what his readers ought to like). The other side of the coin
is what James Jolly, the editor of Gramophone drew attention to. He
exaggerated when he said that the productions at Glyndebourne,
generally as fine as any in the world, were presented to an audience half
of whom would not be able to tell the difference between what is on offer
at Glyndebourne and a run through at a village hall. But he did not
exaggerate much. Many in the audience are there to confirm their
proprietary rights rather than to enjoy the music. Glyndebourne, like
Ascot, is one of the places to be seen.

53
PHILOSOPHY OF MUSIC

There is an important lesson to be drawn from the first part of this

chapter. To know what a work of art is is to have the concept of a work of
art and this is to understand the hard cases for what they are: hard cases.
To the extent that you are in the grip of an analysis, you no longer under-
stand the concept of art. Thus the institutional theory, by determining
what we should say about silent music (it is art music or not because
the qualified say so), does what a theory should not do: it imposes an
answer. Anybody equipped with the institutional theory can then rule a
hard case in or out and any uncertainty he has should disappear. But
surely it will not. He cannot get over the discomfort this case produces
and, to the extent that he can, he has ceased to understand the demotic
concept for what it is. (It is equally true that moral debates and dilemmas
are of the essence of the moral life and that a moral theory, such as
utilitarianism, that supplied a way of settling them, by that very token
fails as a theory.)
To christen a piece of music or recognize it as a work of art is, I have
said, as much a cue as an identification or a recognition. It sets in train a
series of considerations, the music will be seen as offering opportunities
and the act of declaring a piece a work of art has implications, none of
which are easily captured in the jargon of analytic philosophy. It has
something in common with registering with a dating agency or a marriage
bureau. The object becomes available for differing sorts of intimate
relationships. I intend this analogy to be taken seriously. We live our lives
with art. The books, the films and the music that I love are what I continu-
ally return to. I want them in the house on hand. We speak unashamedly
of love here, even though, it will be argued, the relationship is single-
sided, a claim that will occupy my attention later. The works do not adapt
themselves to me as my friends do, it will be said, although I, of course,
may adapt myself to them, being prepared to take time, trouble and
thought to see what can be seen in them. On this view, perhaps the love of
art is more like a love of God than love for friends and family. For on the
usual conception of God, my love for Him changes me but not Him, for
He is immutable. Similarly, works of art change us but, it is usually
thought, we do not change them.

The work of music

Over the years I have heard about half a dozen recordings of Count Basie
and his Orchestra playing Lester Leaps In; they all vary in the actual
notes played, some very considerably. Two versions are very similar in the

54
 THE WORK OF MUSIC

notes played but one is so evidently the superior that the other sounds like
a draft version of it. The former is one of those pieces of music in which
every note sounds in place; alter and we would damage it. The ending in
particular is so exquisitely managed that it is impossible to suppress a
laugh of delight at the elegance with which Basie and Lester Young
manage the final bars. Are all of these performances of the same work?
Or are they performances of different versions of the same work? Asked
what Count Basie and his Orchestra are playing in every case we would
answer Lester Leaps In and we do not think that the same title is being
used to cover half a dozen different works. What do we say about Charlie
Parkers She Rote? The opening is arranged but the rest is pure
improvisation on a tune not mentioned in the title Beyond the Blue
Horizon but it isnt the melody of this well-known tune that is used, but
only its harmonic base. Is Parker playing Beyond the Blue Horizon?
Such complicated cases apart, in jazz, generally, it seems as though
identity is given by the theme around which musicians improvise or
compose. Contrast that with classical music. The notes are very largely
predetermined. The failure to play all the right notes in Mahlers Fourth
Symphony means that a mistake has been made; but even among accurate
versions, those which have been edited by recording engineers, there are
differences, but they are differences not in the notes but in the nuances of
performance; they differ in tempo, dynamics and in the degree of rubato
(the varying of pace within a basic tempo).
In between these two examples, Lester Leaps In and Mahlers Fourth
Symphony, there are cases where a certain latitude is accorded to the
performer in playing ornaments. Earlier music is less fully scored. The
eighteenth-century instrumentalist playing a concerto would have been
expected to provide a cadenza, perhaps improvised, or a vocalist would
have been expected to add ornaments to an aria as she saw fit. Yet the
former remains a performance of a Mozart concerto even if at points the
notes vary from other performances of the same work. What constitutes a
work seems to vary in as much as what the performer adds varies in
extent. There are conventions that provide a context within which clas-
sical music is performed. A decent edition of Bach or Couperin shows
you how the ornaments should be played and gives you an idea of lower
and upper limits as to tempi. Over the centuries the notation of Western
classical music has become increasingly specific. This has several
explanations. First, the music is no longer played within a sort of guild
whose members know the ropes, in the way an eighteenth-century soloist
would have known and have been expected to know the conventions of
performance. Nowadays music will be played by musicians who will not

55
PHILOSOPHY OF MUSIC

be in personal contact with the composer, at least after the premiere and
the first cluster of performances. Secondly, music is now assumed to
outlive its composer and the composer will almost invariably want control
over the manner of its performance. Increasingly precise notation enables
him to obtain this.14
At the other end of the spectrum there are electronic compositions in
which every element is decided by the composer and not varied. Each time
the music is played it is exactly the same. There are other cases that closely
resemble this. A piece of rock music might be created in the recording studio.
Vinyl records or CD copies are made from the master recording. A live
performance of the music may amount to no more than lip-synching. The
artists are not really performing at all and the music is the same at each
performance. Incidentally, a pop record producer may stand in relation to
the finished product quite unlike that of the sound engineer on a classical
recording. The producer may play a creative role in the recordings he
supervises through the various techniques of over-dubbing, sampling, adding
tracks and all the other means of editing at his disposal. It is tempting to think
of the producer as somehow analogous to a film director but the role played
by Phil Spector or George Martin is less that of an auteur and more that of
a primus inter pares, it seems. How, then, should we think of such a
recording? The rock or pop recording is a sort of prime instance rather like
the autograph of a novel or the original of a lithograph. Copies are taken from
it by a purely causal process. Indeed, if classical recording came to supplant
live concert hall performances then it is conceivable that the recording
might take on something of the status of the pop or rock recording, as a work
of art in its own right. But at present, the Beatless recording of A Day in the
Life is very different in its status vis--vis a work of art from a classical
pianists recording of a performance of a Beethoven sonata.15
Although the view that, in rock music, the work is the recording is
now very widely accepted, in many cases the piece will go on developing
through live performances (where the band does not lip-synch to an
existing recording). Here what philosophers call the ontology is differ-
ent again. There is a steadily developing conception of a piece of which a
recording only offers a snapshot of one particular phase in that develop-
ment; the situation compares closely with that of the various versions of
Lester Leaps In.
Some jazz musicians may settle into a standard way of doing Laura
and they play it that way on successive gigs. The notes will be the same
each time, more or less; departures from the norm will not count as an
interpretation of the piece. Performance is closer to playing a recording
over and over again. Parenthetically, a great deal of recorded jazz sounds,

56
 THE WORK OF MUSIC

to the classically trained ear, like hugely competent note-spinning and its
interest is limited. Music that seems to have a more lasting appeal, such as
Benny Goodmans AC-DC Current, which I shall discuss at greater
length in Chapter 4, has a structural and motivic organization that brings it
closer to Western classical music. But this is fairly exceptional in jazz. It is
fun to go to a live jazz concert, where the pleasures of watching and
listening to a performer, noticing how mistakes are embroidered into the
text of the music and used, is rather like watching an experienced lecturer
deal with a slip of the tongue by making a joke or weaving it into what he
intended to follow. I have heard it suggested that Rachmaninovs Third
Piano Concerto ought to be listened to in the frame of mind one listens to
Miles Davis and not Brahms. Deep structural unity is lacking in this work,
although the constant varying of the melodic motifs, which are drawn
from Russian chant, does give it the sort of moment-by-moment logic that
Gurney thought was what we mainly require from music. This piece
simply celebrates the delights of performance and display. There are other
cases again. Some music in the Western classical tradition is nowadays
created by the composer improvising at the piano; a microphone in the
piano is used to make a recording on a mini-disc, which is then edited on a
mixing desk. This might then be notated later on. Which, here, is the work
of music? At the other end of the spectrum, there are unrepeated improvis-
ations by musicians trained in jazz, in Indian music or Western European
classical music. I have described some of the forms that a work of music
can take. I have hardly exhausted their variety. Peter Gabriel has produced
a CD-Rom, Xplora I (1993), which allows his songs to be remixed by the
player.
What, then, is the work of music? It looks fairly evident that there can
be no simple answer to this question. The work of music may be a record-
ing engineered and created in a studio. On the other hand, what began as
an improvisation becomes more and more rigid in performance until we
have merely the illusion of improvisation but not the substance. It may not
be written down but, to all intents and purposes, it is as if it was played
from a notation that has been memorized, although, while it remains un-
notated, the role of interpretation may be more exiguous. Between these
two extremes we have the cases in Western classical music on which
philosophers have tended to concentrate: a notated music that has to be
interpreted by the performer. Thus the criterion of the identity of a piece
of music varies according to the sort of music we are dealing with. A
discussion of this issue needs to be historically and socially informed. It is
no good imagining that the criterion in use for Western classical music for
the past two centuries applies universally. As we shall see in a moment,

57
PHILOSOPHY OF MUSIC

that conception is itself a product of various philosophical and ideological

forces and, partly because of this, the debate about what a work of music
is has all sorts of implications for its standing as an art and for the relev-
ance of certain ways of listening and performing. These are issues that
have considerable importance outside the ivory towers of academe.
Classical music reflects a dichotomy to be found in philosophy at large
in a way rarely found in world music, rock or jazz. The sharp distinction
between work and performance raises questions about the status of the
work itself, questions akin to those that philosophers ask about ideas,
properties, mathematical concepts and mathematical and logical proofs.
Plato famously thought of goodness, the Idea or Form of goodness, as
something distinct from the goodness of a single act of kindness or
generosity. This universal Good had an existence that was separate. If
you believe that moral values have an existence over and above the
morality of individual acts, an existence that can be apprehended in an act
of discovery, then you take, in some form or other, a Platonic view. Such
values exist independently of what we judge or think and the morality of
our actions is to be calculated with reference to them. Likewise many
believe that the work of music exists separately from and independently of
the individual performance or the score from which the performer works.
Is there any reason to suppose that the status of classical music has
been enhanced by its replication of time-honoured philosophical posi-
tions? Well, as some have persuasively argued, the notion of a work of
music as an object comparable in permanence to a sculpture, a painting or
a performance puts it on a par with already established works of art.
Performance, by contrast, is evanescent, as evanescent as the flowing
stroke play of the Indian batsman Sachin Tendulkar. Although it can be
captured on film, seeing it is like hearing a record of a jam session; it
preserves an action. There is no work of football performed in an
individual game and even though a game capable of aesthetic sublimity,
like cricket, is often watched for that quality, it is not and never can be like
a work of art. Part of the Platonic inheritance is a belief in the value of
what is permanent. How often are works of art described as inexhaustible
and their creators as immortals, not of an age, but for all time as Ben
Jonson wrote of Shakespeare? The distinction between work and perform-
ance enables the work of music to claim the same status as works of
literature. It has been argued recently that the development of the concept
of a work of music developed as Western classical music flowered in the
late eighteenth century with the first Viennese school. It is, I am sure, sig-
nificant that the metaphor of the concert hall as an imaginary museum
developed in the nineteenth century.16

58
 THE WORK OF MUSIC

In some form or other, realism, at least about classical music, is not to

be gainsaid. The alternative is to reduce, as philosophers would say,
the work to the performance. Reduction programmes in philosophy are
traditional. You reduce a physical object if you show that it is no more
than a possibility of sensation and not something with an existence
independent of perception. More perniciously, a reduction of the mind to
the brain shows that thoughts and emotions are really only electro-
chemical patterns in the brain. Reduction programmes are no more
successful in the case of music than they are elsewhere in philosophy. You
cannot reduce a work either to the class of its performances or to the
possibility of a performance. The work is not the same as a performance.
Lets begin with a few truisms. A work can exist unperformed, as long as
the music has been written out and preserved in a library or a study. It
might, as well, be remembered accurately by somebody even if no notated
copy exists. While it is remembered, it exists. If the last copy is destroyed
then it continues to exist as long as the last person to remember it can do
so. After that the work is lost. (Whether it is right to say that it no longer
exists is something we shall consider shortly.) So if the work could be
performed but is not, there is no set of performances; but nobody would
conclude that the work does not then exist. Indeed, if the work is the set of
its performances, the more performances there are, the larger the number
in the set. But a work does not grow bigger the more often it is performed.
As far as the idea that the work is the set of performances is concerned, in
order to allow for the possibility that a work exists without ever being
performed we would have to treat the set of performances as a class that
may be a null class (i.e. a class with no members, but a class none the
less). But even if this escapes the present difficulty, a further trap remains.
The notion of a work embodies certain normative requirements. It brings
with it the possibility of errors in performance.
A second consideration, then, is this. Certainly, two performances
might be of the same notes but differ in respect of tempo and dynamics, all
those subtle little variations that make it interesting to hear different
performances of the same work. Both performances, we can agree, are
accurate, but since the notation is not specific with respect to the small
nuances that are the difference between a notable performance and a
boring one, it is possible for the two performances to differ. There may, as
well, be mistakes. Probably no live performance is perfect. But how can
we have the concept of an interpretation or of a mistake without having
some prior notion of what the work is such that this performance diverges
from it or interprets it? The problem, then, is that both these definitions of
a work (as either performance or as a set of performances) leave us with

59
PHILOSOPHY OF MUSIC

no way of making the important judgements about accuracy and interpre-

tation we need in our everyday musical life. To be able to make these we
need the concept of a work distinct from its performance.
For these reasons we cannot rest content with a description of the work
as the possibility of a performance or as a set of performances. If you say
either then you need to say more for there is a normative element in the
notion of a work of music. It prescribes how a performance should go and
departures from that prescription are what define mistakes. If it was just
the possibility of a performance that constituted the work there would be
no room for the idea of an error. We could be rough and ready about what
counted as performing the work. You might argue that our ability to
identify it as the work in question would suffice for the preliminary task of
picking out the work. But we want more from the concept of a work; we
need the element of normativity.
We also describe the work and the performance in different ways. A
work can be good but a performance of it bad. Perhaps, too, it is possible to
give a good performance of a rather mediocre work, making the best of it.
Sir Thomas Beecham was sometimes said to make second-rate music, such
as that of Andr Grtry, sound better than it is. Now I dont know whether
this is a fault in Beecham. If it is, it is venal. The worst one can say is that
he misled us about the works stature. Again, a performance can be faithful
to the character of the work. There is no possible parallel construction
relating the fidelity of the work to the performance. There is no symmetry.
There are further complications for the thesis that a work is a possible
performance; it will not cover most improvisations that exist, it seems, only
as long as the performance. There is no possibility of a second perform-
ance. Here the notion of a possible performance seems to collapse into a
singularity. If the improvisation is neither written down nor remembered
note for note by the player or hearer nor recorded, it ceases to exist after the
performance. (The improvisations of Beethoven and Bach are in that
category.)
If the improvisation is recorded, of course, the situation becomes
different, although not in a way that saves the thesis. Then, each copy of the
record will be an instance of the piece in a way we shall shortly define, and
it will be possible to play the work by playing the recording. The situation
is, as we have seen, close to that of the multiple printings of a novel or a
poem. Barring accidents, each copy is identical with each other in the
edition and nothing is gained or lost by listening to one copy rather than
another. It is unlike the performance of a sonata, which may be more or less
revealing, more or less successful and more or less artistically praise-
worthy.

60
 THE WORK OF MUSIC

So the idea of a work as a possible performance or as a set of perform-

ances does not apply to those works that exist as tapes, or as CDs. These
can be played but not performed. More significantly, perhaps, it will not
cover cases where a work cannot be performed because we dont know
how to do it. Perhaps we dont know what instruments it requires or
perhaps the instruments no longer exist or perhaps nobody can play them
any longer or perhaps we do not know what the performing conventions of
the time were. In the limiting case, a work might be physically impossible
to play. But in such cases we would not deny that the work exists.
Could we, then, identify the music with the notated score, the copy that
enables the pianist to produce a performance of the sonata? Again, this
cannot be right. A score could be inaccurate. Scores sometimes contain
contradictory instructions for tempo and dynamics; we then have to
choose which characterizes the real work. Sometimes they are crucially
vague. In the last movement of Beethovens Piano Sonata op. 110, at bars
1667, Beethoven writes nach und nach weider geschwinder (poco a
poco pi moto, or little by little faster again). Now, is the tempo supposed
merely to accelerate to the opening tempo of the movement by bar 172?
Tovey says it would be unthinkable to accelerate beyond that.17 Or
would it be right to go on steadily accelerating to the fortissimo close of
the movement at bar 207? Myra Hess did, although one runs the risk of
creating a merely superficial thrill in this majestic movement. But
although a score may be vague, we do not, I think, say of a work of art that
it is vague, at least not in any sense relevant here. Furthermore, as in the
case of identifying work with performance, if we identify work and score
we immediately run into the difficulty that the score might be lost and yet
a pianist who has memorized the music may still be able to play it; and if it
can be played it must exist.
At this point we need to introduce a little technical terminology from
philosophy. The great nineteenth-century American philosopher C. S.
Peirce distinguishes between type and token. The type can be roughly
understood as an abstract object with which a particular, the token, com-
plies (for want of a better word). This paragraph began with the word
At; this is a token of a type; the type of the word at is located nowhere
in particular but one token appears here. Among its many tokens are
occurrences elsewhere in this book and, of course, in the newspaper I
read, in books in the library and so on. Types and their tokens can be
complex. The Welsh dragon is a more complicated type with which
individual token flags comply. Prousts la recherche du temps perdu is
an exceedingly complicated type containing millions of words with which
individual printed copies comply. Of course they may contain misprints;

61
PHILOSOPHY OF MUSIC

there is no precise answer to the question: how many misprints may it

contain before it ceases to be a token of that type?
Let us then regard the work of music as a type with which an individual
performance, or a manuscript copy complies. The type is Beethovens
Piano Sonata op. 101. The performance I give, with lots of mistakes, is a
token of that type because the mistakes are not sufficient to disqualify it as
a performance. (The peerless Beethoven pianist, Solomon, gives a per-
formance in which there are, as far as I can judge, very few errors). Now, I
have mentioned mistakes. In the notion of a type there is a certain
normativity involved. The type prescribes how a token should go and a
departure from what is prescribed means a mistake has been made.
A printed copy of the score is also a token of a type. Should we think of
the printed copy as a token of the work? Certainly the typetoken contrast
seems to be applied in two ways, something that can occur when we apply
the contrast elsewhere. My utterance of At is a token of a type just as the
printed At is a token of a type. Does it matter too much whether we
decide that the printed or written score and the performance are different
tokens of two different types or different tokens of the same type? We
need to pay attention to the difference between a score and a work.
Notation gradually developed over centuries from an aide-mmoire for
performers to a notated recipe for performance that was increasingly
prescriptive, although without ever removing the need for interpretation
by the performer. Perhaps the closest parallel today to earlier notated
music is the printed version of a pop song; this is far more of a memory
aid than a set of instructions to produce a performance. In a jazz standard
with a subtle harmonic structure, like Laura, it is the harmonic sequence
that the notation helps you to keep in mind. The tune can be altered as the
spirit moves you. This is leagues away from the latest ur-text18 edition
of Bachs Partitas, from which I depart at the peril of my teachers
indignation. One would, I think, need to be very relaxed about how a
token complies with the type to allow that various renderings of the
standard Sweet Georgia Brown on different occasions by different
bands are all tokens of the same type. However, from what I said earlier, it
is clear that each LP or CD including the Beatless A Day in the Life
includes a token of a type, of which the master recording is not the piece
of music itself but merely the first token. In some ways, the master
recording is comparable to the autograph copy of Prousts novel. Neither
is privileged as far as the reader or listener is concerned. If either were
lost, the work would not be lost. It is very different from a painting. A
copy of this is a mere copy and if the original is lost we generally regard
the work as lost.

62
 THE WORK OF MUSIC

My own predilection is for calling music copy not a token of the work
but only a token of that type which is the score. The music copy is, after
all, obtained from another copy or from the composers autograph by a
causal process and there should not be any mistakes in transmission. Of
course, an editor may make alterations but this is in an attempt to repair
what looks like a mistake in the process of transmission or a slip on the
part of the composer. The relation between work and performance differs.
Many musicians work out how a piece should go; once such an interpreta-
tion has been formulated, they play that interpretation each time they
perform it (give or take a little); but it would be a mistake to think of this
as a causal process of the kind that holds between a notated piece and a
photocopy of it or between a score that has been proofread and corrected
by a composer and the printing of that piece. The intentions of the
performer are far too closely bound up with what goes on. The interpreta-
tion, too, should be pervaded by the musicians knowledge of the style and
of the context in which the work was produced and even the acoustics of
the building in which he happens to be performing. All this feeds in to the
interpretation. It is as far from a mechanical reproduction as could be
imagined.
I have laboured the point that notation has never become so precise as
to rule out the need for decisions by the pianist or conductor. Indeed, if
you listen to performers of even the most precisely notated twentieth-
century music you may be surprised by the way they vary, even while the
interpreters are assiduous in respecting what is written.

Realism: for and against

As I have suggested, I think we are bound to accept some form of realism

respecting classical music. The work of music exists in between perform-
ances. It exists at least as long as there remains the possibility of its being
performed (without it being identical with that possibility). But is its
existence even less circumscribed? Might it not exist eternally? This
question revolves around how we view the notion of a type. Should we
think of the type as created or initiated at some time by some individu-
al? Did Beethoven create or initiate the type of his Archduke Trio or was
it there from eternity just waiting for Beethoven to identify it? What would
lead somebody to adopt the extreme realist view that musical works exist
eternally? I think that the motivation is to do with the notion of music being
essentially some sort of pattern of notes, or structure of sounds, and there
are, obviously, many such possible patterns and structures much as there

63
PHILOSOPHY OF MUSIC

are many possible geometrical shapes, ranging from the triangle up to

shapes of a thousand sides, some with curved sides, some with straight and
so on. Platonists think that these exist prior to their being discovered by
geometricians; if you agree, then you will be tempted to think the same
way about musical shapes and structures.
A musical shape is thus something like the philosophical notion of an
abstract particular. It is not a universal like greenness, because that
correlates with a single property whereas a musical shape may be extremely
complex, running to many bars and many, many notes. Neither could it be
a kind in the way tigers form a kind. For, again, the musical work is
normative; it prescribes how an instantiation or a token should go. The
concept of a tiger is not currently normative in prescribing how people
should create an instance of it. It is not so much that the notion of a kind is
never normative. After all, in the case of an animal there may be a
biological norm as to how many teeth it should have and a tiger who has
lost teeth does not come up to the norm.19 But the notion of a biological
kind is not prescriptive. Now the idea of a work of music as a set of
instructions for producing a performance brings such prescriptive features
to the fore. But this, again, cannot be the whole truth. A work is not just a
set of instructions; it has a character that needs to be reflected in perform-
ance. It is hard to think of a set of instructions as being epic or tragic, for
example. Finally, one consideration militating against extreme or Platonic
realism is that the notion of discovering something normative is itself
rather puzzling. Did Samuel Johnson or William Wilberforce discover
slavery to be wrong? What sort of facts are discovered in this way? (The
problem will arise again in Chapter 4.)
Furthermore, a kind is a class of entities. But we have seen that a
work is not a class; it is a particular thing. For roughly similar reasons
we do not usually speak of the composer as inventing his music; fairly
central in the notion of invention is the idea of a pattern that can be
repeated. The diesel engine was invented and its construction repeated
over and over again. But performances of a piece of music are not
separate works of music. They are just that: performances of a work.
Perhaps somebody invented the fugue, thereby inventing a form in which
many different works of music can be composed, but that is about the
extent to which we can make sense of the notion of invention in music.
Significantly, Arnold Schoenberg was rather foolishly proprietorial about
the twelve-tone system. He thought he had invented it and, about this, he
was either right or wrong; it may have been simultaneously invented by
another musician. But he did not need to be proprietorial about his
String Trio.

64
 THE WORK OF MUSIC

The implied analogy with mathematics that lies behind the idea of
musical works as eternal Platonic patterns of notes or sound structures has
deep historical roots and it is highly significant. Clearly Platonists about
music are inclined to think that musical works are not so much created as
discovered. So the suggestion we are considering is that just as variste
Galois made a mathematical discovery so Beethoven made a musical
discovery;20 he discovered the series of notes that make up the Archduke
Trio. The possibility of combining these notes certainly existed before in
the way that possibilities are generally thought to exist. It only fell to
Beethoven to make the discovery. Now advocates of this view do not have
to think the discovery could be made at any time. They could take the view
that the culture of music had to be in a certain state, that of late classicism
verging on romanticism, for somebody to discover that work. It is also the
case that not just anybody could have discovered that work. It needed a
musical genius like Beethoven. The situation is thought analogous to that
in science. It took a man of Darwins genius to discover the principles of
natural selection and such a discovery could not have been made much
earlier than it was. It needed the work of Charles Lyell and others to set
the scene.
Platonists think of the eternal existence of musical works as the
existence of possibilities that are then made actual by the discoverer. Now
you may feel uncomfortable about the idea that possibilities exist and, I
think, you would be right to be so. Certainly one objection is that this
theory plays fast and loose with our notion of possibility. Possibilities
may be said to exist in some etiolated sense but it is a robust part of our
concept of a possibility that it does not exist in the usual sense until it is
brought into existence. Treat possibilities as though they exist in any fuller
sense than as mere possibilities and you collapse a useful distinction.
Indeed to say that the Archduke Trio was a possibility before 1811 is
precisely to imply that it did not exist before Beethoven composed it.
Furthermore, there is a perfectly respectable sense of possible in
which we might say that this work was not possible before Beethoven
came along. First of all it has the fingerprints of Beethoven and not his
precursors. It is hard to see either how Bach could have conceived the
opening page or what he would have made of it had he produced those
notes while idling at the keyboard. Equally, had the famous Tristan chord
occurred to Mozart, it would have been thought of as a progression that
required resolution. He could not have thought of writing it as Wagner
did. Indeed, the greater the composer and the greater the music, the less
likely it is that anybody else could have written it. A banal tune might
occur to anybody. Fairest Isle is unlikely to occur to anybody short of a

65
PHILOSOPHY OF MUSIC

Purcell. The better the music, the more identifiable it is as the work of the
master who produced it.
Think again about the case in science. In science more than one man
may make a discovery. It is a familiar fact that both Darwin and Wallace
fell upon the notion of natural selection at about the same time. A more
striking instance is the periodic table. In the 1860s there were six indepen-
dent classifications of the elements, all pointing to the periodic table we
now associate with Mendeleyev. His account was superior but any of the
other scientists could have made the final breakthrough. But it is
unthinkable that, had Beethoven not existed, somebody else would have
composed the Archduke Trio. Composers really do not worry about
precedence. They are not perturbed by the prospect of somebody else
publishing their discovery before they do. For what they do expresses
something deeply individual. (I argue later that this is profoundly
involved with the way we value art). It is true that had Beethoven not
existed music would almost certainly have changed over that period of
thirty years when he was the greatest living composer. But it is unlikely
that it would have changed precisely in the way it did.
But not only does the notion that musical works exist from eternity to
eternity in the way some mathematicians think the natural numbers exist
collapse notions of possibility and existence, the notion that these works
are discovered plays fast and loose with our idea of discovery. Think
again about Galoiss work in mathematics. The mathematical theorems
Galois discovered count as discoveries because the expressions are true. If
the expressions were wrong or there were mistakes in his proofs, these do
not count as discoveries. Change the proofs and they will, almost cer-
tainly, no longer be proofs. But if we were to alter Beethovens Arch-
duke Trio, as an editor might, we have neither another work of music nor
have we turned something that is true into something that is false. The
Archduke Trio is still the Archduke Trio; depending on the number of
alterations we have made, we might think we have a new version. It is
important to see that notions of truth and falsity, central in mathematics,
do not apply to music. The mathematical and musical cases are, crucially,
not analogous.
Consider another disanalogy. A proof in mathematics can be lost but
cannot go out of existence. A musical work might be lost and found again.
But it is not lost if the score is destroyed by the composer. Brahms tore
up many of his scores shortly before his death. These works are not lost.
They were destroyed. They no longer exist.
Certainly a discovery may be made by accident. Digging in my garden,
I might discover a Roman coin. If I am searching for it, I already know

66
 THE WORK OF MUSIC

something about what I am looking for and shall be able to identify it

when I find it. I may come across a coin and think that it is Roman where-
as it is in fact of later date. But these cases are very different from the
cases in mathematics that seem to provide the closest analogy to music.
For one thing, although I can quite properly say that I discovered a
Roman coin in the garden, here I can replace discovered with found.
But we cannot say that Beethoven found the Archduke Trio (except
in the entirely irrelevant sense of finding a mislaid manuscript copy).
Indeed, Beethoven could hardly have discovered the Trio by accident.
That is not a possibility. We can make no sense of the idea that Beethoven
might come across the Archduchess Trio and mistake it for the Arch-
duke Trio.
This is not to say that there might not be some room for serendipity in
the composing of music. Suppose that a composer comes across a tune
in the rather haphazard way that Elgar seems to have hit on the Enigma
theme. After a tiring day teaching the violin he was sitting at the piano
with a cigar doodling towards a tune that his wife immediately liked; he
entertained her by turning it into a character sketch of a friend. Might we
call this a discovery? We might well, but such a tune is not a 35-minute set
of variations; it required time, technique and imagination to work it up
into a substantial orchestral piece. The tune might have been discovered
but Enigma Variations was not.
What I am inclined to say is not that the notion of discovery has no
place at all in thinking about music but rather that its role is very minor
indeed. Most of what a composer does is very unlike normal cases of
discovery. The sceptic about Platonism will point out that Elgars theme is
very Elgarian in its step-wise motion. It is the sort of tune that Elgar would
have hit on just as the sort of passage work with which Charlie Parker
makes something of a banal standard ballad is not the sort of passage work
that another alto sax player would produce, except in mimicry. Of course,
this sort of objection can be dealt with by the realist. Discovery requires a
prepared mind. What Albert Einstein discovered required something of
his genius; it did not occur to William Kelvin. Their backgrounds, their
particular skills and their interests differed. It is no different with music; it
is not surprising that the sort of material discovered by Elgar differs
from the sort of material discovered by Richard Strauss. So much is
common ground with the realist.
Should we think of the composer as selecting the notes of the melody
he composes? Again, in most cases, the usage seems a bit off-centre. A
serial composer might select the tone row on which he composes a
piece. But normally we will say that a melody occurs to a composer.

67
PHILOSOPHY OF MUSIC

Again, however, there are cases and cases. Consider Beethovens note-
books. What he is doing is working laboriously towards a theme that has
the right properties. He doesnt find the theme, he doesnt select it
and it doesnt occur to him in its final form. An analogy, although it is
evidently an imperfect one, is to think of a mathematician as working
towards a proof of a theorem. It is imperfect because the mathematicians
criteria for a satisfactory ending point are the truth of the theorem and its
valid deduction from his starting point, neither of which have any parallel
in music. Interestingly, and it confirms my suspicions about Platonism in
music, we dont say that Christian Goldbach discovered his conjecture
that every positive even number is the sum of two primes (a prime number
being a number divisible without remainder only by 1 and itself), but, if
somebody were to prove it, we would say the proof had been discovered.
Truth matters.
I have made some concessions towards Platonism. What the composer
does seems sometimes like a discovery, sometimes like selection. A com-
poser might quite properly be said to have discovered the theme that is
just right for the last movement of his work. In this case, what he is doing
is quite close to the solution of a puzzle; the manner and thematic content
of his work up to some point imposes constraints within which he has to
work to find an appropriate ending. This may be difficult. Some works
have remained uncompleted and some have even been performed
incomplete while the composer worked on the finale. William Waltons
First Symphony is an example. We may allow that some other features of
what a musician does might be akin to discovery; one might even allow
that a performer could say that he had discovered that a passage makes
sense played this way. But we ought not to rush to the conclusion that an
entire work could be discovered. Even if we grant these restricted roles
to discovery, it does not imply that a theme pre-exists its composition or
that its composition is merely the discovery of a pre-existing order of
notes. There are all sorts of nuances here in the differing locutions that
philosophers rather typically overlook in their pursuit of metaphysics.
The question that is begged in this discussion is the question of what
the identity of a piece of music consists in. Platonists tend to think of it as
a structure of sounds, although such a view would have to allow for the
possibility that it may be transposed up or down a tone and remain the
same piece (or even be transposed to frequencies too high for the human
ear to hear). There is debate as to whether the instrumentation is part of
the music or not. Is it the same work if it is arranged for another instru-
ment? We allow that a lute suite is the same music if arranged for guitar.
Do we allow Ferruccio Busonis transcription for piano of Bachs

68
 THE WORK OF MUSIC

ciaconna for solo violin as the same work? If I arrange Beethovens Fifth
Symphony for steel band is it the same work? Whether we answer these
questions in the affirmative or negative depends to some extent on what
we think is lost or added. There really isnt a clear answer. The further we
go back in the history of Western music, the less it seems that composers
had specific instrumentation in view. Then the less we are inclined to
haver as to whether a particular instrumental version is an arrangement or
not. The notion of an arrangement or transcription applies to music
after, say, 1700, for, by that date, instrumentation is usually specified. You
might, I suppose, insist that a performance of Ravels Daphnis et Chlo on
massed piano-accordions was a pukka performance of the work since the
notes are all there, but all this does is to reveal your assumption that it is
the notes that prescribe the identity of a work. I dont know whether this is
so. Indeed, I dont think that our usage of musical work is that sharply
defined.
If we permit the identity of a work of music to involve its instrumen-
tation, why stop there? We might also include its expressive character, its
gravity or jollity and once we do that further complications enter:
difficulties that raise the question of what philosophers call constructiv-
ism (of which more in a moment). For on these matters there can be
disputes among qualified musicians as to what the expressive features are.
I dont mean the individual markings given by the composer, such as
teneramente or giocoso, although these can be puzzling; it is more the
overall expressive character of the work that I have in mind here. Is it
ironic or do we take it at face value? A decision about such matters is often
critical when playing Shostakovich, for example. Is the imitation of
Tchaikovsky in the Adagio of the Second Ballet Suite, affectionate irony
or parody, for example? The question is to what extent the identity of a
work involves its interpretation and this in turn raises questions about its
expressive nature, its meaning and how that is realized in interpretation.
That Shostakovichs symphonic style involves both a measure of agitprop
and a distancing from such crudity gives it an ambivalence that is very
distinctive and poses problems for the interpreter. It also marks the way in
which the critical discussion of music, its interpretation in that literary
sense, bears on the way it is interpreted in performance. The question of
meaning is a large one and occupies Chapter 3; it cannot, unfortunately, be
briefly but adequately dealt with here. But the question of constructivism
is pressing and it is to that I shall turn.

69
PHILOSOPHY OF MUSIC

Realism and constructivism

We have seen that a natural way to describe the workperformance

distinction is to think of the work as a type but one that has a point of
origin. The type was created by its composer. But we have not yet finished
with the complexities. In classical music a work is interpreted, and this
adds further complications. Before we consider this, I want to make a
distinction between two forms that realism can take, for a second concep-
tion of realism, one that we have not yet considered, is germane to the
question of the interpretation of music.
What is this second form of realism? So far we have considered realism
as what might be described as ontological independence: specifically,
the idea of a work having an existence distinct from its performance. But
another form in which realism appears is in the supposition that there is an
answer to a question we may put prior to the raising of that question. The
issue arises in a very clear way in the philosophy of mathematics. For
centuries, Fermats Last Theorem awaited a proof. Now, a realist will say
that Fermats theorem is either true or false prior to any investigation and,
of course, prior to Fermats formulation. If we find a proof we shall dis-
cover the truth of this theorem. A non-realist will argue that, since mathe-
matical truth depends upon the existence of proofs of a given theorem,
until such a proof is established, the theorem is conjecture and is neither
true nor false. I have, of course, over-simplified a complex matter. Is proof
something itself to be discovered, as a realist might say, or is proof a
concept that essentially involves showing to somebody that something is
the case? Does it involve a psychological element?
I want to look at a partial, although only partial, parallel to these
questions that arises in the domain of artistic performance. As soon as a
performer considers playing a work, she has questions to ask and decis-
ions to make. Look at the opening page of Beethovens late Piano Sonata
in A flat, op. 110. It is marked Moderato. Now, bearing in mind that it is
followed by a scherzo-like movement marked Allegro molto, and that
the ensuing slow movement is brief, (a recitative-like opening followed by
a short arioso, would it be best to consider this the slow movement and
take moderato as close to andante, thus balancing fast and slow
movements in the work as a whole? This is what some great pianists have
done. The marking con amabilita (sanft) means with tender affection
(easily or gently). How long is the pause in bar 4? Should the leggier-
amente passage from bar 12 onwards be played a bit faster?
All these decisions may be thought out in advance. I remember a
distinguished pianist saying that most of his decisions about interpretation

70
 THE WORK OF MUSIC

came when he was away from the keyboard, perhaps taking a walk. If he
decides on the best way to play a movement he might stick to that decision
in successive performances. We say that he has thought out an interpreta-
tion and that he repeats the interpretation in various performances. In the
philosophical jargon, the interpretation itself becomes a type of which the
performances are tokens.
I have not suggested that the way it is played is integral to the work
itself. The work is one thing; its interpretation in performance is another.
The notes are all being played and given their proper values. But it is true
to say that they underdetermine the performance. Especially with a work
like this, we can imagine it being played in different but equally valid
ways. But we need not think of the directions as vague. They are sufficient
unto the day.
But suppose now that we identify the work with the work interpreted
so that each interpretation of a work in fact individuates a new work. To
put it crudely, the work comes into existence with the interpretation much
as a mathematical truth comes into existence as it is proved. This view is
known as constructivism.21 It can be formulated in different ways. Thus
we could, indeed, identify the work with the performance so that different
performances would be identical where their interpretation is concerned
but count as different works. But this is not a popular option. For reasons I
shall explain, and that derive from broader features of modern culture, it
has been usual for constructivists to identify the work as the work-as-
interpreted. So the Beethoven sonata becomes the sonata-as-interpre-
ted-by-Brendel, for example, and Alfred Brendel may repeat this
interpretation in different performances. As recorded performances have
made it possible to compare interpretations to a degree not possible
before, the discussion of varying interpretations has become of consider-
able interest outside the world of professional musicians. Ordinary music
lovers have become interested; this, after all, is the way we experience
music. We experience it through the minds, hands and voices of interpre-
ters. So constructivism, it might be argued, actually brings to the fore a
significant feature of our life with the art. Music is always experienced
through the interpretation of somebody or other. (Of course, it is a view
that applies primarily to Western classical music; I hardly need to reiterate
the point that the requisite distinction between work and performance is
not always found elsewhere. We have seen the differences between jazz,
improvised music and rock music on the one hand and Western classical
music on the other). More broadly, constructivists have been given
succour by the strong tendency in our culture to press the thesis that we
see and hear everything under an interpretation; as it is sometimes said,

71
PHILOSOPHY OF MUSIC

interpretation goes all the way down. There are no naked historical
facts, for example, that it is the historians job to reach. The historian is
both the proponent and the victim of narratives and worldviews. This
widely accepted thesis has unquestionably influenced thinking about the
arts, although it is open to very obvious objections in disciplines like
history.
A question that has much exercised philosophers is whether an inter-
pretation can be true or false. Is a certain interpretation of King Lear, as a
play about blindness, say, true? There are parallels, although not quite
exact parallels, in music. Is there only one right way of playing
Beethovens Piano Sonata op. 110, and is it then proper to say that this
interpretation is the correct one and that the statement This is the correct
interpretation is true? A corollary is the question of whether an inter-
pretation is correct prior to its being considered by the player (and here
there is a close parallel with the issues in the philosophy of mathematics).
On the whole, the consensus is that musical interpretations, much like
literary interpretations, are multifarious, and that no single one is the valid
interpretation, although it is fair to say that not all philosophers agree about
this. Since constructivists allow no independent uninterpreted work, it is
hard to see how the question of correctness or validity in interpretation can
arise for them.
There are, as I said, various forms of constructivism. You might identi-
fy a work with the class of its interpretations, a view that seems a particul-
arly counterintuitive form of constructivism since it, inevitably, makes the
work of art inconsistent; for of many works there are interpretations that
are mutually incompatible. This is particularly evident in performing arts
like music; you cannot play a movement at two different speeds simultan-
eously nor, at a certain point, both make and not make a pause. You may
narrow the class that constitutes the work to include only defensible
interpretations, thereby excluding eccentric judgements about tempi or
dynamics (a jazzed-up version of Bachs Goldberg Variations is then,
arguably, not a viable interpretation). If you are a singularist and believe
that one correct interpretation exists, then you reach the vanishing point of
constructivism. The correct interpretation is then true of the work of art in
much the same way as a description of its other properties, such as its
length, instrumentation and so on. Again, you may want a distinction
between the idea that interpretations change their objects and the idea that
a new object-of-interpretation is created in interpretation (a distinction
that I do not find clear). Or you may think of interpretations as complet-
ing the work. What constructivists have in common is the idea that a
work of art is the part-product of an interpretation. Thus Beethovens

72
 THE WORK OF MUSIC

Moonlight Sonata is not just the score nor the note structure but it is the
sonata under an interpretation and the interpretation gives the possibility
of a performance. Non-constructivism maintains that the work is separate
from its interpretation.
The principal objection to constructivism is that since interpretations
are the stuff of critical debate, we need a distinction between the work and
the interpretations it bears. Without this we do not have the wherewithal of
a debate. If we identify the work as the work plus interpretation, different
interpretations merely individuate different works and a critical debate
becomes merely a confrontation between different works. But surely there
must be a difference between preferring Brendels interpretation of
Beethovens op. 109 to Daniel Barenboims and preferring the op. 109 to
the op. 110. In any case, we can say that some interpretations are not
viable; you may not play the opening movement of Beethovens Fifth
Symphony as slowly as Boulez did in a notorious 1970s recording.
Let me describe a second objection before I continue. I said that some
constructivists think of a work as being completed by an interpretation.
The idea that works need to be completed by interpretation suggests that
they are half-formed before. But we do not think this. In the greatest cases,
we feel both that the creator has got things absolutely right, and at the same
time, because the greatest works are the most multiply interpretable, we
think they are open to many ways of seeing, reading or listening. The great
works of art are both utterly distinctive and open to various ways of
understanding. Their distinctiveness, or uniqueness, is connected with the
way that we think that the creators got it right because we think that cuts or
alterations would be damaging. Mozarts mature string quartets seem to me
among the most perfect of human creations but there is no doubt that they
can be played in different ways. This might look like a conflict. But it isnt,
not as long as we remember the distinction between work and interpreta-
tion. It might be worth remarking en passant that different ways of playing
a work may not amount to different interpretations; there are many
different ways of playing Monteverdi but they do not, I think, constitute
different interpretations. The idea of interpretation really becomes
appropriate with Bach or, possibly, Purcell and thereafter. This is, of
course, all a matter of degree. As scholars and performers converge on
decisions about the correct vocal style and instrumentation, it may become
evident that there are different interpretations abroad. Against this, we need
to remember the far greater range of expressive predicates that can be
properly used to characterize post-baroque music; it can be bitter, ironic,
kitsch, sardonic and so on, and, as I shall argue later, the variety of these
expressive descriptions relates to the variety of ways in which the music

73
PHILOSOPHY OF MUSIC

can be understood. Interpretation comes to the fore once we have to decide

whether this passage is to be taken at face value or not.
A change of locution offers us another way of making a distinction and
thereby suggests a third objection. Let us replace convincing here by
true to the work (which does not mean that I want to say that interpreta-
tions can be true simpliciter). An interpretation can be true to the work
but the work cannot be true to an interpretation; therefore the work and its
interpretation cannot be identical. Symmetry fails.
Fourthly, there is a powerful argument against constructivism that der-
ives from the complicated relationship that exists between the value of the
work and the value of an interpretation. If you are tempted by the thought
that the work of art is either completed by an interpretation or, stronger
still, comprises a class of interpretations, consider the following. A work
of art can be good and an interpretation of it bad. Examples abound.
Could a work of art be bad and an interpretation good? I do not see why
not. Bellinis Norma is an opera of conspicuously poor invention but it is
performed because there are a couple of splendid arias in it; we might do
our best for it. So the constructivist does have problems; because he
excludes the existence of the work of art uninterpreted, he cannot make
the distinction which I have just referred to, between a good work of art
and a bad interpretation of it.
Finally, the possibility of a multitude of stimulating interpretations all
of which are grounded in the work is one thing that confers value on the
work itself. That is another possibility lost to the constructivist. He cannot
rate Beethoven or Mahler as greater composers than Delius on the
grounds, among other things, that Beethoven and Mahler can be played in
many more convincing ways than can Delius.
The major and distinctive problem for the constructivist who wants also
to retain a concept of the work as the locus of interpretations is one I have
adumbrated: the problem of consistency. On the one hand, if we argue that
the work is completed by the interpretation we multiply works, and then
have no means of comparing and contrasting interpretations. But if we
allow that interpretations can really conflict but want to retain the identity
of the work as currently conceived, then the more we draw the interpreta-
tion into the work of art, the more we render the work of art internally
inconsistent. A work of music cannot possess contradictory properties.
There cannot be inconsistencies within it. (If we find inconsistent metro-
nome markings, we assume a slip in transcription.) Some interpretations
may, of course, be compatible and thus jointly sustainable. But it is a point
of enormous importance about the arts that they can be understood in
different ways that may not be mutually compatible and yet may be equally

74
 THE WORK OF MUSIC

plausible. The common-sense realist assumption has always been that

while there are questions as to the properties a work of art has that are
matters of fact and to which there are pre-existing answers, these will be
questions such as whether Hamlet has five acts or whether Mozarts Eine
Kleine Nachtmusik is in sonata form, but not interpretative questions such
as Does Mahlers Sixth Symphony express his reaction to his own life
crises or is it about the collapse of bourgeois society? The answer to the
latter is neither yes nor no, although reasons can be given for one answer or
the other. We shall see, in Chapter 3, what considerations make one such
interpretation plausible and another unacceptable.

Is there a truth in constructivism?

Now the arguments against constructivism, as we have seen, look over-

whelming. Yet might there be a truth in constructivism? Picassos studies
of Las Meninas might alter the way Velasquezs original is understood and
hence change its meaning. On one understanding, of course, this is
uncontroversial. A writer, painter or composer might not understand all
that might be encompassed within the work he has created since there may
be unconscious factors in the creative process. The hard cases are where
the work of art picks up features neither consciously intended, uncon-
sciously intended nor available to the creator in the culture in which the
work originated. Now, I think there are at least some cases where a work
changes in that an interpretation of it becomes so much part of the way of
understanding it that the interpretation becomes the starting-point for an
engagement with it; and although these are often cases where the creator
seems to have an inexplicit intent, not all are. So sometimes constructiv-
ism seems to give an accurate description of the engagement with art.
Some features of an interpretation become part of a tradition. Later in this
book, I shall discuss the performing tradition for Beethovens Moon-
light Sonata, a tradition in which it has become customary to give the
first movement peculiar weight. It is partly connected with a romantic
interpretation of the meaning of this movement, with tales about Beet-
hovens romantic attachments and with the view of it as a sort of nocturne;
no evidence for any of this can be found in the score nor in obiter dicta.
But a tradition has built up which is pretty nigh inescapable. There is an
element of constructivism about all this. The work has been endowed with
features that it did not originally possess. Decisions as to where to place
the climax of a single movement may also come into this category.
Different interpretations of Rachmaninovs Second Symphony may well

75
PHILOSOPHY OF MUSIC

place the salient moment at differing points. But for many works, there is
an agreed tradition and, in many cases, there will not be much intelligent
argument as to where the climax should be; the structure makes it plain.
For many works, interpretative questions can be given clear answers. But
some of the greatest music raises questions to which there are no very
clear answers. In saying that sometimes a work seems to acquire a
meaning, a way of interpretation, during its history of performance, I am
not saying just that it has come to be viewed differently. That would be a
triviality that nobody would deny. It is also a matter of the works identity
that is at stake here. This comes out in the way in which a failure to give
the first movement of the Moonlight Sonata its accustomed weight will
be taken to be an error in performance or, at least, an eccentric interpreta-
tion that needs defending, possibly by a recasting of the later movements.
Reflections on this may suggest a stronger form of constructivism.
Consider this analogy. What I see in my friend is how she reacts to me and
I to her. Does it follow that she has, as a matter of fact, those characteris-
tics? Yes. Then is this how she really is? Well, what significance do you
give to the word really? Sometimes it may be said in extenuation She
isnt really like that; she was rude but it was an aberration she was upset
at the time. What is undeniable and easily observable is that other people
bring out different characteristics in us. With some people you may be
reticent, and with others outgoing. Of course, other features may remain
constant. You are not simply the creation of context. Why should not art be
likewise? It takes on a different character as different people encounter it.
Some characteristics, expressive and interpretative, are permanent. They
will not change. But others do.
For to deny that a work of art really possesses features that attract and
move us, if those features are not seen by others, is no more plausible then
for a mother to agree that her son is not really kind just because others do
not see that quality in him. The non-constructivist might say we see
meaning in music or we project meaning on to the music but we do
not, I suggest, think of art as something we invest with meaning and, were
we to think of it in that way, it would be a very different business. And yet,
puzzled, we acknowledge that others see in art features that we do not see.
(The very considerable implications of this for the way we value art will
become a little clearer, I trust, in Chapter 4.)
It is true that our experience of music is characteristically to be thought
of in expressive terms. We may immediately categorize it as plaintive or
spirited. If you wish to categorize such judgements as interpretative, then
interpretation does indeed, as they say, go all the way down. But, I sus-
pect, to do this transmutes interpreting into something else, something still

76
 THE WORK OF MUSIC

intentional in character, as philosophers would put it, but more inclusive.

So although the initial impact of a work may be exciting, awe-inspiring or
moving, I think of these as basically causal judgements, albeit causal
judgements that lead to expressive judgements about the nature of the
work. They are not properly interpretative because the role they play in any
debate about the nature of the work is distinct. Interpretations are, of their
nature, contestable and a good deal of debate may be needed before we
cancel our initial judgement. I do not have a choice as to whether or not to
hear the opening of Mozarts Symphony in G minor as grief-laden. This is
a reaction and this distinguishes it from interpretation proper, which may
be largely within my power, in as much as I can call upon the relevant
considerations. Of course, we can imagine listening to a great interpreta-
tion, like that of Beecham, and find the music more uneasy than grieving:
neurotic and anxious rather than tragic. After this, we might find it difficult
to characterize the music in other ways and feel that other interpretations,
by comparison, are rather superficial. Then the interpretation moulds our
reaction. We subsequently hear the music as unsettled and if the character
of another interpretation is different we may feel the interpretation is
inadequate to the nature of the work. Nor should we ignore the fact that
there can be uncertainty about the initial reaction. In some cases there may
be no initial reaction at all; perhaps another hearing will enable us to form
an impression or perhaps reaction has to be guided by somebody leading us
to hear what can be heard. In which case, we might well have a case where
interpretation is at the foundation. It does go all the way down. But quite
commonly it will not be like that. There will be some sort of ur-reaction
and when that reaction is very strong we may be reluctant to have it
countermanded by someone else, even when he or she is more experienced.
To summarize, then, sometimes music does acquire features that are a
product of an interpretative history. Sometimes, too, we describe it in
ways that others would dispute, but that we cannot retract. To this extent
constructivism holds. But, then, it is quite common for a great interpreter
to strip away the accumulation of tradition and show a work in a new light.
This introduces questions about performance, and I now turn to these.

Authenticity

What would it be to be a realist about performances, as opposed to works?

Well, in one way, of course, we are bound to be realists about perform-
ances. The performance has the features it has and we, the listeners, grasp
that or fail to. We do not, as listeners, do any interpreting. We simply

77
PHILOSOPHY OF MUSIC

recognize or fail to recognize the way the pianist plays it. We dont endow
it with features that we contribute to the occasion. If we think that it is a
somewhat detached performance then we are either right or wrong about
this. Like tables and chairs, the features of a performance are independent
of us. Constructivism about performances is not much of an option. We are
not tempted by the thought that, as listeners, we add something to the
performance so that the performance becomes the performance-as-
heard. Still, we might wish to raise another question. Can performances
be correct or incorrect? From what has been said about the multiplicity of
interpretations, the answer will seem to be negative, although we may
always raise the question as to whether a performance is true to the
interpretation conceived by the performer. After all, a performer could fail
to produce the interpretation she planned through technical inadequacies or
a loss of concentration.
For those who favour the idea that a performance can be correct, a
natural unreflective assumption is that the correct interpretation is the one
the composer intended or would have endorsed had he settled down to
work out an interpretation. (The latter proviso is important because there
is no reason to suppose that a composer can encompass all the varying
ways in which his piece might be played; it might well be that he would
endorse a number of these, in which case there would, on this inten-
tionalist thesis, be no single correct interpretation.) The matter of the
creators intention has been most widely debated with respect to literary
interpretation; what a poem means, intentionalists propose, is what the
poet intended by it. The opposition argues that the poet is merely the first
reader of his poem. What it means is up for grabs; the reader brings her
own background to the reading. Now Western classical music is different
from poetry in one crucial respect: performance is essential if people are
to encounter the work of music. We can read a poem to ourselves but, in
music, we need the work of the performer before we encounter the music.
Admittedly, some have argued that reading the score is better than listen-
ing to the limitations of a live performance but it is hard to believe that
music heard in the minds ear is equivalent to the experience of hearing
the music played. Brahms famously claimed that reading the score of Don
Giovanni was preferable to hearing the music in the theatre; I dont
believe him and, even if he was telling the truth about his own experience,
for very few musicians or music-lovers could this be so. If performance is
essential, then intentionalists will claim that the correct performance is
not only accurate to the score but is interpreted in accordance with the
way the composer intended on the instruments he chose. It will sound as
he intended. Such a performance will be authentic.

78
 THE WORK OF MUSIC

There is an alternative. This second approach is less concerned about

the business of getting the music to sound as the composer intended. It
stresses the importance of a performing tradition. Pianist after pianist has
played Bach and Beethoven and modern players can draw on their
accumulated wisdom. Interpretative ideas are passed down from teacher
to pupil. Certain technical problems are solved and the newcomer learns
how. This might lead us to prize the inherited musicianship of a modern
player above historical re-enactment. Why relinquish the wisdom of the
ages? There are two things to be said about this. First, there need be no
conflict with the first approach; there is no reason why at least many of the
interpretative insights that Clifford Curzon, say, brought to Mozart cannot
be reproduced when Mozart is played on an authentic instrument such
as the fortepiano. Secondly, and sceptically, I suspect that the performing
tradition is not really cumulative. Indeed, Mahler called it Schlamperei
(sloppiness). On this view the tradition looks more like a gradual
accumulation of performing habits until there is a sudden reversion to a
fresh approach based on an examination of the original score, studied with
fewer preconceptions. This is how conductors like Arturo Toscanini and,
to some extent, Beecham, were seen by their contemporaries. Like the
restorers of great paintings, they strip away decades of varnish. The tale is
much more like Kuhns account of the history of science, with scientific
revolutions establishing new paradigms at various points in its history.
Indeed the history of performance is probably a good deal more like what
Kuhn imagined science to be than science is.22
Why are modern musicians so interested in what is called the
authentic performance of music between 1700 and 1900? The term has
been much bandied about and readers of this book will probably be
familiar with the authenticity movement in the playing of pre-classical
music and in the idea of the historically informed performance. It is, of
course, a modern interest with an historical explanation. Generally,
musicians prior to the latter half of the twentieth century did not think
they were in any way obligated to perform music other than how it
sounded best to them. The nineteenth-century development of an
historical understanding of old texts and what they mean was extended in
the twentieth century to the study of old music. It seemed equally
important that we should treat old music in the same way as historical
documents. We want to ask questions that, as far as possible, parallel the
questions asked by historians: for instance, what could this text have
meant to contemporary readers? The two questions that need to be
separated are How did it sound? and How did people hear it? These
are evidently different. We might, by reconstructing the eighteenth-

79
PHILOSOPHY OF MUSIC

century oboe, produce the sound Handel would have heard. It is then, of
course, an additional question as to what tempo, dynamics and
ornamentation are in order. Suppose we can settle this so that we do have
the sound that these musicians and their publics heard. It is also a fact that,
at that time, the sound would still have had the pastoral connotation of the
shepherds pipe, something that we need to learn and internalize. And then
the whole idea of the pastoral, a concept that played such a large part in
Classical and Renaissance literature, is relevant and has to be
reconstructed. This is something we can learn, although it will not be
second nature to us because we are not children of that culture. Still, we
can learn it sufficiently well so that it no longer requires a laborious act of
imagination to enter that world. These things are not impossible; but they
may require painstaking work from us while the original listeners
acquired it simply through growing up in that society.
If we reconstruct the original sound and pin the correct interpretation
to that which the composer intended or might have allowed at the time of
composition, then we identify an interpretation or a class of interpreta-
tions as authentic, or historically informed, as they now say. Since
this is often put forward as an ideal towards which the performer should
strive, we should consider what arguments might be offered in its defence.
First, it might be argued, the music is the creation of the composer and
we owe it to him to present his work as he conceived it. This prompts one
important question: can we find out how he expected it to be heard? Let us
assume, as seems plausible, that we can get half decent ideas about how he
conceived the music from scholars working on instruments of the period,
from reading contemporary manuals on performing practice or descrip-
tions by contemporary listeners and travellers, and from looking at
contemporary pictures of musicians. Thus the musculature of the face
might show how vocalists produced their sounds and the pictures of
players show how they held the instruments. So the first difficulty is not
insuperable. We can go some way towards meeting it. Of course, the
earlier the music the more difficult this is. But that it is difficult is no
reason for not trying it and there seems no reason to deny that we can be
partially successful at least.
Some argue that since the composers in question are dead, they no
longer have rights over how their music should be played. I dont counten-
ance this argument at all. The death of my mother did not cancel my duty
to respect her wishes about the disposal of her remains. An author cares as
well, that if his books are read after his death they should not be distorted.
For me this argument may trump any consideration that the music might
be improved by a later hand, though it depends somewhat on the stature of

80
 THE WORK OF MUSIC

the music. In one of the works I quoted earlier and which appears in the
discography at the end, I suspect that an Italian original from the seven-
teenth century has been heavily arranged in such a way as to make it
attractive to a modern ear. The result enchants me but I dont approve. No
more do I approve of some of the performances of the Domine Deus from
Bachs Mass in B minor where the conductor has been unable to resist
making prominent a particularly luscious string melody with the effect of
unbalancing the whole. Such an expressive treatment may tickle the ear
but its stylistic inappropriateness damages the whole. Some, convinced
that the point of music is to give pleasure, are impatient with the whole
authenticity business. If you like it what does it matter? The answer to that
depends upon a more responsible and, I argue, morally informed view of
what the business is all about.
Thus I am uneasy about an interpretation that improves a work in
what may, at first sight, seem a more defensible way. Suppose a performer
plays the music in a way that the composer did not and would not have
endorsed and yet can be justified in terms of the score, the composers
cultural background and influences and so on. What then? Suppose you
play the bombast in some of Richard Strausss tone poems as self-
mocking or ironic. Would this be impermissible? Suppose you treat the
triumphalism in Elgars First Symphony as hollow, something to which
the later Elgar might well have consented. In such a context, Elgar seems a
greater figure but it is not obvious that such were his intentions at the time
of composing. In both cases we no longer see the work in its true light and
that is a reason for misgivings.
And these considerations raise the question of the way the piece hangs
together. I have heard a Mahler specialist new to Elgar attempt to play
Elgar as though it is Mahler, that is by treating the music as essentially
contrapuntal, displaying lines of approximately equal weight. Played this
way Elgars music does not add up. The line is melodic and the inner parts
supporting. Where music is primarily contrapuntal, it is important not to
bring out a tune in such a way as to obliterate the other lines. Bach is,
and this is elementary, not to be played on the organ as though the organist
was playing Messiaen.
Why should we care? Suppose we like the grand sound of a Cavaill-
Coll or Father Henry Willis organ and dont care about the authenticity
of the Bach as long as a thrilling sound is being produced? Well, anybody
interested in the arts will be aware of the way the very great astonish us.
The notion of genius is not a Romantic invention.23 The ridiculous
argument that Shakespeares plays are intrinsically no better than those of
his Elizabethan contemporaries and that only the Shakespeare industry

81
PHILOSOPHY OF MUSIC

ensured his pre-eminence evaporates, for all save the cloth-eared, by

comparing a half-decent performance of any mature Shakespeare play
with a very good production of anything by Ben Jonson. What Bach,
Purcell, Mozart and so on achieved is rarely something on which we can
improve. So we need to hear what they composed and not something
factitious. Their pieces hang together. Indeed it is noteworthy that very
few arrangements survive in the repertoire. How often are Leopold
Stokowskis Bach transcriptions played? A few are. Ravels orchestration
of Musorgsky is the most prominent example of a successful arrangement
of a piano piece by another hand and some transcriptions of Bach chorales
for piano are regularly played. Liszts transcriptions of Schubert songs are
played but perhaps ought not to be for they are more show-cases for the
virtuoso than pieces of genuine musical worth.
It may take time for the virtues of an historically informed perform-
ance to become apparent to us. We may need to learn new habits of
listening; we may need to pay attention to what we overlooked before.
Perhaps the music will not appeal to us in its new guise, even after careful
and knowledgeable listening. The conclusion we should draw is not that
we have regrettably lost a pleasure in the music that we once enjoyed.
Rather, we should be glad that the scales have been lifted from our eyes
and that we are no longer under an illusion as to its real value. After all,
there is nothing virtuous in false pleasure; we should not take pleasure in
music that does not have the qualities that we thought it had, music that is
being falsely presented to us as other than it really is.
So there is both a moral and an evaluative reason for playing the music
the way the composers contemporaries would have heard it. We owe it to
the composer to play it in the way he conceived it and it normally sounds
better that way. Admittedly, unlike moral considerations elsewhere, these
vary with the stature of the music. The first argument has far less purchase
with composers who are not of the first rank. It seems less of a crime, if a
crime at all, to mess with Sir Henry Bishop than it does with Bach,
although a few misgivings show a basic decency. The moral argument is
less significant than assumptions about the stature of the music for, in the
matchless, every individual element is part of the total effect. We cannot
revise Mozarts finest works without damage and this extends to the mode
of performance as well. Great works were conceived to be played in a
certain way. Undoubtedly, within these constraints there will be more than
one way of doing it. Bach might not have worried as to whether his key-
board music was played on the harpsichord, clavichord or early piano. He
might have had preferences but any of these instruments would do. But I
suspect that ornaments would have mattered as intensely to him as they

82
 THE WORK OF MUSIC

did to Rameau, and we know Rameau cared, given his instructions on the
matter. If you iron out the notes ingales in Rameau you ruin the effect.
Notes ingales require what is notated as of equal value to be played as a
sequence of alternate long and short notes or vice versa producing a
skipping effect. Styles of playing cannot be disjunct from notation. The
two complement each other and when the first is known and expected, it
will not be notated. The more widely the conventions of performance
were known among the anticipated performers, who were, after all, the
composers contemporaries, the less needed to be written down, and the
more incumbent it is for our scholars to discover what they were.
It has been suggested that authenticity is achieved when the music
produces the result in the audience that the composer intended it to
achieve. Authenticity then becomes a matter of the effect the music has.
This has been called second-level intention. So Haydn, wanting to
shock his audience by the sudden fortissimos in the Surprise Symphony,
may need to have his orchestration beefed up in order to produce the same
effect in a modern audience in a larger hall. The problem then is that other
elements in the music are lost, for the precise timbre of a smaller ensemble
disappears when forces are increased. Besides, this example is somewhat
exceptional. For most music, no such specific effect has been calculated.
The composer is concerned to get a piece right. The reactions of the
audience and their characterization of the piece may surprise but not
bother him. If they find spirituality in his work and he is an agnostic, he
may not be too concerned. Members of an audience may, in any case,
react differently. I do not think there is any good reason to suppose that the
composer composes with an ideal audience reaction as his goal and so I
dont think such a conception can play the role in authenticity that this
theory suggests.
For some Western music the question of authenticity does not arise. It
can only arise when we have a notation without a recorded legacy. It will
never arise for the music of Britten and Stravinsky who recorded all their
major works. We know how they expected their music to sound and future
generations will know how contemporaries interpreted it. The only
question is how much variation from that is permissible. The question is a
question about the limits of interpretative licence and not about how the
music should sound and, although these questions may not always be
easy to separate, nevertheless they are distinct.
Still, in the end I dont wish to rule out entirely an appeal to what
sounds best. How do we choose between the various versions of the
Bruckner symphonies? Bruckner was persuaded by some of his friends to
make alterations. Should we take these changes into account when

83
PHILOSOPHY OF MUSIC

performing these works or assume that he was too easily led? The
authenticity movement has led to a prejudice that first thoughts are best
thoughts. There may even be an element of expressionism in this with the
unspoken assumption that the music will be best if it is closest to its
original inspiration, as though the composer was trying to communicate a
mental state immediately prior to the creation of the music, a mental state
from which he is increasingly separated by the passage of time. But surely
we ought not rob the composer of the capacity to revise. Since the 1960s,
Leo Janeks Jen fa has been played without the orchestral revisions of
the conductor and composer, Karel Kovaovic. These alterations are
particularly striking in the closing pages of the score. Yet Janek intially
approved Kovaovics revisions in letters to his wife, praising them very
highly. Later on, in correspondence in the 1920s, he objected strongly,
although how much of his animosity was due to a dispute over royalties is
hard to gauge.24 I much prefer Kovaovics version. In other ways I think
we need to be less than purist. Another example or two will help. When
conducting Verdis Otello with the NBC Symphony Orchestra, Toscanini
replaced a pianississimo with a piano, explaining that in Verdis day
Italian orchestras played everything mezzo forte and that the only way to
get any variation at all was by exaggerating the markings. Erich Leinsdorf
said that the six horns required in Prokofievs Romeo and Juliet reflected
the weakness of Russian players and that in an English or American
orchestra, four would do. Surely in such cases changes are justified.25
Still, when all is said and done, I find the arguments of those who
espouse authentic or historically informed performance very difficult
to resist. All this relates closely to those questions of value that are
discussed in Chapter 4.

84
Chapter 3

Meaning

Aunt Lily sat down ... breaking off every now and then to say, Its
the thought I appreciate, shes such a thoughtful kiddy, ... It did not
escape us that there was a certain falsity, a greasy and posing self-
consciousness, about these expressions. We had very often been
sharply warned against sentimentality, and though we might have
been able to define it only vaguely as the way one should not play
Bach, we recognised it. Rebecca West, The Fountain Overflows

Music and language

What can music mean? The question arises most directly with respect to
instrumental music; indeed, it seems to have occurred to thinkers after
music emancipated itself from text and dance, so becoming a fully-
fledged art in its own right (what is described as pure or absolute
music). When music accompanies words, its significance customarily lies
in underlining or counterpointing the meaning of the text. In such cases
the question of meaning does not arise in the same way.
So the problem comes to our attention when we think, say, of
Beethovens late quartets or Bachs Goldberg Variations. These sublime
works seem to say nothing for which we could find verbal expression, at
least in the sense that we cannot say after hearing them what we know that
we did not know before. Indeed the fact that music cannot make statements
explains, in part, why it was so tardily embraced as one of the fine arts.
While the arts were expected to serve a moral purpose, to teach and instruct,
it was hard to see music as one of them. Nor can music issue commands or
question although it may, as we shall see, have a commanding manner or a

85
PHILOSOPHY OF MUSIC

questioning tone. The questioning manner of the opening of the last move-
ment of Beethovens last quartet, op. 135, is, in fact underlined by his
writing Muss es sein [Must it be]? in the score. However, the music itself
asks no question that could be put into words. But to conclude from that
that the music is meaningless seems dismissive. Certainly we can describe
it in expressive terms. The music has a character and the character may be
noble, charming, thrilling or whatever. So much we can say, properly, of the
music in verbal terms. But in characterizing it in this way we are certainly
not saying that these are statements that it makes. Of course, we might find
it equally hard to say what we know now that we did not know before
reading Wilfred Owens Dulce et Decorum Est have we discovered that
war is a waste of precious human life? We knew that before. We might say
that a great poet brings these things home to us, and that is so. But
although we may find examples of cases where he tells us what we did not
know before, these can be quite hard to find. What we wont deny is that
statements may be made within a literary work of art statements that may
themselves be true or false and that it may be possible to put what a poem
says into other words, although if the poem is any good it would be
destructive to do so. But the Goldberg Variations does not contain a mes-
sage, for a message that is not communicable in words cannot be a message;
any attempt to persuade us that there is one will involve some distortion of
the word message. Message certainly will not be taken in its usual
sense. However I shall suggest that you may be able to say what it means,
though in saying what it means you wont be saying anything true of the
piece of music. In this respect the case is not so different from making a
non-banal claim about Dulce et Decorum Est. This sounds paradoxical.
I shall explain, but before I do there is some preparatory work to do.
One of the first things that strikes us when we begin to think about
music is its power over us. Why does it affect us so deeply? It is not hard
to see that our natural sympathy for others makes us feel pity for people
who suffer and, the closer they are to us, the more we share their misery.
Although it is certainly puzzling that we should feel sadness for the fate of
Mercutio in Romeo and Juliet when we hardly have to think to remind
ourselves that he does not exist, it is surely not such an apparently
insoluble problem as our being moved by music. After all, Mercutio is a
character, albeit fictional, with cares and worries. We find it more
puzzling that, when Vaughan Williamss Serenade to Music was first per-
formed at a Prom in 1939, the composers wife turned to Rachmaninov,
who was sitting in the same box, to find him in tears, saying it was the
most beautiful music he had ever heard. Mighty sounds can make us
tremble.

86
 MEANING

I am not sure we can do very much to explain this effect. The effect of
music in opera, where the dramatic situation is underlined, is easier to see.
I am very moved by the end of Janeks opera Jenfa because the spectacle
of forgiveness and reconciliation moves me. The modulation of the music
into a major key breaks the dam and it is hard not to weep. Equally a text
will determine the expressive function of the music. When Billie Holiday
sang Strange Fruit, the words and the voice spoke of the sufferings of
black people and the music underlines that. Sometimes quite unpreposses-
sing music tugs at us. The little string phrases in the opening of Act II of
Jenfa express a domestic tranquillity; Jenfa and her stepmother sit quietly
at home. Of course, their poignancy depends on our knowing that Jenfa is
pregnant by a worthless fellow. But music also moves us when it is not
underpinning a text or action.
Perhaps it is worth mentioning, as well, in case anybody is troubled by it,
that we need to distinguish associations from meaning. Some music is
irrevocably bound up with a particular time in ones life. In particular pop
music, just because it is evanescent, and because it is something that
especially appeals to the young, becomes, in later life, suffused with nostalgia
and as potent in releasing the past as the taste of madeleines dipped in tea was
for Proust. But such idiosyncrasies, which typify the power of cheap music,
as Noel Coward characterized it, have little or nothing to do with its meaning.
They are private; meaning is public. They vary from listener to listener,
but the expressive character of music, its lightness or gravity, its stealth or
laconic stride, is something open to any competent listener to identify.
To the other question that is likely to strike the layman who begins to
think about music, there is an answer, or more properly, a set of answers.
This question is very familiar to philosophers who reflect on music. It is the
problem of how we can apply what have been called expressive predic-
ates to music. Examples of these have already surfaced in the discussion.
Philosophical writers have tended to concentrate on the application of the
word sad to music. But there are many others; how do we justify calling
music exuberant, exhilarating, exotic, gentle, grave, funereal,
detached, distant, grandiose, grandiloquent, resigned, comic,
ironic, sentimental or dispassionate? Purely instrumental music can
be any of these without our being able to identify anything that it says in
the way that a piece of ordinary prose says something. Yet we may find this
form of description apt.
Lurking in the background is one salient feature of our relation to
music. We seem to have a need to capture its character in words. Consider
this passage from Beethovens Piano Sonata op. 109, variation IV from the
final movement.

87
PHILOSOPHY OF MUSIC

88
 MEANING

The temptation is to think that because we cannot describe exactly the

expressive nature of this passage, it must be saying something too deep
or too precise for words. What it communicates is ineffable. Certainly
some writers have said this. Very famously, Mendelssohn said that what
music expresses is too precise for words. We can say that the bars follow-
ing bar 8, remote from the main key of E major yet constantly leaning
towards it, express yearning. But there is also something curiously
oblique about the passage the music seems to look at us askance and
there is a consequent air of detachment, which links with the otherworldli-
ness supposed to characterize late Beethoven. We want to characterize it
more exactly, to capture the je ne sais quoi, but it somehow escapes us; we
feel uncomfortable and uneasy at being defeated, as though we have not
quite got there.
But should we feel thus? At one level the music is deeply satisfying. If
we listen to the recorded performance by, say, Solomon, we will feel that
at last a great pianist has the measure of this very difficult and elusive
work and that nothing needs to be said. No qualifications or reservations
are required. But then our urge to describe insinuates itself and we are
frustrated once again. But music might be no different in this respect from
the ambience of an evening spent in beautiful surroundings with
somebody you love or the perfection of a fine wine.
Might we say that some music, at least, refers and its meaning is a
matter of its reference? Some writers have been prepared to say that sad
music refers to sadness. Such music then becomes rather like a string of
proper names. A cheerful passage refers to cheerfulness, it is followed by
a passage that refers to sobriety and so on. But as soon as we spell this idea
out in such detail it looks highly implausible. Normally I refer to
something or somebody in the course of saying something about it or her.
I refer to the Queen in the course of saying that she has celebrated her
Golden Jubilee. The mere uttering of nouns one after another sounds like
a form of linguistic insanity. There are only a few occasions where such an
activity would have any sense; at a reception, a man at the door may call
out Colonel and Mrs Blimp, The Right Honourable James Golightly
and so on. In these cases the giving of names in sequence does, I think,
constitute acts of referring. But normally, even where lists are made, the
entries do not refer. Beachcombers Directory of Huntingdonshire
Cabmen did not refer to these fictional individuals any more than the
dramatis personae at the beginning of the text of Macbeth refer. My late
father-in-law used to discipline his children when they used vulgar
language by insisting that they read sections of the London Telephone
Directory, inserting the offending word after each entry. But, in this

89
PHILOSOPHY OF MUSIC

witless attempt at reforming their usage via boredom, the children were
not referring. (Of course, it reduced them to fits of helpless giggling.)
Offhand, I cannot think of any context that would entitle us to suppose
that the solemnity of a piece of music refers to solemnity. Neither does
Beethoven, in imitating the sound of a cuckoo in his Pastoral
Symphony, refer to an individual cuckoo he might have heard in the
Vienna woods or to cuckoos as a species; no more do I refer to a bear in
particular or in general when I amuse my grandchildren by pretending to
be a bear or by imitating a bear. A conductor once observed that in the
slow movement of Haydns Symphony No. 93, the bassoon imitates a fart.
But we dont say that Haydn or his music referred to a fart. To imitate is
not to refer. When an impressionist imitates the halting speech and
confused syntax and semantics of George W. Bush, he does not refer to
him. Apart from the way music can imitate bird song or the sound of
trains, particular instruments have long established conventional associ-
ations: the oboe with the pastoral, the horn with both hunting and with
cuckoldry. But none of these constitute anything like a referring function.
Most significantly in music, because it does not make assertions, there is
no space for placing referring within an act of communication.
So if music was, in fact, a list of names following each other in
sequence, it would be hard to see that this constituted the music having
meaning. We might equally well call the music meaningless just as we
would call the gibbering of a madman who endlessly repeated names
meaningless. What we might do, of course, is to ascribe meaning to music
by interpreting the sequence of characters that the music possesses in terms
of some overarching narrative. When we describe Beethovens Fifth
Symphony as the triumph of mans spirit over fate, we do just that. Perhaps
this is not a very insightful or interesting interpretation, but it does show
how music might have a meaning via the process of interpretation. But a
full discussion of this must await later developments in this chapter.
Perhaps, you will say, this is all a very big mistake. Perhaps those who
think that music has no meaning at all are correct. This is what, in a
different area of philosophy, is sometimes called an error theory. On this
view it is an illusion that music has meaning. Unsurprisingly, of course, this
depends upon what we understand by meaning. As I hinted earlier in this
chapter, if meaning is construed with the meaning of sentences or
statements as the paradigm, then it is pretty clear that music does not have
any meaning. But music does not have to be construed that way alone. We
also speak of what somebody meant by a certain gesture or a certain tone of
voice. We speak of what a certain action meant. In all these cases, there is
a supposition that the speaker or the agent intended her words or her action

90
 MEANING

to have that meaning, although whether she intended or expected uptake

by the listener or onlooker is another matter. An ironic tone of voice signals
that ones words are not to be taken at face value and it is quite proper to
speak of what somebody meant by using that tone of voice.
So talk of the meaning of music might be premised on the latter sense
(or senses) of meaning. That was a gesture of consolation, I say, and
you misinterpreted it when you took it as ironic. Equally a musical pas-
sage might be consoling. It is a short step to saying it expresses consol-
ation and a somewhat longer step to saying that it has that meaning.
Perhaps the latter is acceptable if marginally unidiomatic. But it is cer-
tainly idiomatic to say that a passage by Schubert is meant to be consoling
and that, if we miss this, we misinterpret or misperform the piece. I shall
have more to say about this sort of meaning later. All in all, though, I think
talk of the meaning of music here uses meaning in a somewhat vestigial
way. We can find locutions where it is proper to talk about the expressive
features of music as a matter of what the music means or what the com-
poser meant by the music but even here there is a slight feeling of discom-
fort, a sense that language is being stretched to accommodate a figure on
which it is not a comfortable fit.
Famously, Stravinsky denied that music expressed anything at all,
either a mood or a state of mind, and this, a fortiori, would prevent it
having a meaning even in this somewhat attenuated sense.1 Stravinskys is
what I take to be an error theory. It is not plausible. It flies in the face of
what we do say about music; later on, he modified it considerably,
stressing the conventional connections with extra-musical phenomena. If
the connections are conventional then of course, you might argue that
there is a parallel with the way in which the meaning of words is
established. In one sense the meaning of cat is conventional in as much
as other languages have different words for cat and we can introduce
technical terms into a language simply by specifying what they are to
mean. But it is seriously misleading to suggest that the expressive
significance of music is conventional. That music that seems to rage is
expressive of tempestuousness is not unconnected with the fact that
someone in a tempestuous mood shouts and is voluble, their voice rises
and falls, and they wave their arms about in animated gestures. This does
not seem to be a matter of conventions that might or might not be found in
a particular society. As will by now be clear, I reject an error theory.
However, we have yet to determine in what sense we use the word
meaning when we say that music has meaning. We still await an answer.
Set aside, then, the question of whether music means in the way that
language means. For the expressive descriptions that we make of music

91
PHILOSOPHY OF MUSIC

we may use in characterizing people independently of anything they say.

Their exuberance may show as much in the way they move, their gestures
or in their faces as in what they say. The gist of the problem is this. To be
sad or to be happy, to be bold or to be fearful is not just a matter of having
a certain private feeling that we may, or may not, betray to others by our
actions and behaviour and by what we say. It is widely agreed among
philosophers that there is what is often called a cognitive component to
our feelings. To be angry with somebody normally implies that you think
he has injured you or somebody else in an avoidable way. He is guilty of
a misdeed. You have beliefs about the situation. When I call somebody
angry I imply that he is likely to behave in certain ways. I imply, as
well, that he is angry about something, which in turn implies that he has
certain beliefs about the situation and so on. But music seems to lack the
requisite cognitive elements; it has no beliefs and no propensities to
behave.
Before I go on to discuss the various ways philosophers have tried to
account for the difficulties in applying these descriptions to music, let me
recapitulate. What is the significance of all this? Because of these ways in
which we characterize instrumental music, it is not a decorative art but
one of the fine arts. It is, we might say, a humanist art, much like the
representational arts: literature, drama, film and painting. It is not like the
patterning of a wallpaper or of the arcading of a great building. Certainly
these can be described expressively but in a much more limited
vocabulary. A pattern might be frenzied or calm, possibly, but not ironic,
nor sad tinged with sentimentality, which music can be. Indeed it is highly
probable that it was the great expansion of the ways in which music can be
characterized that led to its being embraced as one of the fine arts towards
the end of the eighteenth century (although the terms in which the matter
was being debated suggested that music was then being thought of as
expressive or representative of mental states). To describe a mere pattern
such as the architectural arcading of a blind-storey as melancholy would
be to place an onus upon the speaker to defend this claim. As it stands it
looks a bit like a metaphor and needs some teasing out. She will need to
point to some detail that justifies the claim. But in the case of music such
talk is unproblematic, save to philosophers, and it is not thought of as
especially figurative.
Now it is not just the range of expressive predicates that makes music
unequivocally one of the fine arts. Like the arts whose medium is
language, we can follow music. We usually have some expectation, not
necessarily consciously formulated, as to how a phrase will end. At the
simplest possible level, we expect the phrase

92
 MEANING

j j j
= b= = = = =

A D E

to end
j
= b= = = = =
A

I give the letters of the notes for those who cannot read staff notation but
know the names of the notes on the piano. The figure is contained within a
single octave. It is the ground bass in an anonymous southern Italian
seventeenth-century Passacaglia della Vita, the text of which is in
Italian but bears the Latin title Homo Fugit velut Umbra [Man Flees like
a Shadow]. (Details of the recording are in the discography.)
As has been well understood for half a century or more, this mirrors an
important feature of language. If I leave a phrase uncompleted, you may
well be able to complete it for me. With clichs and proverbs this is easy:
In this day and ... [age]; Too many cooks spoil the ... [broth]. Our
ability to fill in this way helps us to combat distortion or interference
when we are listening to public announcements in an airport or on the
railway. This feature of language is partly a matter of grammar or syntax
(grammar may call for a noun at a particular juncture) and partly
semantics. We dont have too much trouble filling in the missing words,
although, of course, we can make mistakes. A famous pub sketch from the
television series The Two Ronnies makes use of the latter. Ronnie Corbett
has a propensity for getting lost at the end of his sentences and Ronnie
Barker offers suggestions as to the word or phrase on the tip of Corbetts
tongue.

RC: I always used to work with ...

RB: With pride?
RC: No.
RB: Within reason?
RC: No.
RB: With an overcoat on?
RC: No. With Harry Hawkins.2

93
PHILOSOPHY OF MUSIC

But with music it is the syntactic features alone that enable us to supply
what is missing, for there is no semantic element. Different notes and
chords do not approximate to nouns, adjectives, prepositions or pronouns.
If we attempt a glossary of music, we will take either whole phrases and
whole harmonic sequences or notes in relation to the tonic and give them a
character of the sort I have described above. Thus we may think of a rising
major sequence as optimistic or a chromatic descending passage as sad.3
But the syntactic element that enables music to be followed is crucial to
the way we value it. Music that is merely a concatenation of agreeable
sounds does not have the same appeal. The very memorability of music
connects with this capacity to be followed. The recognition of this feature
has led some to think that the debate over the meaning of music should be
formulated in terms of a contrast between those who think that the mean-
ing of music is internal to the music itself and those who think that it is
external, in the sense that music can mean what is extrinsic to the music
itself. In fact there is some truth in both positions, although the second
needs careful formulation.
I shall not spend too much time on a theory that was once popular. It is
the idea that a piece of music is impassioned because the composer was
impassioned when he wrote it; the objections to this are fairly obvious and
have been around for as long as the theory has been around. First, empiri-
cal evidence suggests that it is false. From what we know of composers,
they do not necessarily have to get themselves in a certain mood in order to
write impassioned music. It is a familiar point that a composer can write a
triumphal march for a state occasion to order. Secondly, a more philosophi-
cal objection is that we establish the character of the music independently
of anything we know about the composers mental condition. We do not
need to consult him or his biographies in order to find out what the music
is like. This, of course, does not preclude the possibility that his mental
state causes him to write the sort of music he does; but, as I have just said,
there seems evidence against such a causal connection being general. The
idea that music has to be an outpouring of the soul in which the composer
works out his own emotional turmoil is a romantic one. It is the product of
a culture that encourages people to look inwards and in which narcissistic
self-concern goes unquestioned. It underplays the element of craftsmanship
of artists, the sheer pleasure to be taken in making something and getting it
right, so that the transitions are inventive and neat, that the harmonic
structure and design maintain our interest, that it is not merely a collection
of clichs and that the various parts fit together. To get it right is to produce
something that holds and interests the listener. This aside, my main point
here is that it is a mistake to suppose that you could either deduce the

94
 MEANING

character of the composer from the character of the music or find out what
the character of the music is by initially finding out how the composer felt.
You cannot. From learning that a composer was depressed over a period of
months you cannot reliably deduce that the music he produced over that
period was sad nor from the tragic nature of a funeral march can you
deduce that the composer was downcast when writing it. Any auto-
biographical account of how music has the expressive significance it has
must come to terms with the fact that most of the major settings of the
Requiem and Mass after Beethoven have come from unbelievers.
This is not to deny that we may be particularly moved by a work
which strikes us as authentic or sincere, or as the raw portrayal of a
real human beings emotions it is hard to dissociate the music of
Shostakovich from the life of the composer, for example and we may
love music for the power it has to bend us to its mood. The point is rather
that not all the music we love and value is either autobiographical or
causally active in this way. Expressionists tend to over generalize about
this and, I think, hidden here in some forms of expressionism is the
thought that Hanslick rejected: that the value of music lies in its power to
move us to the emotions it expresses. I shall argue, to the contrary, that
while the value of music lies, in part, in its power to move us, this is not to
say that what it moves us to are the states that it expresses.
Perhaps, you might think, a more pallid version of expressionism
might fit the bill. When we listen to music that is uneasy or disturbed, do
we imagine a person who suffers these mental states that we ascribe to the
music? Is there a sort of musical equivalent to the narrator of a tale, a
narrator who may or may not be identical with the author?4 Perhaps we
may listen to music, aware that the composer probably did not suffer the
angst that we ascribe to the music, but listening to the music as if there
was such an individual. Behind the music is an imaginary persona
suffering these states. But for many of us, music is not listened to in this
way; there is just the music, which may be wonderful in itself and requires
no props or make-believe. Indeed much music resists this imaginary
exploration. It needs an effort to listen to Bachs Forty-Eight Preludes and
Fugues or some of Stravinskys late music in this way. We might imagine
an austere and rather neutral persona. Even if we do, the familiar difficulty
re-emerges. The persona has to be calculated from the character of the
music. As an explanation of the expressive features of the music, this is
question-begging. We already have to recognize that character in order to
invest the imaginary persona with the relevant qualities.
The expressionist theory I discarded earlier often goes with a belief
that the composer recreates his feelings in the listener via the music. This

95
PHILOSOPHY OF MUSIC

is an element in what has been called the infection theory of art and it is
often associated with Tolstoy.5 Detach the second part of this theory, the
theory that the music recreates its own character in the hearer and you
have a view that I have just adumbrated, which is commonly described as
arousalism. On this view, a view that has become popular in the past
decade, what character the music has it has in virtue of its effect on the
listener. Such a theory owes us some investigation of the nature of our
responses to music, and I shall categorize it as phenomenological in line
with the philosophical movement that tried to investigate the character of
human experience of the world. We may contrast this with what is called
cognitivism; cognitivists claim that it is the music itself that is grave or
exuberant and not the composer or the listener (or the performer for that
matter).

Arousalism

We are now, then, in a position to make a preliminary distinction. Contem-

porary answers to the question What is it for a piece of music to be sad?
fall into two groups. Either they ask what is characteristic of our experi-
ence such that we should call music sad, that is, they look for a phenom-
enological answer, or else they ask what is it that entitles us to call music
sad, and this leads into an investigation of metaphor and enquiries as to
what extent this music is sad is metaphorical. Some of the answers I
shall consider are answers that belong to the phenomenological category
and some to the metaphorical. My main contention will be that no one
account does for the entire range of these predicates; we need different
answers for different contexts.
So one way in which we might explain the grimness of certain music,
such as the opening of Shostakovichs Eighth Symphony or the reflective-
ness of the Jimmy Giuffre Trios The Train and the River is that it makes the
listener grim or reflective on listening to it.6 Now, as I have said, we can
hardly deny that sometimes music has this effect upon us. Indeed, it might
be a testimony to the power of great music that it affects us in just this way.
We are overpowered by the mood it expresses. It arouses our feelings. You
might describe the effect of Purcells Hear My Prayer, Oh Lord as a kind
of twisting in the vitals. But arousalism is not a very interesting or
controversial thesis if it merely maintains that sometimes music moves us
to the moods it expresses. It then is merely a causal claim and one that is not
even universal; sometimes music moves us to the corresponding states and
sometimes not. In order to be a thesis of philosophical interest, it needs to

96
 MEANING

claim more: to propose that in general we describe music in these terms

because of the effects it has on us. It needs to give priority to the response
over the musics initial character. Some of the arguments I shall assemble
are knock-down arguments and I shall end by rejecting arousalism;
nevertheless, as I have just remarked, there is something deeply plausible
about arousalism simply because its emphasis on the effects music has on
us relates to the value music has for us. For no philosophy of art can afford
to stray too far from issues of value. What we prize in music may connect
with its power to move us to a particular mood; this gives us a close link
between expressive power and value in music. So if it seems curious, a sort
of corollary of my initial remarks in this chapter, that we should place such
a high value on such an abstract art, then the fact that music should be so
imperious in its ability to move us may help us to find a connection. We
admire and acknowledge such a power, although I suspect we admire and
value this power more than we value the states it puts us in.
A few caveats are required here; we may be ambivalent about this
power. I might admire the power of a particular piece of music but feel
uncomfortable about that very power. I may even dislike the effect it has
on me. Some of us feel uncomfortable about the effect of Wagners music
not because of his anti-Semitism and the unpleasant and foolish ideas he
frequently tried to propagate in his music dramas but just because he
overwhelms. We feel we have no private domain left. We do not go out to
meet Wagners music as we do that of Mozart or Bach or Miles Davis. He
envelops us. Just because of this I may value it in one way but not in
another. I think it is great music but I dont want too much of it in my life.
There is too much egoism involved in this demand for our submission.
This is the manner of the demagogue or the evangelist.
Now the form of the phenomenological account with which I am
concerned is the arousalist thesis and I have conceded that some music
arouses passions in us that match the music. Reflective music may make
us reflective or grim music grim. There are two questions:

How widely does this apply?

Is such a response a necessary feature of an adequate response to the
music?

Before we examine these issues, I should warn the reader new to the topic
that in no area of the subject is she more likely to encounter very
approximate philosophizing than on the topic of music and the emotions.
Anxious that music that they love should not be seen as trivial, otherwise
good thinkers make claims that are hard to understand and, once

97
PHILOSOPHY OF MUSIC

understood, are seen to be patently false. A very distinguished philosopher

claimed recently that music has profound connections to our emotional
life and is ultimately experienced by the listeners as about them and their
emotional life.7 On any understanding of what our emotional life
means, it is preposterous to suggest that music is connected with it save
superficially. The sequence of rages, infatuations, jealousies and the rest is
barely affected by listening to music and cannot be represented in it. What
emotions we experience in listening to music are invariably evanescent
and even the very self-absorbed could hardly imagine that the music is
about them.
Let us turn to the difficulties with arousalism. First of all there is a
great deal of music that has an expressive face that simply could not be
mirrored in the listener. Music may be sad or happy, detached, grave,
impassive, reflective or exhilarating. But it can also be ironic, bombastic,
grandiloquent or fey. I can see that Charles Minguss Thrice Upon a
Theme is grave and thoughtful in character and I could become grave and
thoughtful in response. Exhilarating music exhilarates me, given that I am
in a receptive state of mind, though at least part of my delight lies in
appreciating the fact that such music exists. But I appreciate the wit and
economy of Benny Goodmans AC-DC Current without being made
witty myself. I can even appreciate the growing sense of tragedy in the
first movement of Mahlers Tenth Symphony, his last, without being made
tragic, whatever that could possibly mean. How could the irony of
Mahler and Shostakovich make me ironic? What am I to be ironic about?
And if some of Elgars music is bombastic, how could that make me
bombastic? Bombast is something displayed in my speech and manner.
But I listen quietly and silently. It is not very plausible to suppose that my
character is temporally changed to being bombastic whereas before it was
shy and retiring. It is certainly implausible to suppose it could be perman-
ently changed. Since classical times, few thinkers have made that sort of
claim for music and it is certainly not part of the arousalist thesis. They
make a claim about the mental effect of music for the duration of the
music and, perhaps at most, for a short time after. Indeed, I can admire and
want to return to a work that does not move me to any particular state just
as much as I may to a work that regularly moves me to tears. As I shall
argue elsewhere, there just is no general pattern.
Let us consider some parallels. You might find the flat lands of the
Netherlands on a wet winters day a glum prospect and be inclined to call
it a glum landscape. Does it make you glum? After a while it can get you
down, but if you are on your way to a winter break in the Bahamas it may
be registered as glum without making you glum. In normal circumstances

98
 MEANING

probably it will make you feel glum principally because nobody much likes
cold rain. Should we then say that subdued music has a tendency to make
you subdued unless there are contrary features of the situation? This is a
natural move. But should we take it? After all, the assumption is that the
proper response to music is to move with its mood. But why should your
appreciation of the composers craft, your pleasure at the way he handles
the form, be any less important, and if it neutralizes the effect of the musics
sadness on your state of mind, why should we dismiss that as an improper
reaction? In such circumstances, we ought not to be prescriptive about
reactions. Music appeals in many ways. A good listener combines alertness
to the technical skill with an appreciation of its expressive character.
Should we perhaps say that we hear the bombast in the Pomp and
Circumstance marches or hear the wit in Lester Leaps In rather in the
way that we might say I heard his misery in his voice? This, however,
suggests a way of construing the bombast or the wit of the music in a way
that rapidly misleads. If I hear the bombast then the suggestion is that I
might fail to hear the bombast if I am inattentive. That is certainly the
case. So to ask Can you hear the bombast in the music? is to draw
attention to what might be overlooked. But, aside from such cases, why
would I want to say I hear the bombast in the music rather than simply
that the music is bombastic. What is achieved by drawing back to a con-
sideration of my response rather than the qualities of the music? After all,
in most cases hearing the bombast in the music is tantamount to recogniz-
ing that the bombast is there and the cognitivist position is conceded
thereby. It is the music itself that is bombastic. Further, to say I hear the
music as bombastic carries the suggestion that it can be heard in other
ways and it further suggests that I am making no strong claim that the
music really is bombastic. The locution is only appropriate in a restricted
range of cases; namely, those where the music has an ambiguous character
that is a bit difficult to settle.
In the end, however, there are even more compelling reasons to reject
arousalism. The vast majority of pieces of music have an expressive face
but they have no effect upon us. They are dull and incompetent. Listen to
half an hour or so of current pop music. The music alternates between the
frenetic and the maudlin. The harmonic structure and melodies are crude
and banal, the music moves from clich to clich and the level of
invention is desperately low. We can give the music an expressive face
while at the same time being bored or repelled by it. The character it has
does not depend upon that character being reflected in us.
Now the arousalist might reply that the fact that we grasp the expres-
sive nature of the music independently does not show that the character of

99
PHILOSOPHY OF MUSIC

the music is not determined by its effect upon us. After all, this is basically
an epistemological matter; we all agree that we discover its character by
listening to it. But that may be compatible with our grasping it in virtue of
an unconscious registering of the music through its effect, an effect that
may not rise to the level of conscious awareness. The problem now is that
we seem to have rendered the theory uncheckable and, in many, if not
most cases, unmanifestable. If we describe a piece as tumultuous in
virtue of it making us tumultuous without our being aware of it, all that we
have is an ad hoc move to preserve a theory under threat. We would need
very strong arguments in favour of arousalism for this to be a defensible
strategy. It would need to be the case either that other arguments yet to be
provided or the intuitive plausibility of arousalism or both have a force so
great that the price of making it unfalsifiable is worth paying. I do not
think that is so.
So, to recapitulate, the character of the music is independent of its
effects upon us. We do not inspect our feelings to find out what expressive
features it has. Any form of arousalism that claims that there is a proper
response to sad music and that is for the listener to be sad (all things
being equal) faces a further difficulty. If the arousalist needs a notion of
an appropriate response, the upshot will be that an inattentive listener,
or a listener who is not properly prepared, or one who is weighed down
and distracted by cares outside the music, may respond wrongly or fail to
respond at all to the character of the music. But the normative element
introduced requires the notion of a match between the descriptions of the
music and the description of its effect upon us. That requires that the
character of the music be something independent of and prior to the
response. We have to know what the expressive character of the music is
independently of our experience of it in order to know that our reaction is
appropriate or inappropriate. It might be that we see this by seeing how
others react but this is neither necessary nor usual. Indeed, we might
criticize the reactions of all our fellow concert-goers as wrong-headed.
On other occasions and in other circumstances, of course, arousalism
gives the right answer. If music is exhilarating or amusing, it exhilarates
or amuses the receptive listener. If it is disturbing, it disturbs her. Some
sections of our expressive vocabulary are arousalist because they mark the
power of the music to affect us. We cannot deny that there are occasions
when music does move us to the expressive states that we detect in it and
there is a group of expressive descriptions that attribute to the music a
causal power to bring about in us just those states. But it is worth
remarking that what we do matches the music as often as what we feel.
Listening to Sviatoslav Richter and Britten playing Schuberts great

100
 MEANING

Duet Fantasie in F minor, I cannot resist a raised eyebrow, a shrugged

shoulder and a half smile at the knowing way Richter (I think) phrases a
rising figure in the bass. Wittgenstein famously remarked on the relevance
of gesture to understanding music.8
The plausibility of arousalism must, in the end, depend on claiming
that these cases are central to our relationship to the arts, even if they are
not universal. The art would be very different, perhaps an art of decoration
only, if it were not the case that sometimes music moves us to states that it
expresses. The music is properly described as grief-filled and it produces a
response in the listener that, if not exactly grief-filled, is nevertheless
sombre. Perhaps it makes your heart feel full. This is music that we
particularly value. If this is so, and I cant think of a philosopher who has
argued this thesis, it would perhaps be the truth in arousalism, as strong
a case for arousalism as could be reasonably made.
Even if this is true, we still have to decide how we justify many,
perhaps most, expressive descriptions of the music independently both of
composer and listener and this leads us to the second family of explana-
tions: those that treat these descriptions of music as posing a problem
about the development of ways of talking about music, about their
justification and their point.

The basis for expressive descriptions

Whatever causal relations exist between the composer, the music and the
listener, or performer, music and listener for that matter, they do not
answer the general question: what is it for music to be sad? Many thinkers
have come to the conclusion that the gravity of such music is something in
the music itself. As has been said, it is like the redness to the apple and not
the burp to the cider.9 But this suggests that the sadness is heard. We
cannot, however, be very comfortable with the suggestion that the sadness
is a heard property of the music in the way the redness of the apple can be
seen. If we hear the sadness in the music it is more like the way we see that
a friend is sad. The nuances are important.
However we describe this, we already know from Hanslick what dif-
ficulties it produces. How can music be grief-laden? I may be made grief-
laden by the death of a friend and the death of my friend is what strikes me
with grief. There is both a cause and an object here. Something causes my
grief and there is something to grieve about. The cause and the object are
the same in this case, although in other cases of moods or emotions the
two may be distinct. (A frequently cited case is that I may fear death but ex

101
PHILOSOPHY OF MUSIC

hypothesi my death has not occurred so, in this case, the object of my fear
cannot be the cause of my fear since a cause cannot post-date its effect.)
But the music cannot have an object of grief nor is the grief caused by
some loss. Music simply does not make judgements nor possess beliefs.
Furthermore, someone who grieves feels something; she is in what
psychologists call an affective state. But the music feels nothing.
Certainly there are many expressive descriptions of music for which
the corresponding mental states normally lack objects; these apply to the
music without raising this difficulty. I might be sad without being sad at
something in particular or without some particular event making me sad.
There is a case, then, where music might possess sadness in a relatively
unproblematic way. No object is required.
But for many such expressions there is no objectless usage and for
many of these it is the beliefs about the object that give a precision and
determinacy as to precisely what the mental state is. It is the differing
beliefs about the nature of the object that distinguish anger, fear, hope and
jealousy more than the nature of the internal feelings of the person
concerned. If music did have objects for its expressive states, then there
would be a precision and determination about its description. But such an
object is almost always missing. In the case of human emotions, the
objects of those emotions together with the beliefs that the holder of the
emotion has about those objects introduce a degree of determinateness
about the emotion that has no parallel in music.
Perhaps we should try another approach. Many writers, faced with the
difficulties of expressionist or arousalist theories and with the difficulties
of taking descriptions of music literally, have assumed that they must be
metaphorical. Think of the mewing of a European buzzard. This does
seem melancholy (although nobody is inclined to deduce that the buzzard
spends its life in fits of depression while the skylark is a generally happy
bird). Perhaps, then, the relation of sadness to music is like the relation of
sadness to the crying of a buzzard. When we say that the mew of the
buzzard is melancholy, do we mean it metaphorically? I dont think so. So
when we say that music is grave, do we mean it metaphorically? Again, I
would say not. Part of the problem is that thinkers about the topic of
metaphor have been too ready to assume that a sharp distinction can be
drawn between the literal and the metaphorical. There isnt one. When
Macbeth says, My way of life is fallen into the sear, the yellow leaf, we
have a clearly metaphorical phrase on which we may reflect. Sear
means dried-up here, and we might think about what is meant. There are
no new currents of life coursing through Macbeth. His life is drawing to
its conclusion just as leaves turn yellow in autumn. But when metaphors

102
 MEANING

are dead, their use in a remark may be straightforwardly true or false. If I

say that Bertrand Russell had an incisive mind, the notion of cutting that
once belonged with the adjective incisive has long become inactive. The
statement is simply true. Indeed, we may well question whether there is
the vestige of a metaphor here any longer. Incisive simply has an extra
usage. But a sensitive reader of Thomas Hardys much-loved poem
Afterwards will reflect on how the line When the Present has latched
its postern behind my tremulous stay works. She may, after such reflec-
tion, decide that the word tremulous is peculiarly apt to describe the
fragility of life. Such reflective consideration is or is akin to interpretation
and its product is less a judgement of truth or falsity and more a
judgement of aptness. Would we say that life is tremulous is either true
or false? This is a metaphor in full fig and if, when pressed for an answer
one way or the other, we say that it is true, we are merely registering its
insight.
Lets return to music. When I describe music as grave, do I speak meta-
phorically? Well, as much or as little as when I call a mind probing or
incisive, I would say. We need to think of language as occupying a
spectrum that runs from the evidently literal such as George W. Bush is
the current President of the USA to the clearly metaphorical as when I
say This rain is icicle-sharp. But our use of language, particularly the
usage of those who use language well in speech and writing, is immensely
flexible and capable of doing very different things in different contexts.
Vary the context and an ordinarily literal utterance may be clearly
metaphorical. To say that somebody is dogging my footsteps is normally a
dead metaphor with a single literal sense. But if I described somebody
whom I had asked to follow behind me on all fours looking for a lost
contact lens, as dogging my footsteps, the weak joke would play on a
similarity between being on all fours and moving like a dog and would
then carry a metaphorical overtone. If his or her face had any overtly
canine features, the joke might be a bit better, although it would then be in
questionable taste. We move away from the literal in all sorts of contexts,
in punning and word-play, in jokes that turn on ambiguities and on the
occasions when we find something difficult to describe but make an
attempt nonetheless.
Often our description of music is clearly metaphorical, of course. Here
is Tovey on the first movement of Brahms Double Concerto:

Soon the syncopated theme reappears in a pathetic calm in the

orchestra, while the solo instruments weave around it a network of
trills. Before long the calm becomes a stiff breeze, and the breeze a

103
PHILOSOPHY OF MUSIC

storm, through which the first theme cries out angrily in the
woodwind. 10

Sometimes description is not metaphorical at all. Quite often it is hard to

say. One listener or reader might take it as metaphorical and another as not
metaphorical but straight. The speakers intentions are not going to help
here. How often do we knowingly and deliberately speak metaphorically?
In any case, on many occasions it will be a matter of decision to take it one
way or the other and there will be no prior fact of the matter about speakers
intentions as to which way it should be understood.
So perhaps we should set aside the matter of metaphor as rather
unhelpful and consider rather what might make somebody decide to call
music grave. Such ways of talking about music do have a history and
the history is of a widening range of epithets from the medieval dulce to
the much wider range of descriptions to be found in writing on nineteenth-
and twentieth-century music. Many of these will not be true or false of the
music so much as opening gambits in a process of interpretation. Such
interpretations are of their nature debatable and may well be challenged
by the reader or hearer. But more of this later.
Suppose we stick to relatively uncontroversial descriptions, describing
a slow movement by Schubert as sad for example (they usually are).
What makes us accept such a description? In calling it uncontroversial, I
commit myself to its being true. (Perhaps we should remind ourselves
once more that mood-words and emotion-words are only a sub-class of
the words we use to describe music, although philosophers have fastened
upon them to the virtual exclusion of others.) So what can we say to
explain calling such music sad? Such a question has a fairly standard
answer and I am not going to depart from it save by means of a few
provisos. I should, by now, have said enough to make it clear that the
problems are less severe than philosophers have sometimes supposed.
Language does spread itself in such a way that a command of it requires
knowing both central and peripheral applications of words as well as a
readiness to use a word in a novel context. (We saw that this was so with
the phrase work of art). What we need is some explanation of how such
expressive descriptions have come to be applied to music with such
regularity and such acceptance. Certainly, they are not quite central
applications of expressive terms. Music is not made sad by the contin-
gencies of life, the news, bad luck, domestic problems or mental
instability. But I may describe a ruined cottage as a sad sight without
anybody feeling sad and my hearers have no trouble in understanding me.
As I have said we should be wary about calling these metaphorical usages.

104
 MEANING

If they are metaphorical they are on their way to becoming dead

metaphors. The metaphorical content is certainly not obvious. The
descriptions look, for one thing, clearly true or false and this rather tells
against their being fully-fledged metaphors.
So much of all this depends on the context in which the speaker makes
her observation and how we read it. In poetry we look for locutions that
we interpret; in everyday speech we may expect more often to be
informed. Lets return once again to music. Why should I describe a piece
as sad? Think about setting words to music or choosing or writing music
to accompany a dance. If you want to produce music to accompany a
funeral, what music would you produce? Something slow and stately. In
our culture, wild expressions of grief are not quite the thing at funerals, so
the melody would be subdued rather than angular and likely to move step-
wise from one note to the next; the rhythm will not be jagged or
syncopated. (How hard it is to avoid metaphors even in this attempt at
explication!) Again if you are setting a lively text, then a faster tempo, an
energetic and more agitated line will be appropriate. This is largely to do
with the vocal manner we think appropriate for delivering sad news or its
opposite. On such slender foundations, our expressive vocabulary is
extended to music. We do not have extensive points of similarity between
the manner of sad or lively people or the manner of sad or lively music.
But we have enough. I recently heard a violin sound described as sweet,
too sweet for the piece the reviewer thought; what does sweetness here,
which we readily understand, have in common with the sweetness of
tiramisu or the sweetness of Shirley Temple or the sweetness of some-
bodys temper or the sweetness of a kitten? Not much, I imagine. But no
one of these usages is clearly literal and the others clearly metaphorical.
No one is paradigmatic. What these cases have in common, arguably, is
that in all of them there is a quality of being ingratiating but this explana-
tion only helps a little. Somebody might refuse tiramisu on the grounds
that it is sweet but hardly refuse a kitten on the grounds that it was sweet.
Now the sweetness of the violin tone is more like the sweetness of
tiramisu than the sweetness of the kitten. But again the reviewer did not
object to the sweetness of the violin tone because of its sweetness per se,
but because it was too sweet, cloying in fact. The complexity of all this
does not prevent a child from picking up the word sweet, possibly by
hearing and using it for one or two of these cases and then extending it
effortlessly to new cases.
Parenthetically, I think the way in which we speak of pitch as high or
low and of a theme as rising or falling has the same sort of explanation.
Philosophers have found this somewhat puzzling. A flight of stairs

105
PHILOSOPHY OF MUSIC

ascends (if we are at the bottom) and a scale ascends if we start on the
lowest note. Language might spread itself from the first group of cases to
the second because effort is required to reach higher notes, or because the
head voice is required. Even if the precise origins of these shifts is a
matter of speculation, I am sure that the explanation is to be found in such
factors.
Now it may well be that sad does have an established central use and
that usage is that it is people who are, primarily, sad. Possibly this is a
paradigm case of sadness. But it is an easy extension to describing news as
sad, a face as sad, a wreck as sad, and, further, to music as sad and land-
scapes as sad. The procedure isnt different. We certainly do not think that
we speak metaphorically when we say that Schuberts Death and the
Maiden Quartet is sad.
So some expressive descriptions of music are straightforwardly true.
We cannot deny that much music has a character that is determinate and
uncontroversial. En passant, it is worth noting that this expressive
vocabulary is itself culture bound. We cannot describe music of the seven-
teenth century as bourgeois, neurotic, kitsch, sanctimonious, maudlin or
sentimental. Or, perhaps more accurately, were we to do so we would have
to defend such claims in the ways we defend metaphorical descriptions.
We would have to show why one or other of these is le mot juste and we
can imagine debate as to whether such a description is or is not anachron-
istic. In the same way when some critics argue that the final speech of
Othello is sentimental, we may feel that sentimentality is not in the
repertoire of Shakespeare or his characters and that this criticism cannot
be fair. However it seems right to describe Richard Strausss Symphonia
Domestica as bourgeois, Schoenbergs first Kammersinfonie as neurotic,
Saint-Sanss Organ Symphony as kitsch, Charles Gounods Messe
Solennelle de Ste Ccile as sanctimonious, parts of Tchaikovskys Sixth
Symphony Pathtique, as maudlin and Schumanns Liederkreis as
sentimental. We do not always appreciate how much information we
introduce into our assessment of this music. We have some common
understanding of the history of our culture that remains unformulated but
which determines the range of descriptions we think apt for a particular
piece of music. Certain expressive possibilities exist at a particular time
and place. Thus Richard Strausss settings of poems about classical
Greece have an extraordinary mixture of wildness and opulence that must,
I imagine, owe something to Nietzsches revisionist views about the
classical world. (I am thinking especially of the Gesang der Apollo-
priesterin op. 33 no. 2, although the style is more apt for Elektra). We
judge a piece of music according to our knowledge of its historical setting

106
 MEANING

and if we stray beyond this range, what we say will look more metaphor-
ical, more a case to be mulled over.
In attempting to show how expressive descriptions become attached to
music I have made much of the fact that what we may call concert music
for instruments only is comparatively recent and comparatively restrict-
ed to Western and Indian cultures. Music has been and still largely is a
mixed media affair. Apart from its use in setting words, the other main
connection with extra-musical features is, of course, the relation of music
to dance. With some music it is difficult to keep still and dance can
exhilarate. The sound track of the cult movie The Blues Brothers is
infectious and the dancing hilarious and liberating but the exquisite music
that Rameau wrote for Les Ftes dHb also demands movement, albeit
movement of a more sedate form. Recall the adage that the farther that
music is from dance, the more it is moribund. For most of its history,
music either set texts or was an accompaniment for dance and it may well
be that purely instrumental music would not have the effect it has if we
were not brought up in a culture where our first experience of music is
likely to be through singing and being sung to and moving in time to
music. Perhaps this is where we should look here for the primitive basis
for our expressive descriptions of music.
This rather familiar idea of music as evolving into the purely
instrumental has tended to involve a rather controversial idea of progress,
stigmatized by English-speaking philosophers as Hegelian; it is some-
thing we should resist. We ought not to assume, as was assumed quite
widely in the nineteenth century, that music marches onwards and up-
wards towards pure or absolute music rather like the notion, contemporary
with this, of evolution as the ascent of man. For one thing, if we think of
music as making progress towards a greater independence from words and
movement, we may think of pre-Renaissance music in the wrong way.
The sacred music of Ockeghem and others certainly owed little to the
constraints of the text. These composers were not illustrating or matching
the words in any way. To all intents and purposes the elaborate and
intertwining vocal parts where many notes were devoted to a single
syllable was an instrumental music in which the instruments happened to
be human voices. Be that as it may, the assumption that instrumental
music is superior is no more than a prejudice.
To summarize, then, I suggest that the application of words like
grave or lively to music is a modest extension of the way we describe
human beings and animals in a way that is neither unusual nor particularly
puzzling. Although the sadness of music is not very much like human
sadness, the points of contact are sufficient to make such an extension

107
PHILOSOPHY OF MUSIC

easily acceptable. Sometimes it is pretty obviously metaphorical; more

often, not. In the end the problem of expression comes, like so much
else in philosophy, from a faulty view of language.

The limits of musical description

If we can solve this problem of expression, albeit in this rather messy and
non-generalizable way, then we are left with another issue to which
philosophers have not devoted much attention at all. Just how far can we go
in describing music in either expressive or non-expressive ways? For
example, one writer says of Sibeliuss Fourth Symphony that the tritone
destroys the music of this work and leaves it bereft of melody; in this he
sees Sibeliuss comment on the destructive force of Schoenbergs atonal-
ism.11 Consider another example. A correspondent wrote to Sibelius shortly
before the composers death, asking whether his symphonic work related in
some specific way to landscape and the geology that underlies it. Sibelius
confirmed this very abruptly in a reply. It is difficult to make this claim
precise but the use of metaphors like large-scale contours or slabs of
material does seem peculiarly apposite to, say, the finale of the Fifth
Symphony, where the great horn theme is mirrored at a much slower speed
in the bass. Some listeners to Mahler, noticing his penchant for little
fanfares on horns or trumpets that seem to come from the distance, find
themselves imagining a landscape with distant hunting calls. The Scherzo
from the Fifth Symphony comes to mind. Indeed, it is a rather striking
feature of some of Mahlers music, and some of Dvoks as well, that it
possesses a sort of three-dimensionality. In some of Dvoks Slavonic
Dances, one theme seems to disappear behind another only to reappear
later, like dancers circling. The Trio of Slavonic Dances op. 46 no. 5 is
an excellent example of this. These are examples taken from non-
programmatic music that does not set a text. Are these reflections justified?
What sort of illumination is being offered?
This is a problem of practical import. Some analysts of music regard
such reflections as self-indulgent and irrelevant. Whether we are dealing
with Western classical music, jazz or world music they will sternly point
us in the direction of formal analysis and key structure as what is crucial
to musics quality. But this puritanism robs us of something significant
about music: that it is open to interpretation in a sense other than that in
which it is usually employed. When we talk about the interpretation of
music we usually refer to the decisions a performer has to make about
tempo, dynamics or phrasing. A performer of classical music, unlike an

108
 MEANING

improviser, has to decide how to play the notated music. When the
composer writes a hairpin showing that he wants a crescendo at this point,
just how much louder are we to make the music? The notation does not
prescribe in such detail. This is what interpretation in music currently
means, and because the term has been commandeered to apply to
performance in this way, we have been lulled into thinking that verbal
accounts are no more than descriptions of it, metaphorical or literal.
But one form of musical interpretation is different. We may be
concerned with the significance of the music and that significance may be
a matter that takes us outside the individual notes to broader issues
connected with the culture of the time. Holsts Hymns from the Rig Veda
are not well known, perhaps because they are difficult to programme. The
second series contains a funeral chant for womens voices and orchestra.
The text, translated by Holst himself, ends:

To those whose souls are born of fire,

The poets of a thousand songs,
The Holy Ones who guard the Sun, unto the Fathers,
May he go forth!

By a gradual overlapping of entries in closer and closer imitation,

combined with a gradual diminuendo and a move to remoter keys, Holst
creates an astonishing sense of a multitude of voices welcoming the spirit
in regions of greater and greater space. It is music that succeeds in
capturing something of what mystics are supposed to experience. We
could understand the music in this way even if we set aside, if we can, the
clues the text offers.
Is this a justified interpretation of the significance of the music? Of
course, it goes beyond the notes and this would be enough for some
theorists to call a halt. Music cannot have this sort of significance, they
would claim. It is a formal and abstract art. Certainly, unlike one of the
interpretations of Sibeliuss symphonies that I mentioned, I know of no
obiter dicta to support my reading, although Holst was composing at a
time when there was some interest in Eastern religion. But then, I also
know of no extra-textual observations that legitimize reading King Lear as
a reflection on the various ways in which blindness may be literal or
metaphorical and how blindness damages love. Nor do I know of any
remarks by Shakespeare that show it to be a reworking of a folk tale or a
reflection on the old adage Love is blind. But the absence of clues does
not prevent us from interpreting. That is the job of critics. They are
licensed to go beyond the text and the great body of critical writing about

109
PHILOSOPHY OF MUSIC

painting, sculpture, literature and drama is precisely this: a defence of

certain interpretations that can be made of the works in question. When
we are told that King Lear is about blindness, we see how we can see, read
or produce the play in the light of that claim. We highlight or foreground
certain passages. We dont for a moment think that this rules out other
approaches and other interpretations and, in the end, it may be a testimony
to the greatness of the play that it can be understood in different ways.
Music is not different. Operas, of course, can also be understood various-
ly; but so can purely instrumental music, at least since the dawn of the
Viennese classical era. Given that Western classical music has been
composed with a serious intent for three or more centuries, why should
not the business of interpretation be equally permissible?
It is, of course, and once we see this we should see that the two forms
of interpretation are linked. Decide on the general character of a piece, a
movement or an episode and you see that certain decisions are required
from the performer to bring that character out. As I intimated, performers
sometimes say that the crucial thinking about a work takes place away
from the practice room when you see the overall character of the work and
the phrasing or structural characteristics that display this.
The more basic expressive description is primitive in a sense; it is this
that leads us into the interpretative maze. Lower-level expressive descrip-
tion of the music as yearning or opulent are the precondition of
interpretation in the sense we are now considering. What an interpretation
will typically try to capture are those moments that initially move, delight
or impress us. I said earlier that I may not have a choice as to whether or
not I hear the opening of Mozarts Symphony in G minor as grief-laden.
These are reactions and it is this that distinguishes them from interpreta-
tion proper, which is something that may lie more within my power. A
particular passage may deeply affect me and become, for me, the centre of
the work. The rest of the work I see as leading up to that moment or lead-
ing away from it. For most listeners and most interpreters there is a
moment in Sibeliuss great Fourth Symphony when the theme of the slow
movement, assembled very gradually, eventually emerges in all its
majesty. (In the end it is cut short in a way that leaves you feeling short-
changed at one level but not at another.) I find it hard to imagine a
satisfactory interpretation that does not make that passage central. Still, it
may very well be that what somebody proposes as an interpretation will
foreground a different section of the work; I might be swayed by this
and then the interpretation has moulded my reaction. I now have a differ-
ent view of the work. Furthermore, we might subsequently revise a view
not because we were persuaded that our reaction was in some way

110
 MEANING

deficient but rather because subsequent experiences did not highlight

those lines. Then we will foreground these other features and capture
them in a revised interpretation. Or we may consider that any of the these
interpretations is satisfactory. Again, the original view may prove resistant
to persuasion or subsequent experience.
I shall assume that you will allow me all this and reject, as I do, the
critics who think that music should only be analysed as a form of mathe-
matical set theory. To the question as to what limits can we place upon
such interpretation, I give the same sort of answer as I would give for
literary texts. In general, we should not venture beyond the culture of
origin of the symphony, quartet or opera. Thus one reason why a link with
geological mapping is apposite in the case of Sibeliuss later symphonies
but not Beethovens Eroica Symphony is that knowledge of geology
was sparse and certainly not general in Beethovens day. In the case of
Sibelius, it provides a connection with the stark landscape of Finland that
is part of the aura of associations that may legitimately surround a hearing
of the work. The unique feature of music is that it can be afforded a
meaning in this way without there being any lower level of statements
or representations from which the interpretation begins.

Meaning

All this has directed us towards an answer to the question: what is the
meaning of music? What people are thinking when they ask whether
music has a meaning is not at all clear. Do they expect music to express
some spiritual truths, whatever they may be? Sometimes they do.
Messiaen evidently thought that his music stood in some intimate relation
to Roman Catholic doctrine. But if Messiaen thought that music can
express Catholic dogma in the sense of stating it, he was egregiously
mistaken. To say that music expresses spiritual falsity would be to
denigrate the music qua music. It would be an indirect way of saying that
it is sentimental or sanctimonious. I have never heard this charge made
but I can well imagine some might think it a justified charge against
Wagner, Liszt, Elgar or perhaps Mendelssohn. It is a way, and perhaps
not much more than that, of objecting to an overblown idiom. In the case
of Elgar it might be more precise as a criticism of The Dream of
Gerontius. I know that for some listeners the work produces a distaste, an
almost palpable feeling of disgust. The text airs what they think should be
private the death of an individual and is embarrassingly mawkish. To
say that Elgars idiom, which sounds to them like John Stainers The

111
PHILOSOPHY OF MUSIC

Crucifixion overhauled by a student of Wagner, matches this is hardly to

praise it.
I doubt whether people who ask whether music has a meaning have a
very clear understanding of what the question might mean. They certainly
do not usually have an understanding of what would make it an intel-
ligible and, indeed, intelligent question. For music under-determines
the meaning that people ascribe to it. It was long ago observed that the
same music can illustrate a tempestuous love affair or a storm at sea. In
The Dagenham Dialogues,12 Pete and Dud, mistaking the title of
Debussys La Mer for La Mre, declare that you can almost hear the tea
being poured by Debussys silver-haired old mother, followed by the
adding of the cream.
My reference to under-determination provides a clue. If the question of
what music can mean (as opposed to what expressive properties it has) is a
question about interpretation, then music clearly can have a meaning and
the meaning is suggested by the context in which it was produced and by
its history in very much the way that the meaning of King Lear is deter-
mined by history and context. The meaning of a Beethoven Symphony
is like the meaning of Hamlet and not like the meaning of O, What a
Rogue and Peasant Slave am I.
Robert Graves can be a difficult poet. In The Challenge, the
following lines occur:

The Moons the crown of no high-walled domain

Conquerable by angry stretch of pride:
Her icy lands welcome no soldiery.13

His biographer suggests that these lines mythologize his lover at the time,
Laura Riding, describing how a young prince flies his hawk at the moon,
only to lose both bird and kingdom.14 Graves apostrophizes Lauras intel-
lectual authority and supremacy but in chilling images of coldness and icy
remoteness, and in ultimately destructive terms. You will not find this
explicitly in the poem. You might well be able to spot the ambivalence of
the poem, but this cannot be read off the page directly. The critic has to
think what the poem might mean and, with a writer like Graves, link the
poem to his preoccupations at the time of writing. Edward Cone15
connects Schuberts Moment Musical in A flat D.780 with Schuberts
realization that he is suffering from syphilis; his letter of the same year as
the piece, 1824, expresses his despair at the loss of his health. An insidious
foreign element creeps into the music, much as a disquieting thought to an
easy-going nature. Eventually this element takes over. As in the case of

112
 MEANING

the interpretation of the poem by Graves, the critic steps outside what can
be directly read off from the poem or the music. He or she uses biographi-
cal information to construct a meaning that goes beyond the obvious and
this information is extraneous to both poem and music. Knowing it you
see how the work has that character. If you were ignorant of it, you might
still be able to see something of that character but without the degree of
specificity the interpretation allows.
But an influential critic may stamp his authority on one particular
interpretation making it part of the meaning such that later interpreters
start from there. I doubt whether any informed reader of Oedipus Rex can
ignore the Freudian interpretation that is now commonplace. That has
entered the text, so to speak. Beethovens Moonlight Sonata offers a
musical parallel.16 What is called its history of reception recounts how
the nickname became attached to it and how it picked up various
significances over the first century of its existence. (Liszt played it, in the
presence of Berlioz, in complete darkness, not even in moonlight.) The
opening Adagio was given increased weight as the focal point of the
work. It came to be associated with love and, in particular, with
renounced love. It was thought to relate to Guilietta Guicciardi, with
whom Beethoven was romantically involved during its composition.
Tolstoy made it central in his novella Family Happiness. It is no longer
easy for us to hear it with the opening Adagio simply as a first movement.
Indeed, perhaps it is now impossible for us to foreground the two
following movements.
It is always open to a hostile critic to accuse the interpreter of
anachronism, of course, and of reading in what is not there rather than
of drawing out what is latent. There is a divide between those who think
that art of significance is open to constant reinterpretation so as to make it
relevant to use the catchphrase of a decade or so ago and those who
think that great art tells us, inter alia, about the culture in which it origin-
ates, a culture that may be very different from our own, and speaks to us
today only because human nature varies between fairly circumscribed
limits. The rapprochement between the two will come through the claim
that the intentions of the artist may be unconscious. An artist of the past
may write of Oedipal themes without being aware that he is doing so, and
what he writes may be decoded into a Freudian vocabulary that he does
not possess. Certainly Freud believed this. In this respect, music does not
differ from poetry, or from film, drama and painting for that matter. It is
worth remarking here, although it comes up elsewhere in this book, that it
is a mistake to contrast too sharply performance and writing and talking
about music. For the character that a good critic finds in a work may well

113
PHILOSOPHY OF MUSIC

directly affect the way it will be played.17 Characteristically, great inter-

pretative performers also think deeply about what they play.
Interpretation does not have to be so keenly dependent on biographical
information as my musical examples might suggest. Poets and musicians
may create works that have much less to do with their own personal
preoccupations. Sometimes, too, a work can display features that it is very
unlikely the creator would have intended. It might come over as a parody
when this was not intended; it may seem to support a cause contrary to the
one she favours. If it is ironic, of course, then we assume that this is an
attitude that is adopted by the creative mind. Failure to notice irony can
mean that a quite contrary meaning is given to what is said. But if a work
is ironic it is intended as such. (You cannot make much of a case for
unconscious irony.)
So when we are asked what music means, we would be advised to
invite our interlocutors to say what piece of music they have in mind. We
then engage in the same process of interpretation familiar to us from the
other arts. To repeat, asking what the Eroica Symphony means is like
asking what Hamlet means. You might question whether those who ask
what music means are likely to be satisfied with this answer. I dont know,
mainly because different people will have different ideas as to the signifi-
cance of the question, as is obvious from our previous discussion. Suffice
to say that relatively few people have a sufficiently clear idea of what they
are asking for it to be correspondingly clear what answer is required. The
arguments of this chapter rule out some answers to the question. The
answer I give here is an answer to a question that can reasonably be con-
strued in such a way that the answer I give is a plausible one. It is about the
best construction I can place on the question What does music mean?:
the construction least likely to lead to philosophical difficulties. In this
context people sometimes speak of significance rather than meaning
in order to maintain a distinction, a distinction that can be obliterated by
using meaning to cover both the meaning of an everyday statement or
instruction such as Pass the salt and the meaning of Hamlet. The
motive is laudable but the word significance is not without baggage. It
tends to import thoughts about value for a significant work is one which is
important. Consequently I prefer to use the word meaning.
But a problem remains. Somehow we have to find a way of permitting
some interpretations while rejecting others. It may be all right to see
Schuberts mixture of guilt and terror at the onset of syphilis in one of
his piano pieces. Is it equally permissible to see Beethovens Fifth
Symphony as a celebration of phallic power and its climax as a paean to
the male orgasm? Such licence in interpretations is characteristic of what

114
 MEANING

is called the new musicology and it raises the question of how we draw
the lines between interpreting, reading into and using a work.18
Parallel questions arise, notoriously, with respect to opera production.
Opera production is not commonly in the hands of musicians. Possibly as
a result, producers seem to have a very considerable degree of licence.
But when a producer sets Berliozs Les Troyens in 1960s America, and
has the Trojan horse drawn by a procession dressed as cowboys and
native Americans, is he any longer even reading into the work? He is
using it to press upon the audience some banal thoughts about
imperialism that have nothing to do with a nineteenth-century French
composers understanding of the classical world. There are, I think, two
questions to be asked. First, what does an interpretation offer us apart
from a means of whiling away half an hour or so? Secondly, what does it
contribute to our understanding of the music? One way in which an
interpretation can help is that it might direct us to certain details in the
music. Cones interpretation of Schubert does this and the pianist who is
persuaded by Cone may then accentuate certain details so that a
particular structure becomes more apparent. He may play a certain
section with anguish rather than mere melancholy. The listener who
accepts Cones interpretation understands the work anew; she will look
for a reflection of this understanding in the pianists characterization and
judge the performance more or less harshly if it fails to make the nature
of the piece plain. (Of course, the pianist could be instructed to play the
piece this way without being told why but, in a sense, it would be a
hollow interpretation, akin to an actor mouthing the words without
understanding their significance.) Could the interpretation of
Beethovens Fifth Symphony as phallic have a parallel effect? Could it
lead to a difference in the way the piece is played? I doubt it.
Some writers assume that the point of all this is to help us to see the
work in a better light. But, of course, the consequence of an interpretation
may be that we judge the work more harshly. At least one critic finds
Elgars First Symphony so objectionable that he cannot listen to it. He
sees in it an unintelligent and uncritical celebration of British imperial-
ism; it is not easy to exclude this interpretation partly because of the idiom
the symphony shares with the Pomp and Circumstance marches. The
Second Symphony is rather ambivalent. The quiet, reflective close may
anticipate the dissolution of the imperialist dream or it might represent the
ideal of a pax Brittanica. Similarly some find Strausss Metamorphosen
objectionable because it dwells on the death of the German cultural
heritage at a time when something far worse than the destruction of the
Munich opera house was happening in Germany.

115
PHILOSOPHY OF MUSIC

We are, to an extent, retreading paths trod in the discussion of

constructivism so the next point should not come as much of a surprise. A
way in which interpretations might be kept on the straight and narrow is
to restrict acceptable interpretations to those that the composer intended
to be possible interpretations. Faced with a sexual interpretation of
Beethovens Fifth Symphony, we might point out that nothing in
Beethovens letters or conversations suggests such a view. Judging from
his attitude to Mozarts Don Giovanni, Beethoven seems to have been
something of a prude.19 This cuts little ice with its advocates, who will
insist that such elements can enter unconsciously into the compositional
process. Suppose we are liberal enough to allow unconscious intentions,
what limits should we place upon these? Suppose an interpretation allows
something alien to a composers outlook. It might be anachronistic or it
might reveal something of the composer that he would, had he known,
have preferred to keep quiet. Had Renoir appreciated that his female
nudes, through combining the facial features of a pre-pubescent child
with the voluptuousness of a young woman, suggest a perverted sexual-
ity, would he have painted them? More controversially, had Wagner
realized that the anti-Semitism of The Mastersingers or the Ring might
pass some twentieth-century listeners by, would he have made it more
overt?
There is, then, the possibility of an interpretation revealing something
that the music might plausibly be thought to represent, something that the
composer might not have intended and, in one sense, could not have
intended since, had he known it would be taken that way, he would have
composed differently. For this reason, it is not even feasible to restrict
interpretations to those the composer could have intended. In some cases,
of course, the interpretation is dependent on the composer having specific
intentions. If it is a correct interpretation of Schumann to read the
encoding of certain names in his music, it is because Schumann inten-
tionally encoded them. If they were an accident then any interpretation
that revolves around the occurrence of these names is illicit.
The situation is, indeed, complex. It is natural to want to rest our
interpretation in the culture that provided a seed-bed. We are right to be
anxious about the viability of an interpretation of a Beethoven sonata that
owes so much to a high romanticism obsessed with sexual love. Unless we
relate the work to its culture we risk creating an artefact that merely
reflects back our own preoccupations and never can show us different
ways of seeing our world. But neither can an interpretation merely reflect
what the composer intended to put in. Not only may he have unconscious
motives, but he may also be unaware of currents in society that surface in

116
 MEANING

his art. Interpretations cannot be circumscribed by what the composer

could have intended to be understood by it.
You can deduce from all this that I am not very sympathetic to
formalism (no longer would be le mot juste.) Always latent, even when
listening to instrumental music, are images, a biographical or cultural
environment, and associations that make the music capable of a more
specific meaning. In so far as we are dealing with pieces of music as
works of art, the task of investigating the meaning is always in the offing.
For this reason, I do not think the sharp distinction between instrumental
or pure music and opera, ballet, programme music and music for film
(what are called mixed media or mixed arts) is tenable.20 This is not to
say that what I have represented as fairly general interpretations of music
is the same as giving music a programme. From time to time, and particu-
larly in the nineteenth century, the composer might provide a narrative
that the listener is expected to bear in mind while the music is played.
Among the more elaborate of these is the story that Berlioz provided for
his Symphonie Fantastique, recounting the artist wandering in the country
and his opium-fuelled dreams and nightmares. In the case of some
programme music, the transitions do not make much sense without the
story. The structure of Berliozs Romo et Juliette only seems ineluctable
in the light of Shakespeares play. Where a programme is more important
then, trivially, the understanding of the music is weakened by ignorance
of it. In opera, of course, the importance of knowing the dramatic situation
is obvious. The music may have its own logic but a fuller understanding
requires seeing how the music underpins and comments on the drama and
text. It might be more accurate to say that the dramatic persona is created
by a combination of words, action and music. Notice how the Count in the
finale of the second act of The Marriage of Figaro, adopts a deliberately
plebeian manner when quizzing Figaro about the letter of assignation. The
music is blunt and commonplace, although commonplace music in
Mozarts hands still delights us.

You might say that instrumental music can be understood and enjoyed
knowing nothing of all this. But that does not show that our understanding
cannot be enriched by hearing the music with a knowledge of its setting
and the interpretative possibilities this may set off. If, like some poetry, it

117
PHILOSOPHY OF MUSIC

has to be interpreted in the sense discussed in this section in order to be

followed at all, then that is for the worse. The very greatest music, like
great literature and drama, can often be appreciated in many ways, at dif-
ferent levels, as they say. Indeed, we can even identify music as profound
without knowing what interpretative conceptions cast light on it just as we
can judge a performance as profound without knowing the conception that
led the performer to play it that way. But profundity is a puzzling
phenomenon in music, and it is one we need to address next.

Profundity in music

So, as I draw to the end of this chapter, I want to concentrate on one

particular difficulty that has occasioned a good deal of comment from
philosophers in the past decade. We sometimes speak of music as
profound. What could this mean? We have already decided that music
contains no propositions and therefore cannot express thoughts. But it is
surely thoughts that we think of as profound and thoughts we express as
propositions. If we want instances of profound thoughts we look to
literature. Typically it is remarks that strike us as wise that we will most
often think of as profound. When Samuel Johnson remarks that there is
very little hypocrisy in the world but rather an abundance of self-
deception in its various forms, he displays wisdom and insight. He helps
us to be more charitable in our judgement of others. In imaginative
literature, perhaps such nuggets of wisdom are rarer. Indeed, in a novel
they can come over as self-conscious. It is rather striking that when one of
Shakespeares characters gives a list of such injunctions, the author is
clearly satirical. Polonius is a bumbling, officious and rather pretentious
old fool whose attempts at wise advice to his son fall flat.
But to suppose that literature is always profound through the senti-
ments directly expressed is a mistake. After all, Lears closing lines are so
moving that many of us feel that they are too much to bear. When Lear,
with Cordelia dead, asks Why should a dog, a horse, a rat have life, And
thou no breath at all?, the sentiments are commonplace. Yet if anything in
literature is profound, it is King Lear. Now, perhaps you would point to
something other than the fact that these lines move us as the locus of
profundity. After all, to be profound is not the same as to be profoundly
moving. And that is true. It is when King Lear forces us to reflect upon
love, on the chilling effect on relationships of demanding proofs of love,
on blindness both literal and figurative, that the profundity of the play is
brought home to us. If this is right then profundity has much to do with the

118
 MEANING

possibility of interpretation and the possibility of certain sorts of inter-

pretation; namely, interpretation in terms of the major preoccupations of
human life, love and death. Where music does offer this it does so through
opera, of course, and we immediately think of The Valkyrie, of Brunn-
hildes recognition of human love between Siegmund and Sieglinde, of
Siegmunds decision to choose earthly life and inevitable ageing and
death with the woman he loves rather than immortality as a hero in
Valhalla; we think, too, of Wotans farewell to his daughter and wonder
whether fatherly love has ever been more precisely captured.
But what about pure music, as it used to be called (with the illicit
implication that it was somehow superior to other forms)? What about
instrumental music? Some of it is certainly described as profound. Talking
to a pupil about the opening of the slow movement of Beethovens
Hammerklavier Sonata, Dvok said if anything is profound in music,
it is this and Tovey describes Bachs The Well-Tempered Clavier, Book 1,
no. 18 as one of the profoundest in the whole Forty-Eight. From another
branch of music, let me cite Charlie Minguss Thrice Upon a Theme.
It might seem as though the standard account of the expressiveness of
music does the job here. Profound music has the features that profound
utterance has, albeit without the content that profound utterance has. The
manner of a profound utterance is slow and weighty with a sense of the
importance of the matter. Should we then conclude that if that is all there
is to profundity in music, then the profound in music cannot be dis-
tinguished from pseudo-profundity? It is merely portentous to utter
platitudes with the manner of speaking deep truths. Much wisdom
literature does this, in fact. (Skim through the Old Testaments Book of
Proverbs again). The essays of Francis Bacon often say the obvious with
an air of self-importance and it may be this that led an early critic to
complain that he wrote like a Lord Chancellor; he would have been an apt
prototype for Polonius. If music utters no propositions but merely has the
manner of depth, then its profundity is merely show. As Samuel Johnson
remarked, To be grave of mien and slow of utterance; to look with
solicitude and speak with hesitation, is attainable at will; but the show of
wisdom is ridiculous.21
Yet this cannot be an accurate picture of the business, not simply
because we resist the thought that music never is profound but only
pretentious, but because we can make a distinction within music between
the profound and the pseudo-profound. I am afraid that much of the music
of the holy minimalists strikes me as exhibiting pseudo-profundity; the
music of Henryk Grecki and Taverner is slow and weighty, and the texts
are spiritual but there is nothing very much there. It would need music

119
PHILOSOPHY OF MUSIC

of greater melodic and contrapuntal interest to underpin these pretensions.

And here lies a clue. Music that seems too easily achieved cannot make
any claims to profundity anymore than the mouthing of mere clichs dis-
plays wisdom. Furthermore Beethovens late piano sonatas and quartets,
like Shostakovichs late works, are the products of a composer we know to
have suffered terribly and, in addition, the product of a composer whose
technique is not in question.
In the end, however, the profundity of music connects closely with the
question of its meaning. As we have seen, there is a sense in which music
has meaning without it having any propositional content. Profound music
is music that makes demands upon the interpreter and listener. The word
profound can suggest that which is deep or concealed. In literature and
the other arts, what is concealed is brought out through interpretation.
Interpretation is the means by which we can give a meaning to something
that can be understood in many ways, and I suggest that there is a
connection between depth and the possibility of a multiplicity of inter-
pretations (something that we may recognize without identifying the
varying possible interpretations). This applies also to the more usual sense
given to interpretation in music: the decisions required of a performer.
Thus a pianist, faced with one of Beethovens late sonatas, has decisions
to make. It can be paralysing for all but the very good or the foolhardy.
How do I voice the opening of the slow movement of the Hammer-
klavier Sonata? There seem to be so many ways of playing it. How do I
manage the way the movement gradually takes flight? Every decision I
take has implications for the playing of other passages in the work. By
comparison, playing a piece by Emmanuel Chabrier only poses minor
problems, mainly of a technical sort to be solved by deciding on fingering
and hand position.
There is no denying that music is valued by us in part as a consequence
of its power to express. It would be a different art, an art of ornament
merely, if it could not be optimistic, pessimistic, grandiose, sentimental,
exhilarating or tragic. The peculiar appropriateness of certain music for
certain texts, dramatic situations or public occasions and its power to
move us, sometimes, to the states it expresses are important elements in
our love for the art. Hanslicks thesis that it is false that beauty in music
depends upon the accurate expression of emotions looks like a half-truth.
(Again let us substitute value for beauty.) Of course, the precise
expressive state of some music matters little, but it is one reason why we
think of some other music as good. The expressive state of yet other music
usually substantiates a negative judgement, as when we declare that a
piece is sentimental or cold or faceless. But once again this depends upon

120
 MEANING

the context. Chilliness may be right for this passage at this juncture. And
once again there is no general pattern.
All this has to do with the value of music and this is perhaps the right
place to turn to that complex and difficult matter.

121
Chapter 4

Value

In the field of the arts, impartiality is what reason is in the field of

love; it is the property of hearts which are cold, or only minimally
enamoured. Stendhal, Life of Rossini (1824)

Many of the questions I shall attempt to answer in this final chapter are
quite general questions about the value of works of art and how we
attribute that value, although some of the questions arise in a more acute
and pressing form for music. But before we consider these perhaps there
is an even more general question we should consider. Why, you might ask,
do we need to make judgements of the relative value of pieces of music at
all? After all, doesnt the whole thing smack of university league tables, of
Sir John Lubbocks hundred best books and so on? (The question, of
course, arises for the other arts as well.)
There is a short answer but perhaps it ought to be prefaced with the
observation that talk of value is a little odd here. Its use is, I suspect, just
one of the many ways in which a materialist culture infiltrates areas where
it is manifestly inappropriate. For outside philosophy, it is used mainly to
indicate monetary value. It is largely used to signal the price that a sculp-
ture or a painting might get in the auction room. In the world of music,
nobody much says that this or that symphony is a work of great value.
Instead we talk in terms that suggest a less precise business of comparing
stature. Some pieces of music are great, some mediocre, some engaging
but transient and some poor and so on.
Now for the short answer. Without some comparative judgements, it is
hard to manage in as complex a society as ours where so much in the arts
is available. We do not need to determine whether Mozarts Hoffmeister
string quartet K.499 is the equal of Beethovens Razumovsky string

123
PHILOSOPHY OF MUSIC

quartets, but we do need the sort of judgement that Hume thought obvi-
ous, that Milton is a greater poet than Ogilby or, in the case of music, that
Purcell is a greater master than Jean-Baptiste Lully or that Charlie Parker
is a more significant figure than Ornette Coleman or, in the case of
individual works, that Purcells Hear My Prayer, Oh Lord is finer than
Samuel Wesleys In Exitu Israel. We need comparative judgements
because so much is available and it is necessary to ration our time so that
we afford time only to those works that are likely to reward. If we had
only the handful of books available in the early fifteenth century or even
the relatively small recorded repertoire available when I was a teenager
half a century ago, matters would be different. But we need guidance
when so much is available and the more that is available the more we need
guidance.
There is another point about all this. Composers learn from other
composers. Often they study with them as Beethoven studied with Haydn.
But they learn also from models. Mozart was explicit about his debts to
Haydn. Haydns The Creation was stimulated by the model of Handels
oratorios and so on. In particular, jazz musicians generally acquire their
art by listening to and copying recordings of older musicians. Now, of
course, part of what the listener needs to understand music is a knowledge
of these debts and the way the debts are repaid; sometimes we only under-
stand the music properly when we see this. So to understand a musician it
is sometimes necessary to know his models and good composers often,
although not invariably, use the best available. So in the very intelligibility
of music itself there is a dependence on prior judgements of value. Of
course, great musicians often leave their models behind, as Purcell did
with the French school of the late seventeenth century, or Bach did with
Vivaldi. Sometimes, too, the influences are no longer figures of any great
musical interest. Bach was influenced by a number of minor north Ger-
man composers whose music is now rarely played; any airings given tend
to show that their neglect is entirely justified. Nevertheless some music is
homage to justifiably admired masters, from Stravinsky to Tchaikovsky,
Ravel to Mozart, Faur and Couperin and so on.
Generally speaking, musicians get it right about other musicians, in
particular those who are not their contemporaries. But, like the rest of us,
composers have their blind spots and these blind spots quite often
illuminate their own characteristics. Britten said that he played Brahms
once a year so as to remind himself how bad it is and it was the unfortu-
nate Brahms whom Tchaikovsky called that talentless bastard.
Stravinskys contempt for some of his contemporaries may be put down
sometimes to a quite unnecessary jealousy but some of those he disliked

124
 VALUE

are musicians of a very different stamp whom, perhaps, he could not have
been expected to admire. Performing musicians, in the nature of the case,
have to have a more embracing and catholic taste. They often have to
perform music they do not like.
This mixture of history and model becomes formalized in time. In most
areas of music, a recognized lineage has formed. Read any history of
classical music and you will find a concentration of major figures who are
strung together in patterns of influence and debt. They form a canon. A
history of eighteenth-century classical music cites Bach, Handel,
Scarlatti, Haydn and Mozart as the major figures. Histories of jazz pick
out Louis Armstrong, Benny Goodman, Count Basie, Lester Young,
Charlie Parker, Dizzy Gillespie, John Coltrane and Miles Davis as central.
A canon is less in evidence with popular song but if we were to look at
cover versions of other writers songs we would find that Cole Porter and
George Gershwin and, latterly, John Lennon, Paul McCartney and George
Harrison wrote music that is still used. More to the point, if some recent
writers are to be believed, a rock canon is beginning to form through the
selective reissue on CD of albums of the 1960s and 1970s. The best are
transferred to CD and remain available. In the past few decades, alterna-
tive canons are emerging and the suzerainty of the classical canon, itself
largely German and Austrian, has come under fire. To pursue the pun,
what we have, in effect, is an assemblage of loose canons.
In saying all this I have used a number of locutions that prompt the
most important questions about artistic value. I have spoken of getting it
right, of blind spots and of a canon. We speak as well of likes and
dislikes on the one hand and a consensus on the other, of having a
catholic taste and of the test of time. But what do we mean by these
and how would we explain and defend judgements in these terms? The
implication of getting it right is that there is a true and accurate assess-
ment of the merits of the work that can be obtained. Is this to say that there
is a way in which the goodness or badness of this sonata is something
independent of our judging it to be so, as realists about value maintain? Or
does it merely mean that any judgement I make is open to qualification by
somebody more experienced in music or more in tune with this piece of
music and no more than that? On this latter view, the value of a piece of
music is not independent of human judgement.
This is where the question of value in music debouches into broader
questions in the philosophy of value, as it is called. As I said at the outset,
many of these questions are of quite general significance; their practical
import lies in our need to make a distinction between what the works
value really is and how any individual listener judges it. But although the

125
PHILOSOPHY OF MUSIC

questions are, as I said, general, the divide between work and performance
makes the situation more complicated in music, for here we do character-
istically judge the performance of a work to an extent independently from
the judgement of the work itself.
This all leads us to a perennially fascinating question, probably the
fundamental philosophical question about the arts: is value to be equated
with individual taste or is there any sense in which the greatness of, say,
Bachs Mass in B minor is independent of human judgement? Are we to
be realists or not? We saw earlier that realism can be understood in one of
two ways; both of these have some relevance to the discussion. To
recapitulate, a realist about quarks or charms believes that quarks or
charms exist independently of the researches of the scientist. They are not
merely theoretical entities invented to do a job in explanation. But another
form of realism, and one that does not coincide with that, is that an answer
to questions may exist prior to the questions asked. Thus a realist
philosopher of mathematics of this second persuasion might believe that
every even number being the sum of two prime numbers is a matter of fact
that is independent of our proof procedures but he probably will not
believe that there are numbers to be inspected in some Platonic heaven. A
realist about artistic value could think that whether Berliozs Requiem is a
masterpiece or not is a matter of fact independently of what any musician
thinks but he would not expect to find that proposition lodged in a
Platonic inventory or, for that matter, in the mind of God.
Some non-realists about value have tended to construe the question on
the first model. Its rejection is then for them a fairly straightforward
matter. John Mackies book, Ethics: Inventing Right and Wrong,1 begins
with the assertion There are no objective values; his celebrated argu-
ment from queerness complains that moral values would be strange
entities indeed. Where could we observe them? What would it be like to
discover them? But this is to construe the debate about value as though it
paralleled the debate about whether quarks exist; the question is, then,
whether they are objects like viruses or bacteria. But, as I have indicated,
the debate about realism in the arts is certainly not primarily a question
about the independent Platonic or quasi-Platonic existence of artistic
values; that question rarely arises in post-Platonic times. It is first and
foremost about whether saying that the Eroica Symphony is a great
symphony is saying any more than that I admire the Eroica Symphony,
or, more plausibly, that a conspectus of qualified judges would, yawning-
ly, assent to this judgement. It is not, either, merely a matter of a contrast
between my individual judgement that I like the work and the stature of
the work independently of my judgement. Realists and non-realists both

126
 VALUE

make that contrast, as they have to if they are to do justice to the life we
live with the arts and the way we talk about paintings, literature, music
and so on. The question is: what do judgements about the stature of
the work amount to? Is the greatness of the music like the redness of a
pillar box?
A little tidying up is in order before we continue. Moral values are
properties such as courage, fortitude and compassion. If they are our
values we admire people who display them, ceteris paribus. People who
are compassionate are, to that extent, virtuous. Artistic values may be
originality, vivacity, elegance and so on. They are our values if we prize
works of art that display them.
Those who claim that values are objective (Mackie lists Francis
Hutcheson, Richard Price and Plato as objectivists) probably would not
deny that some values are also subjective (these are the wrong values as
opposed to the right ones). However, I can see no case for values being
either objective or subjective. Neither position looks plausible. If I value
concision and an elegant argument in music while you look for fascinating
textures or, say, very loud music, in what sense are my values objective in
a way that yours are not (or vice versa)? We simply value different things
in music (and dont think that my reference to decibels is intended as a
jibe very loud music does turn some people on). Coming out of a Bach
concert, I heard one man say to another that it was very restful. I dont
value music for restfulness per se but, although I assume that my taste and
judgement in music is superior to anyone who says something so silly, I
would not represent that as implying that my values are objective. The
only sense that one could give to this would be some more or less
factitious sense dependent on a consensus of the informed about what
values to look for in the arts. Perhaps I would say that the concert-goer
who looked for restfulness in music had the wrong values just as a
newspaper film critic recently implied that anybody who values Nick
Roegs Walkabout for the sequence in which the young Jenny Agutter
swims in the nude approaches the film with the wrong set of values. In
fact, we judge a set of values as inferior just to the extent that they either
prevent somebody from seeing what is great in art or to the extent that
they lead him to the wrong set of considerations in judging merit. By
concentrating on restfulness, a listener is missing what can be interes-
ting and stimulating in the music. If we assume that taste is to be cashed
out as a repertoire of values, then that repertoire is judged by the assess-
ments made of individual works. To take values as fundamental in the arts
puts the cart before the horse. The dependency is the other way, as my
examples illustrate. Since values are neither objective nor subjective, to

127
PHILOSOPHY OF MUSIC

make them fundamental and to make the judgement of the individual

work parasitic on those values forfeits the capacity to defend the objec-
tivity of the individual judgement. Not only can the issues of realism and
anti-realism not be raised but the crucial distinction between the
independent standing of the work and individual likes and dislikes is lost.
As we shall see later, it is the individual judgement that must come first.
We work out the values of a listener from the judgements she makes.
Those values are to be criticized to the extent that they obstruct her
appreciation.
The second form of realism I described earlier is where the answer to a
question as to the value of a work is presumed to exist prior to that
question being asked and, therefore, prior to any enquiries we make in the
course of arriving at an estimate of that value. Now, whether or not he
believes in the objectivity of values, a robust realist will say that the object
really is beautiful or ugly prior to the question being raised (unrepentant
realists go for outmoded concepts like beauty). He will argue that where
there is a divergence of judgement, one of the parties is simply wrong.
Indeed, in the last resort, he could say that we might all be wrong about
the value of a work of art, as a realist conceded to me in conversation.
Now, although I am not going to argue this in detail here, I cannot make
much sense of a notion of value so completely separated from what we
love and admire. Such a very strong form of realism I think is difficult to
defend. I think it possible that a work might be fine without anybody
knowing it to be so but I would want to rule out a work being fine where
experienced listeners over a sufficiently long period find against it. If
qualified people have listened to the music fairly and decide it is no good
then I think that the possibility that it is an unacknowledged masterpiece
has been closed off. I would have to add the rider that it is possible that the
interests and concerns of a society might change so radically that what
was once admired is admired no longer. But by that token, we would
conclude to a greater or less extent that they no longer have the same
conception of the arts. The test of time is relative to a certain community
of interest and concerns. But more of this in a moment. What we need for
practical purposes is a means of distinguishing between the real merit of a
work of music and the judgement made about it by any one individual
listener.
Before we go any further, let me explain the sequence in which I
propose to examine these issues. The questions are these:

Does music have value because there are quite general features about
it that make it good, features which can be specified in criteria?

128
 VALUE

Could it be the case that a work is good if and only if it meets with the
approval of an ideally qualified listener?
Is the merit of a piece of music a matter of what philosophers call
inter-subjectivity? Is it dependent on a consensus among the
qualified, an agreement within an elite?
How do we explain differences of opinion, not just within the sphere
of casual listeners but within the elite itself?
Further to this, are there unresolvable disagreements about the merit
of a piece even among the qualified? If there are, does this raise
fundamental difficulties for the philosopher who insists on some
form of realism about judgements of value?
Should we see the judgement of art as, ultimately, particular? On such
a view, single judgements by a qualified listener take precedence;
what generalizations there are are mere rules of thumb, restricted in
their area of application.2

Criteria

Can we have very general reasons as to why a certain piece of music is

good? Sometimes we do say things about music that imply that we think it
good because of certain features it has. We describe it as lively, or as
spirited or moving. Conversely we might describe it as sentimental or
boring. In this are we giving general reasons as to why a piece of music is
good? Some have thought so. If they are right, then the music is good in
virtue of belonging to a certain class of items that meet certain standards.
We know that a good knife is a sharp and well-balanced knife and a good
knife meets these quite general requirements. To start with, then, when a
work of art has value does it have that value in virtue of the work meeting
certain criteria? Is it all a matter of the possession of certain features, such
that if a work of art possesses these it will pass muster? Most of what I
shall say here rules out such a theory. Indeed, the notion that the value of
art is criteria-bound finds little support these days; the traditional
difficulty is that the proposed criteria either seem so vague that it is
unclear what meets them or so precise that it is not too hard to find excep-
tions. Now criteria have to be reasonably specific to count as criteria. A
doctor qualified abroad has to meet certain criteria to be allowed to
practise in Britain. There are specific criteria that a car has to meet in
order to pass the annual roadworthiness test in Britain and so on. These
criteria are not only specific; they do not of themselves incorporate any
positive grading. An extra premise about what counts as a roadworthy car

129
PHILOSOPHY OF MUSIC

or a competent doctor is required for this. But when criteria are rendered
specific in the arts, trouble is inevitable. The rule that one should avoid
consecutive fifths or octaves in writing harmony used to be part and parcel
of the teaching of elementary composition. But it is a rule that, if elevated
into a criterion, would be immediately shown to have exceptions.
Schoenberg explains the rule in the following way. What the rule prohibits
is writing where two parts go from one fifth or octave relation to another
by parallel motion. The assumption is that the independence of the two
voices is thereby compromised. But Schoenberg points out that some-
times octave doublings are used because they sound well or because
avoiding them would take the voice or instrument outside its range.3
Certainly there are many instances of the rule being broken by classical
masters. (Parallel fifths are used by Puccini at the beginning of Act III of
La Bohme to represent winter chill.)
So those philosophers who defend the existence of general reasons for
verdicts about merit usually do so in a way that does not commit them to
very specific criteria. They will talk broadly about unity and variety,
which is hardly precise enough to help. I suppose it might be important to
realize that poor works can be immediately damned because they contain
elementary mistakes that can be picked up via criteria. The beginner had
best learn to avoid consecutive fifths. If the pupil eventually masters
compositional technique then he can decide whether to conform or not.
The problem is, as is generally the case, that in the philosophy of art our
paradigms are the masterly and, in these cases, the satisfaction of general
principles is even more remote from what makes them good or bad. The
peculiar cases of the paradigmatically classical artists, such as Mozart,
really have little to offer in this regard. We dont admire them because
they satisfy certain general good-making rules but because they show
what an inventive mind can do with clichs.
It might seem that to cite qualities like vivacity or elegance offers a
more promising route to criteria. But, given what we have just said about
the nature of criteria, I dont think we would call elegance criterial;
there is likely to be as much debate about whether the work counts as
elegant as there is about the overall verdict. We have not gained enough in
precision and we have, of course, lost by prejudging the issue; we have
imported a judgement of value into the assessment because elegance is
normally a positive feature. I say normally because we might judge that
elegance is inappropriate in a particular context. Elegance would not be a
virtue in a craggy work like Bruckners Ninth Symphony. But elegance
might be a description used to call attention to a passage in another piece,
a passage of significance that might pass the casual listener by. (Perhaps

130
 VALUE

she might not see that it is elegant because the style is unfamiliar.) But,
even here, its merit is hardly likely to depend upon its elegance; calling it
elegant anticipates the conclusion for elegance is a merit. To the extent
that judges disagree about the merits of a piece they will disagree either
about whether it is elegant or whether elegance is a merit in this particular
context.
The way I have explained this might suggest that what philosophers call
the factvalue gap is at the root of all this. How do we get from noting
mere matters of fact about the music, such facts being value-neutral, to the
conclusion that the piece has merit? But what I want to emphasize here is
something different. It is that the judgement of merit is much more
concerned with how a device works in, at the minimum, a passage and, at
the maximum, a movement or a work. We would never say that the famous
Tristan chord has intrinsic merit that would enhance any work into which
it might be placed; in a movement by Mozart it would stick out like a sore
thumb. But it would be wrong to conclude from that that its intrinsic merits
are cancelled out by its inappropriateness to that context. It does not have
any intrinsic merit at all. Enlarge the context to several bars either side and
now it might well make sense to describe it as having merit. It is a finely
composed passage. But inserted in a movement by an earlier composer it
would be unintelligible. What works in one place will not work in another.
Artistic judgement is essentially holistic and individual.
The basic problem, then, is that criteria either have exceptions or are
too vague or they beg the question by importing a judgement of merit into
the considerations on which the verdict is made. Now the defender of
critical generalizations might argue something like the following. Of
course, critical generalizations have exceptions, but in that respect they do
not differ from generalizations elsewhere. Laws of science may hold for
ideal conditions to which the conditions we meet with in everyday life
only approximate. Nobody denies that there are laws relating the tem-
perature and pressure of gases just because, in the empirical case, we are
not dealing with ideal gases. Equally, critical generalizations apply to
idealized cases; in the rough and tumble of ordinary aesthetic judgements
they need to be taken with a pinch of salt.
But the analogy does not hold. Take the examples I have given. In no
sense does the prohibition on consecutive octaves apply to the ideal case.
If anything, the ideal case, the outstanding music of a period, is the music
that is more likely to infringe these rules than not. In any case, the
viability of rules depends on context and may require the judgement of the
experienced hand. Consider performance and consider the rule that a
pianist should not pedal through harmonic changes (because it blurs and

131
PHILOSOPHY OF MUSIC

renders opaque those very changes). This certainly does not apply if you
are playing Debussy. Indeed, even Beethoven often asks you to do just
that. But it is a good general rule in eighteenth-century music (unless you
think that the use of the pedal, or indeed the use of the pianoforte, in music
of this period is a complete solecism, in which case the question does not
arise). Yet even here, there can be disagreements between experienced
musicians; there may, as well, be a way of playing Bach such that pedal-
ling is not an offence. In Beethoven its use needs to be qualified by a
realization of the difference between Beethovens piano and the modern
instrument and this again calls for the judgement of the experienced
musician. So one problem is that the ceteris paribus cases are not
themselves clearly defined. We do not have a clear notion of where the
rule applies and where it does not. For, although I have given some indica-
tions of its applicability, there will always be central cases as well as
borderline cases where instructions on pedalling have to be judged with
an ear to the context.

The ideal observer theory

Set aside, then, the possibility that artistic merit can be assessed by virtue
of general criteria stating non-evaluative characteristics whose possession
automatically makes a piece of music good. What alternatives do we
have? The ideal observer theory is fairly familiar in its moral guise,
although it has not been particularly mainstream in ethics. In essence, the
claim is that to say an act is right entails holding that, if there were an ideal
observer, he or she would approve the act. The ideal observer might, of
course, be an omniscient deity; then the account gives us a way of con-
necting religious belief and moral obligation. The ideal observer theory
has seemed to some to suggest a plausible parallel account of what it is for
a work of art to have value. So the ideal artistic observer, like his, her or its
counterparts in ethical theory, offers a way of defining what is artistically
valuable or what a work of art means without losing sight of the fact that
the way in which art is valuable is deeply connected with what it offers
human beings.4
The ideal observer (or, rather, listener) in music will be somebody who
knows all the relevant facts, experiences the relevant expressive features
of the music and understands all the possible relevant approaches to the
music. In his essay Of the Standard of Taste, Hume sometimes seems to
be proposing an ideal observer theory, although he does oscillate between
that and the consensus theory that we shall be considering next. Any

132
 VALUE

individual critic can attempt to approach the status of an ideal critic by

discounting her individual quirks, the likes and dislikes that she knows are
personal to her. So she might grade the music of Richard Strauss, which
perhaps she dislikes, by setting aside her own predilections and looking at
his mastery of the orchestra, by taking into account the views of other
qualified but more sympathetic listeners, and so on.
Now there are problems; first, there are disputes about what facts are
relevant in the first place. To name only the most obvious, there are
disagreements as to whether the composers intentions are relevant or not
in any debate over what a piece of music expresses or what it means.
Some think they are; some think not. Remember our discussion of the
differing theories about how close performance should be to what we
think the composer intended. Now, disagreements either about what the
creators intentions are or about their relevance need not directly affect
our judgement of the works value. Most works whose interpretation is a
matter of debate are works whose stature is beyond question; the con-
sensus has long since determined that these are outstanding and that they
have canonical status. What these works mean would not be of any
interest if they were not of supreme merit. However, where such disagree-
ments do matter is where they precede the judgements about the merits of
the music. For how we judge the music may ultimately depend on our
estimate of its character; for example, a failure to see that music is ironic
may lead us to arrive at the wrong verdict. The finale of Shostakovichs
Fifth Symphony is a locus classicus for this. The music was long
excoriated as a paean to Stalinist communism; what it really is is a bitter
pretence of rejoicing.
What follows from this is a problem of a different sort for the ideal
observer theory. Given that an assessment of at least some music depends
upon determining an interpretation, the existence of many different inter-
pretations, all of which may be defensible but some of which might not be
mutually compatible, means that the deliverances of the ideal observer
who takes into account all points of view in an impartialist way cannot be
coherently characterized. She has to embrace conflicting interpretations.
Just how serious this difficulty is depends upon what taking into
account involves. It cannot mean accepting for the reasons just given.
If taking into account merely means registering those interpretations
that are viable, then the criticism does not count for much.
We required that the ideal observer be familiar with the expressive,
experiential features of a work. Should she also be moved by the music?
Well, first, intense emotions can get in the way of judgement. Knowing
that this is the last performance of a dying diva might cause her to

133
PHILOSOPHY OF MUSIC

misjudge the scale of the performance. Of course, the fact that she is
moved may be something she allows for or explains away. In other cases,
she might see that the features of the music that excite her are superficial
and will not survive repeated hearings or even see that this is music that
ought not to involve her (although it does) because she can see that it is
meretricious. No real problem here. Secondly, suppose she is not moved;
we might think that somebody who knows what the expressive features of
a piece of very good music are but is not moved by it might not be in the
best position to judge. It is not good enough, it might be said, for the critic
to know that others are moved but not be herself moved. Furthermore,
when a critic, as an experienced listener, sees that it is not the sort of thing
that moves or excites her but judges that it might move someone else, she
ought to be cautious in making a judgement about its value. It might be
technically competent but lifeless. These features complicate but do not, I
think, fatally damage the ideal observer theory. Its advocate would be
entitled to reply that such an individual is not the ideal observer, the
perfect critic.
Although at least the last of these criticisms is not particularly telling, it
points in the direction of the overwhelming problem with the ideal
observer theory. It is that the entire and complex relationship that we have
with works of art, generally and not just music, will be affected if we think
that the position of the ideal observer is something towards which we
should strive, which is how ideal observer theorists see it, even when they
acknowledge that the status of an ideal observer is not attainable. Among
all the works of art available, there are some for which we have a peculiar
affection. Perhaps it is right that we try to broaden our horizons from time
to time and try to understand and appreciate works that are initially
remote from us. But it is also important for our relationship with the arts
that these attempts sometimes fail. For if we were to succeed, the peculiar
way that art is entwined with our lives would disappear. Compare.
Although it might be praiseworthy to try to care for other people who are
remote from me, if this were to be at the cost of the special concerns I have
for my wife, my children and my friends, it would not be an outcome to be
wished for. Such a person would be unnatural and chilling in his lack of
the normal human loyalties and prejudices. And the person whose taste is
not, in some ways, idiosyncratic and peremptory is a bit too close to
Wildes auctioneer for comfort: It is only an auctioneer who can equally
and impartially admire all schools of Art.5 Why should it be preferable to
have a completely catholic taste in music rather than, say, having a passion
for Dixieland jazz or for Indian music? Somebody who has an impartial
concern for everybody, like Dickenss Mrs Jellaby, but who ignores the

134
 VALUE

desire of her children for her special attention does not commend herself
to us as particularly admirable. We might, too, think that somebody who
liked all music equally proportionately, of course, to its merits would
be something of a cultural monster.
I have touched here on an analogue to what in ethics is called partial-
ism. In moral philosophy this is the doctrine that we have special obliga-
tions to our nearest and dearest, to wife or husband, children and parents
and to pets as well. What does being a partialist about art involve? Well,
first, you might love a piece for its associations, for the particular place
that it has in your history. At the same time you may have no illusions
about its merits. Equally, you may share a history with some relative for
whom you may care even though you recognize his faults. Uncle Arthur is
the black sheep of the family; you love him nonetheless. There are no
immediately striking differences between art and morality here. But then,
secondly, in the case of the arts you may think, rightly or wrongly, that a
particular piece is undervalued and have a mission to convert others to an
appreciation of it. Here the differences appear. When you act on behalf of
your children you act in order to ensure that their interests are being met
but, it will be argued, works of art do not have interests. It is in our interest
that they are not damaged, not their own. In this sense the notion of
partialism when applied to the arts is certain to be somewhat etiolated.
Your desire to get other people to listen to a piece of music you believe to
be neglected is more to do with your obligations to other people. You
would not like to have missed hearing this so you want others to hear it
as well.
Hume said, memorably, We choose our favourite author as we do our
friend, from a conformity of humour and disposition. 6 It is interesting
that the partialism that is endorsed here takes, as its model, friendship. It is
proverbial, although perhaps false, that we choose our friends and not our
relatives, but it is certainly possible to love a relative who you find rather
unlikeable. A shared history creates a connection that is not easily dis-
solved. In a somewhat parallel way, as I say, a loved book or recording or
painting being part of my history of taste is often of enormous importance
to my attachment for it and that attachment may survive the realization
that it is not very good. Probably the ideal observer theory can live with
this; the problem is that it does not see an individual relationship of this
sort as anything other than something to be overcome.
My final criticism is another serious objection to the ideal observer
theory; it does not accurately describe how we arrive at an artistic judge-
ment about the status of a work. Certainly a new work may surprise and
absorb my attention and I then may judge that it is of stature, only to be

135
PHILOSOPHY OF MUSIC

surprised when others do not share my judgement. If I accept the verdict

of others, I do not arrive at my assessment by the process of discounting
my prejudice. I arrive at it by looking at the consensus. The consensus will
embrace readers or listeners or viewers of various persuasions and our
assessment of the verdict is not arrived at by discounting their various
idiosyncrasies. These remain in situ. Indeed, how could I know mine or
anybody elses to be idiosyncratic in the first place? Only by looking at
how others judge and react to the music. The consensus theory that, I
suggest, is presumed by an ideal observer theory embodies all this, as we
shall see. But on the ideal observer theory alone, what should I do? It
gives me no lead on what I should discount.
There is a further consideration; if the situation is analogous with
ethics, as its proponents tend to assume, then ought I to appreciate the
music of John Coltrane just as, having found out what the ideal ethical
observer endorses, I ought to act on the recognition that that is what it is
right for me to do? But the notion of a consensus in the arts does not call
for common human reactions or progress towards common reactions.
Those who make up the consensus themselves differ in taste and prefer-
ences and they retain those differences. I can agree on the stature of a
composer or a work while feeling no pressure to like it. On any reading of
the nature of major divergences in taste, these divergences may remain.
The only pressure I might feel is that, in ignoring a work that others
praise, I may be missing something. (Setting aside the much discussed
issue of moral objections to some art; in music such objections tend to
occur mainly in the case of opera.)
In as much as a realist or objectivist account can be given of judge-
ments of merit, it is a sort of surrogate obtained by something closer to
qualified majority rule (where some listeners count for more than others)
rather than by any particular procedure that enables us to arrive at the real
truth, independent of particular judgements about the status of a work of
art. (The ideal observer theory looks intriguingly like the notion of a
scientific method, which, once found and deployed, guarantees that we
shall discover the truth.) We do need the consensus; it is a guide for the
novice but, to a large extent, we throw it away as our own taste forms. What
interests and involves us in maturity may be a sub-class of the music we
enjoyed and admired as we were getting to grips with the art. At that earlier
point, our taste converged with that of the consensus. More than one music-
lover has remarked to me that he doesnt now listen to the Beethoven
symphonies with the same enthusiasm that he did as a teenager.

136
 VALUE

The consensus among connoisseurs

We cannot give any usefully precise and evaluatively neutral features that,
satisfied, make a piece of music meritorious. The ideal observer theory
does not help. How, then, do we justify verdicts? Well, we have already
anticipated a plausible account in our criticism of the ideal observer
theory. Most philosophers who have thought at length about this have
concluded that the answer must lie in the operation of a consensus of the
qualified. The roots of this approach are to be found in Humes already
mentioned, much discussed essay, Of the Standard of Taste. Hume sets
out the requirements for a good critic. A consensus theory allows for
disagreements among well-qualified critics: those who, in Humes words,
have strong sense, united to delicate sentiment, improved by practice,
perfected by comparison, and cleared of all prejudice. Of these, he says,
the joint verdict is the true standard of taste. (The phrase standard of
taste is not helpful here; Hume never says precisely what he means by
this obscure term.)
The general idea among defenders of a consensus theory is that, even if
a well-qualified listener has an antipathy for early Louis Armstrong, the
majority of those who have expended time and trouble on this music and
are experienced in the style, agree that it is of surpassing quality. A few
dissenting voices among them do not seriously damage the authority of
the judgement. Frank Sibley felicitously described it as more a concen-
trated scatter than convergence on a point.7 Thus some good critics think
that Armstrong is the ne plus ultra whereas others give Jelly Roll Morton
the palm. Some think West End Blues was Armstrongs finest
achievement while others rate Potato Head Blues more highly, and so it
goes on. But within the elite whose opinion counts, there is a weighted
verdict (a majority allowing for blind spots, differing tastes, differing
degrees of engagement with the music and so on) that makes Armstrong
the major figure in classical Dixieland jazz. This domination of a
consensus of connoisseurs was described by the philosopher Samuel
Alexander as a conspiracy of the qualified.8 We need, too, to remember
that people rarely spend hours and years in listening to, comparing, and
analysing a body of music unless they find it of very great interest. When
we speak of the test of time as the examination that outstanding works
pass and mediocre works fail, we reflect the way that great works reward
many different players and listeners over a long period. That such a
consensus forms, that a body of qualified listeners constitutes itself, is
already a testimony to the value of the music. Since we cannot specify the
features that make a piece of music fine, the consensus gives us the next

137
PHILOSOPHY OF MUSIC

best thing: the basis for a distinction between the true verdict and
idiosyncratic likes and dislikes.
Unfortunately, the operation of a consensus of experienced and
attentive listeners cannot be the whole story. A consensus theory alone is
an anti-realist account. It does not tell us that there are features in the
music that bring about a response from the qualified listener. A question
remains for anybody with realist inclinations; what is it about fine music
that attracts, interests and moves the music-lover? There surely is a causal
element as well. We must foreclose the possibility that musical merit is
either a delusion of the cognoscenti or, as the philistines of the British
tabloid press propose, a pretence concocted to preserve social distinctions,
to exclude those who are not of the educated middle class. Those of us
who love music know that our response to it is neither a delusion nor a
pretence. But enmeshed in the problem of causality is a thorny issue.

The problem of irreconcilable differences of judgement

For both ideal observer theories and for consensus theories, a problem
remains and the problem is that of irreconcilable divergences of opinion
even among the qualified.9 In order to locate this question, some back-
ground discussion is required. Consider, first, a feature of the arts that
many people find worrying; it has certainly worried me on and off for
most of my life. Faced with, say, a style or school of painting or an
individual artist whose work, in Wittgensteins phrase I cant see into,
what sort of judgement do I make? In the other arts, I might, for example,
be unable to see past the frivolity, as it seems to me, of Fragonard or, like
Samuel Johnson, find the metaphysical style of the poetry of Donne and
his successors strained and artificial. The problem has been particularly
pressing for me in music; by most peoples standards, I have a very catho-
lic taste but there is music I have been unable to make much of, ranging
from Protin and Guillaume de Machaut on the one hand to the second
Viennese school on the other, and also including Miles Davis and John
Coltrane, Harrison Birtwistle, Pierre Boulez and Massive Attack. A good
example that has struck some other writers on this topic is Tchaikovskys
Pathtique Symphony, which is variously viewed as maudlin or power-
ful. Most of you will have similar gaps in your personal taste that worry
you to a greater or lesser extent. As a young man, I was told that familiar-
ity with the music of Schoenberg, Alban Berg and Anton Webern was
required in order to appreciate the cutting edge of new music. Not wanting
to convict myself of a provincial or conservative taste, I duly bought vinyl

138
 VALUE

records of the Variations for Orchestra and Pierrot Lunaire, taped the
string quartets and listened to them over and over again. But the penny
never dropped. I grew to love Stravinsky but I never cared whether I heard
anything by Schoenberg again. As he himself remarked with anguish, we
do not go to sleep with his music echoing through our minds and wake up
longing to hear it again. Or at least I dont, and apparently Schoenberg
himself did not. I was highly motivated to find memorable themes,
entrancing textures, moments of climax and all that excitement that made
a life without Bach, Haydn or Schubert unthinkable. I just didnt find
them. What could account for this misfortune? There are, of course, other
examples. There is still dissent over the status of Berlioz and many find
the operas of Handel bland.
There are various explanations of my failure to appreciate the music of
these composers.

I have not done enough work on the music. Some of this music is not
for the lazy listener. I need to listen more carefully, perhaps with a
score. But then, you ask, how much work is needed? If a couple of
concentrated hearings do not produce a passage of interest, then
mightnt we as well give up? The fact is that very few works offer
rewards to you only after you have given them a decent try. The adage
try and try again is of limited applicability in the arts.
Connected with the first explanation, I am listening to the music in
the wrong way. My expectations are not the proper ones for this sort
of music. I need to approach it with a different gestalt. Its no good
listening to later Tippett in the expectation of the sort of linear
development we find in classical music or listening to Handel with
ears attuned to the virtues of Mozart.
It is simply not to my taste. We all have blind spots. This is no more
problematic than the fact that many people dont much like the operas
of Wagner or that they have idiosyncratic likes and dislikes in the
other arts.
The music really is not all it is cracked up to be. This might be a
tenable view even if many experts judge the music to be of high
quality. For their putative misjudgements can be accounted for by
pointing out that they deceive themselves; after all, there is some
kudos in being members of the cognoscenti and, particularly, in being
the first to recognize a major new figure. Critics thus have a motive
for liking certain music and this can distort judgement. Of course, we
would expect that the negative judgement on the music would be
borne out by the way the consensus develops over time. Eventually

139
PHILOSOPHY OF MUSIC

one realizes, as perhaps people now realize, that Schoenberg is not a

major composer. He is rather what Tovey used to call an IHF: an
interesting historical figure. Perhaps, too, they confuse the interest of
technical proficiency with the more general appeal of expressive
power. A developing consensus would enable us to distinguish
between music of real power and stature and music that is merely
technically accomplished.

Note that the first three of these are cases where we shall consider
discounting our initial adverse reaction to a work of art; we discount
where we set aside our response as grounds for valuing or disvaluing the
work. In these cases we might grant that the work has stature even though
it does not appeal to us, for we know that competent judges rate the music
highly. This is a familiar scenario in the philosophy of art, although not in
aesthetics more generally. (I am not sure whether I dont like this
landscape although I acknowledge that it is really beautiful is a possible
judgement, although there will be argument about this, I suspect.)
Consider the third explanation. It is just a difference of taste. How
significant are such differences of taste? Now the first two explanations
are explicable in terms of an epistemological failing. This can be put in
various ways but each way we put it assumes a connection between under-
standing and valuing. I do not find these works interesting, I do not enjoy
them or I do not follow them because of various interfering factors. This
connection between understanding and valuing is, of course, one of the
features that separates the arts from other cases where our aesthetic
sensitivities are involved. It might be argued, although there may be
exceptions, that generally there is nothing to be understood about a
landscape that is relevant to its aesthetic appreciation. Only if you agree
with those who have recently made a case for aesthetic appreciation being
enhanced or dimmed by knowledge of the geology or history of the area
could that sort of understanding be relevant. Equally, your enjoyment of a
ripe peach seems to owe nothing to understanding, although perhaps the
situation would be different in a Michelin starred restaurant where your
gustatory experience might be enhanced by an understanding of the art of
the chef. But, in various ways, understanding is relevant to artistic
appreciation, and this underlies the centrality of interpretation for the arts.
The Orson Welles film Touch of Evil is generally regarded as a master-
piece. You might find it hammy and clich-ridden. Now this might be a
case of the second explanation. For what seem clichs can be a con-
sequence of foregrounding the wrong sorts of things in a work. For the
film-goer who does not find these to be clichs somehow sees them as part

140
 VALUE

of the quotidian style and sees the way they are used creatively. A super-
ficial listener might see clichs in Mozarts music; the informed listener
sees other things. Re-education may help us to see what is there.
Such a case may be described as one where lack of the relevant
knowledge stands in the way of my appreciation, of my seeing what is
there to be valued. A listener may lack an adequate awareness of the
conventions of the galant style and thus fail to notice how Mozart uses
them in a rather different way. A way of bringing out the epistemological
undertow here would be to put it this way. If every case of a difference of
taste were epistemologically based, then where two of us differ in our
responses to a work of art, those differences could be reconciled by an
appropriate re-education process for one or other of us. This is a line of
argument that naturally appeals to realists about artistic verdicts. Of
course, and this is important in the arts, one of us can certainly allow that
something is masterly while not liking it. If I allow that, then there is no
pressing need for resolution; we might consign the differences to broad
differences in values and not be concerned enough to do anything about it.
The only reason why we should bring our responses into line is because
one of us is missing something important and valuable and this, of course,
is the predominant reason why the matter worried us in the first place. We
want to appreciate what we are told has merit (when we are told on good
authority) and the importance of that ought not to be underestimated.
Suppose, though, that no re-education works. I try but fail to respond to
Schoenbergs Variations for Orchestra. There might be a deeper basis for
this. It might be that we have here a genuine case of the third explanation.
It is not that I need to work harder at the music or acquire more informa-
tion, nor that the music is of poor quality. I simply have a blind spot as far
as Schoenberg is concerned. One explanation of this might be that my
dislike of Schoenberg connects with broader aspects of my taste and to
like this work would require such seismic changes in my taste that I
cannot contemplate it. My dislike of Schoenberg might relate to a sense of
a certain academic stuffiness, a lack of rhythmical verve and a dislike of
that somewhat opaque orchestral sound. (Of course, I may be wrong about
all this. Then make allowances for my deficiencies of taste and find
another example to suit. Remember that the argument at this point
requires a major figure over whom there is dispute.) On the other hand, I
admire Schoenbergs capacity to build huge structures that, despite their
size, are coherent. In all these matters, to an extent, ones acceptance or
rejection of these as grounds for valuing may come and go. We must allow
both that our tastes change over time and that they sometimes oscillate in a
disconcerting way. You may find yourself admiring a musician you have

141
PHILOSOPHY OF MUSIC

not previously cared for. Then the temptation to put down your absorption
in the music to a temporary lack of critical nous or even to deny that you
are really fascinated by it may be quite strong.
What I have been concentrating on is a case where values differ
between listeners and where the difference in values determines a differ-
ence in the judgement of individual works. In as much as you can explain
why you dont like a piece, then differences of assessment will fall into
this category. So my dislike of Schoenberg connects with my dislike of
some aspects of the German musical tradition and that is a fairly broad
feature of my taste in music. (It is important to bear in mind what I said
earlier: we may be dealing here with judgements about the merits of the
work, mere personal likes or dislikes or any one of the many complex
ways in which we relate to art.) A case that I find much more puzzling is
the case of Berlioz, because I cannot fathom why others do not love his
music as I do. How can anybody fail to acknowledge that Act IV of Les
Troyens is one of the most astonishing things in all opera? Berlioz does
divide listeners. There are many, including the thousands who made it
difficult to get tickets for the Berlioz Odyssey in London in 2000, for
whom he is unambiguously one of the greatest of masters. There are
others for whom his music is eccentric, over-the-top, self-indulgent and
harmonically bizarre. In some cases the latter judgements can be
explained in the ways previously described as cases of the first and second
explanations. A solidly German background in music might make it hard
to accept some of his harmonic progressions, which are, uniquely in a
great composer, not based on competence as a keyboard player. The
seemingly strange fact that he is relatively unappreciated in France might
be explained by his passion for Shakespeare, which led, in turn, to an
unacceptably unclassical handling of form.
If our tastes are idiosyncratic in these fairly general ways, then diver-
gences are to be expected; we simply cannot love everything equally and
some things we cannot love at all. We are not Wildean auctioneers who
admire everything equally, proportionate to its merits. We can link this
with the way we might, as we sometimes do, try to offer reasons why we
like or dislike a work. Then to take a feature or features as a reason for
appreciating one work might render problematic the appreciating of
another that lacks it. The idea here is that tastes differ where one or other
of us has a stock of features, a repertoire of considerations that we value
and look for in the arts. It applies quite generally. There are poets,
novelists, dramatists, film-makers and painters who exhibit these. There
are some who do not. A catholic taste may well admire some of the latter
or, more often, some works of the latter. Thus Vaughan Williams did not

142
 VALUE

much like the music of Beethoven but he did admire the Ninth Symphony
and I have heard The Valkyrie described as the Wagner for people who
dislike Wagner. But even a fairly catholic palate wont embrace all the
works of a creator who is, by these tokens, marginal to its taste. If a visual
model helps here, think of tastes as an arrangement of overlapping circles.
Values, as Isaiah Berlin impressed on us, may not be commensurable.10
Tastes, I suggest, may differ and there may be no common measure on the
basis of which some can be judged deficient and others not.
It is worth pointing out that individual judgements of value may agree
while judgements of their basis may differ. We might be puzzled by
Schumanns judgement that Mozarts Fortieth Symphony is the epitome
of Grecian lightness and grace, yet he agrees with us about its pre-
eminence. Many will concur in thinking that the expressive judgement is
wrong. There is an atavistic element of Sturm und Drang11 in Mozarts
mature G minor works that passed Schumann by, although he said
different things about Mozart at different times, but we all agree with him
on its stature. Such differences may be replicated in performance. Most
conductors treat Haydns Military Symphony as genial and high-
spirited; the soldiers are toy soldiers. But turn to Nikolaus Harnoncourt
and you find the symphony treated as expressing menace and disquiet; it
then becomes natural to take it as a protest against the horrors of war.
Defending such judgements turns one towards those issues of authenticity
that we discussed in a previous chapter; you may claim that there is a
proper mode of performance and, at the same time, an informed stance
from which the work is properly assessed. My point here is that diver-
gence in judgements may not only involve the evaluative; they may
involve expressive predicates as well. There may be agreement about
evaluation but irreconcilable difference of opinion about the expressive
nature. I can easily imagine Beecham and Harnoncourt agreeing on the
superlative quality of the Military Symphony but disagreeing about the
expressive features and I can imagine that such a disagreement might not
be settled even if a newly discovered letter from Haydn made it clear
which side he belonged to. For it is open to a performer to insist that the
composers obiter dictum is not the last word.
Recognizing differences of taste often neutralizes differences of
opinion, of course. Once we see that our tastes differ we are inclined not to
argue further. One reason for this may be that argument is no longer
thought very worthwhile, although where I am still under the thrall of a
film seen, a novel read or a piece of music heard, I may not be so easily
satisfied with the nostrum Its just a matter of taste. To the extent that it
matters to me and to the extent that I believe that the judgement I have

143
PHILOSOPHY OF MUSIC

made is not merely a consequence of my particular biography, my defence

of the work may be robust and I shall regret the inability of my interlocu-
tor to appreciate it.
Here we touch on the residual cases: those cases where a divergence
of taste persists despite our best endeavours to come to some agreement
over the stature of a writer, painter or composer. The strongest of these
residual cases is where a discrepancy of judgement is not explicable in
terms of varying repertoires of values in the arts. I dont like this, you do,
and theres an end to it. Our tastes do not differ but our judgements about
this case do. It is these cases that I find most interesting and puzzling.
There are disputes about a works status that do not seem to be explicable
in any of the ways I have described, where not only is there no reason to
suspect that there exists such an epistemological explanation of how the
differences arise but also the difference is not necessarily grounded in the
repertoire of values. Typically these cases, and there may not be many
such, occur where one person has an overwhelming sense of the power of
the work, a power that allows no denial, while another simply does not
respond in this way.
Equally, of course, I may be struck by the ineptness of a work where
others admire it and where I admire other works by the same creator. Thus,
recently, I have had occasion to listen carefully to several different record-
ings and a live performance of Vaughan Williamss A London Symphony. I
have come to the conclusion that, a few episodes apart, it is a terrible work
and I cannot credit how it comes to be programmed and recorded so often.
Obviously others disagree; some readers will; but my initial judgement of
its poverty and crudeness bears in upon me just as strongly as my reactive
judgement to Berliozs Requiem is that this is awe-inspiring stuff. This
coexists with my belief that some of Vaughan Williamss works are as near
perfect as we are likely to encounter this side of Mozart. Of course, I might
acknowledge that my judgement that A London Symphony is a poor work
is idiosyncratic and that the consensual view that it is a masterpiece is
right; but in this case, I am very reluctant to do what I do elsewhere and set
aside my own judgement. This is not a case where the various features I
admire in his music happen to be absent in this particular case. It is rather
that this particular work offends me.
So the really problematic case comes where, while two people agree in
taste, they diverge over the value of a single work that may or may not be
an exemplary case of the style that both appreciate. But there is another and
reverse case; this is where somebody feels that a work, that others do not
value highly at all is one of those works of supreme power that everybody
ought to recognize, regardless of taste. Here agreement or disagreement in

144
 VALUE

taste is not a relevant issue. So Beethovens Ninth Symphony, over which

there is some dispute, is held by many to be a supreme masterpiece and
when Vaughan Williams acknowledges it,12 despite the fact that he is not
very partial to Beethoven on the whole, in their view he properly salutes a
work that transcends differences of taste. Of course, the clientele in
question is assumed to be generally knowledgeable about classical music.
The phenomenology of all these cases is, I believe, much like the way in
which I find it impossible to deny that there is a table in front of me as I
work. Similarly, I believe, there are a few works of art whose status is
disputed but whose power strikes its aficionados so intensely that they feel
we must acknowledge it and will brook no dissent.
I take it that two listeners differing over Vaughan Williamss A London
Symphony may both have listened carefully and without prejudice. I have
said that what is important in this example is that they might share the
same values in music; they like the same sorts of thing but they differ in
their judgement of this work. It might, of course, sometimes be difficult to
establish that they have the same values; the natural way to do this would
be by seeing not only what repertoire they prefer but what considerations
they adduce in defending a particular verdict, what features and what pas-
sages they point to. We would be justified in assuming that more general
aspects of taste did not enter the picture if we found that they did not refer
to such stylistic factors defending their differing conclusions.
Could we say that both judges are right? Well anybody who is a robust
realist and thinks that the value of the work is something independent of
particular judgements of it cannot entertain that conclusion. Those who do
not espouse strong realism have an option. They might say that both are
right in the sense that procedurally both have followed the right path prior
to making a judgement. They have both paid careful attention to the work;
they have both listened to different performances in different contexts;
they have both tried to prepare for a listening by hearing other related
works by the same composer or by those who have influenced him. If they
have done all this, then you might say that both are right. If this seems
rather odd, consider a parallel case in ethics. Two women in relevantly
similar circumstances come to different conclusions as to whether to have
an abortion. Suppose they have both been as honest as they can, neither
giving too much weight to their own futures nor too little; they have both
taken advice and thought hard and deeply about the complex issue. I
certainly feel a strong pull to the conclusion that both are right simply
because it is hard to imagine in this case what rightness could amount to
other than the serious and careful attention to all aspects of the situation.13
So why should not the same hold in the case of my musical example?

145
PHILOSOPHY OF MUSIC

If we do think that correctness of judgement is a matter of following

the right procedure, then we will draw the conclusion that in such cases
not only is there no agreement as to what the objective status of the
work is; there is, in one clear sense, no objective independent status to
be determined. It does not exist. There is no right judgement prior to the
enquiry. However, if the rightness of the judgement is merely a matter of
the integrity of the assessor, then all judgements in this category will be
satisfactory. But clearly this is not the case. If you judge that a piece of
music that lies outside your taste is poor in quality, you will, if you are
sensible about judgements in the arts, allow that the consensus has the last
word. You may then judge that it is good even though you cannot see what
is good about it. The notion that correctness in judgement is a matter of
procedure can only apply to a tiny sub-class of judgements about the arts;
indeed, I suspect that it applies to a tiny number of moral judgements,
even though these may be peculiarly important moral judgements. They
may even be those that are permanently disputable.
Do such cases of unresolvable divergence of judgement occur more
frequently with respect to music than the other arts? I suspect they do,
which makes a discussion of this issue more pertinent to a discussion of
the particular nature of value judgements about music. Let me explain. A
clash of opinion over a novel, a film or a play may often be explained in
terms of differing religious or political standpoints; this is not an explana-
tion so often encountered in differences over music. Thus an atheist might
find the element of religious commitment in Dostoevsky or Gerard
Manley Hopkins rebarbative. You might find the whole genre of American
westerns unattractive because of an assumption that there is no problem so
complicated that it cannot be solved by shooting somebody. (Of course,
some westerns are more subtle or subvert the genre.) This type of explana-
tion is less readily available in music. There are some examples; they tend
to belong to music in as far as music is a mixed medium. Some find much
of the music of Messiaen objectionable because of the atmosphere of
incense, of religious kitsch and of an apparent certitude about dogma that,
in the twentieth century, seems like a pose or worse; many musicians
dislike Wagners Parsifal. (One German critic remarked that the baton
should be gently but firmly removed from the hand of anybody who
attempted to conduct it.)
Of course, there are explanations of negative judgements of pure
music that border on the ideological. The style of Richard Strauss offends
many, partly because the late style, with its classical references, its extra-
ordinary euphony and melisma, its allusions to the great tradition of
German music, seems to have been composed in a different world from

146
 VALUE

that of his fellow Germans. On the other hand, some find the combination
of naivet and a superlative technique rather endearing and what faults
there are forgivable. (I cant think of another composer to whom we
would apply the epithet forgivable.)
Admittedly, differences of opinion over musical style may connect,
closely or not so closely, with broader aspects of our personalities, not just
with our taste but with our values elsewhere. The conspicuously bourgeois
features of Strauss might not gel with radical political opinions. But such
considerations are usually more remote from the judgement of music than
they are from the judgement of film, theatre or fiction. In any case no such
explanation accounts for the sort of divergence I have outlined over
Vaughan Williamss A London Symphony. If we agree to disagree here it
is with a nagging discomfort absent in the cases where it can be explained
away in terms of broader divergences of taste, or moral or political
commitment.
Where does this leave us? The existence of divergences of this sort
leads some thinkers to say that it cannot be a matter of fact as to what
the real status of the work is, a fact that careful and competent listeners
converge upon. For that to be the case there would have to be some point
at which the listener fails, either in being prejudiced against the work or in
lacking the right sort of information or training in taste. She fails to grasp
the true status of the work and the explanation must lie in one or other of
the ways in which she is ill-prepared. But we have found examples where
the listener is well-prepared, is not guilty of any prejudice nor has any
distaste for the style and yet she fails to endorse what seems to be the con-
sensus. Now the assumption that her dissenting judgement can somehow
be explained away in the usual way looks more like a matter of faith than
anything else.
We raise again the spectre of non-realism about judgements of merit.
Fundamental disagreements become a problem for realism when they are
thought to underpin the conclusion this work is great or this is rub-
bish. One way of putting it is to say that realists who subscribe to an ideal
observer theory, believing that the ideal observer grasps certain facts
relevant to verdicts, require that ideal critics do not disagree. For anti-
realists, such divergences are less problematic. For the more robust realist,
of course, agreement alone will not guarantee that the aesthetic properties
are being seen for what they are. The ever present threat of scepticism
means that, for the realist, there is no automatic route from either the
existence of a consensus among the qualified or from the deliverances of
an ideal observer to the object really having the properties ascribed. There
may still be a gap between the evidence grasped by the ideal observer and

147
PHILOSOPHY OF MUSIC

the independent facts of the matter as far as the works merit is con-
cerned. In theory, both consensus and ideal observer could be wrong. But
the crucial challenge to the realist is the case where you feel that you
cannot but insist on the stature of this particular work and yet you know
that other competent judges disagree vehemently, and there is not much
doubt that people react in these different ways to Tchaikovskys
Pathtique Symphony.

The crux for the realist

We have, in this final chapter, been mainly concerned with trying to

understand how we can distinguish expressions of personal taste from
judgements that a piece of music really has merit over and above such
individual preferences. Lets not forget the importance of this. A teenager
is told that the late Beethoven quartets rank among the supreme achieve-
ments of mankind. She does not want to miss them; she knows that her
taste needs to be developed in order that she may grow into them. Her
education in the arts requires the assumption that there may be works of
art that are good and yet, at present, they are to her a closed book. It is the
distinction between what we like and what is good that makes education
in the arts both possible and desirable, an education that is made requisite
by the various ways in which we can fail to understand and appreciate,
some of which I listed earlier in this chapter.
So far I have mentioned various ways in which the distinction might be
understood. The first was that objective properties of works might be
linked via criteria to value judgements and that these objective criteria are
what enable us to define the real stature of a piece of music (or of a
performance, for that matter); the second was that works have a value
independent of ordinary human judgement in as much as that judgement
matches what an ideal listener might decide; and the third was that the
consensus decides. On this third account, what the verdict of the elite gives
is a decision as to the true status. But, while electing for a form of
consensus theory, I only suggested that the consensus is a means of find-
ing out what the status of the piece is. I left open the possibility that it is not
equivalent to that status. Then the merit of the piece may be something over
and above the way in which we settle on what that merit is. So a form of
realism remains possible. But what sort of realism? What makes the work
itself meritorious, distinct from the judgement of the consensus? This, I
think. Works of art have causal features that involve us or move us in the
way that is characteristic of our response to works of art. That is, we do not

148
 VALUE

project on to the work of art what moves and interests us. Nor is any
reductive programme tenable that turns the value of the work into what the
consensus declares valuable. So I incline towards a propensity theory of
artistic value. Works of stature have causal powers to produce in the
prepared mind a fascination with and absorption in the object. To quote
Hume, there are certain qualities in objects which are fitted by nature to
produce those particular feelings.14 Thus they may also produce the
frisson we value, inter alia. I construe the beauty (or whatever other artistic
features are relevant) of the object as I construe poisonous: that is a
propensity to affect human beings or other mammals in the absence of
whom it must be construed counterfactually, as philosophers say. (Poison
isnt a bad analogy; it is a matter of independent fact that arsenic harms
humans, but it also depends upon the way human beings are constituted.)
That is, were I to experience this work in the right setting and with the
appropriate attention and preparedness, then I would respond. Now this
allows a contrast between what an objects value really is and how it might
seem to an individual judge, although not, as I have said, a contrast
between its real value and the consensus of judges as to its value. Certainly
the first contrast has to be maintained in some form or other because, as I
have insisted, it is germane to everyday thinking about the arts.
But the problem with this is obvious. We have spent some time discus-
sing the existence of irresolvable divergences of opinion among the
qualified. The way I have described a causal theory does not allow for
these. For I have described the causal theory as allowing that music of
merit will have certain sorts of effects upon a prepared listener. But the
counter-examples we have been discussing are precisely those: where two
listeners are both equally prepared but where they diverge in their
responses, and, consequently, in their judgement. In fact, the causal theory
fails at an even earlier stage because it does not allow for the possibility
that I may agree that this music is fine even though it does nothing for me,
which is precisely the feature of our life with the arts that gave rise to
sophisticated forms of the consensus theory in the first place. The causal
theory may seem no more than a truism, that works of art involve and
impress us, but it does not seem to help with the problem at hand.
Possibly the sensibility of our culture might change so that greater
unanimity occurs. An authoritative critic might persuade us to see a com-
posers work in a new light, foregrounding what we previously passed
over, or the style of a period might become more accessible through
changes in the receiving culture. A savouring of irony, parody and quota-
tion made Mahler acceptable where he had been ignored. His influence
on Britten and Shostokovich might, through the acceptance of those

149
PHILOSOPHY OF MUSIC

composers, indirectly have brought about the resurgence of his music in

the 1960s. But it is hard to see how these factors could bring about the
reconciliation of divergent judgements over a single work.
Perhaps the causal theory can be no more than a truism. There may be
works that impress some judges but not others. If you consider the variety
of ways in which art affects us it delights, appals, challenges, disturbs,
provokes a revision of fundamental values and assumptions, creates an
imaginative involvement with others and so on it seems likely that any
attempt to expand the causal thesis into a form that is not simply banal
will be open-ended. We cannot predict how the arts of the future will
affect us nor how familiar masterworks will be understood. Perhaps music
is more restricted than other arts in the range of its effects but it is cer-
tainly wide enough to preclude any informative formulation of the causal
thesis.
I considered earlier the possibility that there are some works that have
no objective status; if this is so, we need not be realists or non-realists
tout court. Perhaps some judgements of value or, indeed, judgements
about the expressive character of some works, call for a realist account
while others do not. Do we have to be realists or anti-realists over the
complete range of works? Perhaps not. Of course, in some areas there
must be a measure of conformity. Thus a range of aesthetic judgements
that are not about works of art, the sort of judgements learned by small
children, this is pretty or this is ugly, must show uniformity. So a
child is told that this is a lovely tune and acquires thereby that use of the
word lovely. Such appraisals we learn in the same way as we learn other
words like cat and dog. A child who, shown a daffodil, said it was not
pretty either has not learned the term properly or is a precocious brat, an
infant Oscar Wilde.15 Such basic aesthetic judgements as the flower is
pretty, which contrast with judgements of art and are themselves to be
surmounted by developing tastes in the arts, demand a realist interpreta-
tion; they may even reflect what Wittgenstein called primitive reactions.
Wittgenstein said that the reaction to seeing that a door is too high is
analogous to taking my hand away from a hot plate.16
Still, the problem of irreconcilable judgements wont go away. It may
be true that we acquire our vocabulary of evaluation in the way I describe.
It may even be that in some uses the question of whether the object really
has the properties attributed will not arise. It remains the case that at a
more sophisticated stage, there may be irreconcilable divergences of
judgement. If music means a great deal to you, you back your taste. And
then the problems begin, because other music-lovers may passionately
disagree. The important consideration here is this. If the consensus of

150
 VALUE

judges tells us what the real status of the piece of music is, then a non-
realist will naturally equate the consensual verdict with the real status.
This move is what philosophers would call reductive. When I men-
tioned this earlier I gave as a paradigm of reduction the phenomenalist
thesis that because material bodies are not independent of their being
observed, their status amounts to no more than the possibility of our
having sensations of them. Reduction is the technical term for this and
this classical move reduces material bodies to the possibility of sensa-
tion. Likewise, the reductive move in the philosophy of artistic value will
be to reduce claims about the real value of a sonata to claims about what a
weighted or qualified judgement by the connoisseurs will offer. But now,
if this is so, the individual who maintains, against the tide, that Berliozs
Requiem is a work of outstanding greatness has to recognize, on pain of
irrationality, that he cannot be making a claim about its real status since
the consensus argues against this verdict on Berlioz. His only recourse is
to say that the consensus, over time, will come around to his way of
thinking. But he might not concur with that prognostication. He might
allow that Berlioz will always be a controversial musician. If musicians
from Donald Tovey to Pierre Boulez have denied his stature, then why
should we think that another century or two will make any difference? So
we are left with this impasse. The existence of irreconcilable divergences
of judgement seems to rule out realism because there now seems reason to
suppose that there is no independent fact of the matter on which judges
will converge. And there is no other route to a verdict except through
competent judges. Yet the alternative to realism, which is to reduce
claims about the real status to claims about the judgement of the con-
sensus, thereby preserving the important distinction between the judge-
ment of the individual and the status of the work, requires the dissenter to
allow that his claim is no longer about the real status of the work. But this
he will not do. He thinks the work is really outstanding and demands that
this judgement of his be viewed as an objective judgement.
In a way, I want to stand the realist argument on its head. If irreconcil-
able judgements do exist, then they have to be treated realistically. It is no
good saying that all this shows is that you cannot be a realist about
judgements of artistic value. For if you cannot be a realist about artistic
value you cannot present irreconcilable differences as having any proba-
tive power against realism. The problem is that a non-realist reading of the
judgements in question robs them of what distinguishes them from mere
assertions of taste.

151
PHILOSOPHY OF MUSIC

Are verdicts essentially particular?

One pervasive feature of most of the theories we have been discussing is that
they view artistic judgement as particular. The only exception, and the
theory most summarily dismissed, was the idea that the merit of music is a
matter of satisfying certain general constraints that can be formulated as
criteria. It seems that verdicts are ultimately subordinate to the individual
judgements made by competent individual listeners. There does seem to be
a divide here between moral and artistic judgements, at least given the way
most philosophers characterize moral judgements. Whereas with a moral
judgement, that this action is a case of stealing and falls under that general
heading gives a reason why it is wrong, that this piece is melancholy or in
sonata form or whatever never gives a reason for regarding it as good or bad.
It never carries a polarity. The attempt to find the real status of a work is
secondary, although its importance is undoubted. We are listeners among
other listeners; we share common human responses to music and a common
background in the musics of our society. Music would be a very different art
were discrepancies between judgements more common than they actually
are. As we have seen, in the case of most such discrepancies we have various
ways of explaining why they occur. We can see that we may be missing
what others find worthwhile. But we also recognize, although we are
sometimes surprised by, individual differences that cannot be accounted for
and this, inter alia, underlines the priority of the particular judgement.
So let us consider what is known as particularism in the judgement
of art generally and music in particular.17 Particularism has been much
discussed in ethics recently, although the intense debate over particular-
ism there seems to have had little impact on the philosophy of art, which is
surprising, although it may merely reflect the regrettable fact that
philosophers no longer keep up with developments across the board as
they did forty or fifty years ago. Philosophy has become more specialized.
Still, there are a sufficient number of general surveys to make ignorance
of developments in neighbouring areas not entirely excusable. What I
propose to do, before bringing the discussion of artistic verdicts to an end,
is to examine various forms of particularism and then to try to see how
general maxims might still play a part in musical criticism and practice.

Particularism I

First, a radical form of particularism maintains that there are no common

features uniting the various pieces of music that we call good or bad

152
 VALUE

other than that we call them such. In the same way some particularists in
ethics claim that we see straight off that some action is wicked or
immoral without first grasping the neutral or descriptive features that
make it so, for the simple reason that there are no common neutral
features underlying the application of these words. Attempting to
proceed via a grasp of these neutral features is futile. For what we cannot
do is to describe cruelty, for example, in terms that have no evaluative
overtones. If we try to say that cruelty involves ill-treating another where
there is no guilt on the victims part, we are no nearer to a neutral
specification of what is being done. The description is already replete with
words like guilt or ill-treatment. If we try to describe the action in
terms like hitting we not only ignore mental cruelty but we omit what
makes the action count as wrong. There are very considerable difficulties
in making the necessary descriptions both general and neutral. In parallel,
it is often argued that there are no general non-evaluative features that can
be cited as grounds for any judgement about the goodness or badness of a
work of art. As we saw earlier, when writers have tried to provide general
characteristics, citing qualities like unity, variety or intensity, they do so in
such very general terms that we are not clear as to what will count as unity,
variety or intensity. In any case, it is certainly possible to find works that
have these characteristics but are not very good. Too much unity is
monotony, for instance. A more positive case for particularism may use
the notion of salience or foregrounding here. I have argued elsewhere
that this is something of great significance in the arts.18 Your grasp of a
work usually begins with something in it impressing you. What you
immediately find salient then may be central to your fuller and mature
grasp of the work, although it may also take a lower profile, as they say, in
the light of a more complete understanding.
Another way of putting this version of particularism is to suggest that
there is no shape to the underlying class of musical judgements. But in at
least one form this claim is difficult to sustain. What there is is some form
of commonality in judgement. The business of concert-going may well
reinforce this. Our listener finds herself swept along with the enthusiasm
of the rest of the audience. (When I last heard Richard Strausss Four Last
Songs in the concert hall, I was touched by the fact that tears were running
down the cheeks of the person sitting next to me.) Although the divergent
cases of irreconcilable differences of judgement may not be in any way
marginal to her taste they matter intensely to her generally her judge-
ments do coincide with those of the majority.
What might seem to back particularism is the argument that art can
strike us with considerable force before any sort of reflective critical

153
PHILOSOPHY OF MUSIC

reasoning plays a part. Even if the music did have merit because of
general features that are criterion-bound, that would play no initial part in
the judgement I make. The paradigm cases of valuing involve those
works of music that strike the prepared mind with irresistible force (or
indeed, the unprepared, although these cases are of lesser significance).
As I argued in the previous section of this chapter, a work of art may so
impress me that no countervailing reasons you offer will make me budge.
You may point out that it is derivative or that the effects are cheap or that
the dramatist loads the dice in order to increase the sense of tragedy. I may
not be fazed.
Characteristically in the arts, the verdict precedes the deliberation. I am
seduced by La Traviata and I only resort to analysing it if I want to
persuade somebody else to listen to it or if my judgement is challenged. I
might then look at the ending and think that it is sentimental in the way
Dickens can be sentimental. Then I might change my mind about it. I
might, too, be persuaded of the merit of a work that does not appeal to me
because a number of judges whom I respect form a consensus about its
merit. Note that, in this case, I am not argued into appreciating it but only
argued into recognizing its stature. My recognition of what my peers think
about this music leads me to a judgement about the merits of the work
itself. I dont say that I am never argued into appreciation, but where this
does happen it happens because my attention is drawn to features that I
then see to be powerful. I foreground certain sections. I may find those
aspects of the work that strike others now strike me.
Of course, there might be judges who get the decision right (we know
because it matches the consensus of the informed) without being able to
explain why: without being able to point out the significantly relevant
features of the work or its performance. I think that such an idiot savant
is far more likely in, say, music but I would not rule out people having a
sort of natural good taste in the other arts. We sometimes say I dont see
how it works but it does. Here the task of the great critic may very well be
to articulate what strikes him about a work, and this is a very distinctive
skill.
Lets take an example. Consider Benny Goodmans AC-DC Current,
recorded in 1939. How would you talk about this piece in an endeavour to
help somebody new to swing? For a start, Charlie Christians opening
arpeggio-like figure on the guitar is constantly used at various points as a
sort of punctuation; in its earlier occurrences it is often in a key remote
from the main key of the piece, notably immediately preceding
Goodmans hiccuping intervention (just before the cry from somebody of
Hi Ho, Bernstein!); the main key of G is then resumed without warning

154
 VALUE

or preparation. It is difficult to find highlights in this masterly piece, as

perfect as a Mozart quartet movement, although Goodmans entry is
certainly riveting enough as is his long-held note a little later, reminiscent
of the long-held clarinet note in Doctor Jazz (just before Jelly Roll
Mortons vocal). Perhaps the best you can do is to provide the new listener
with a mental map so that she can keep track of what is happening. Once
done, you think, she surely cannot fail to fall in love with it. Essential in
this is a sort of interplay between attention to the detail, seeing through
that how the piece works, and the overall judgement.

Particularism II

You may resist the argument so far on quite general grounds; you may
argue that drawing parallels between morality and the arts is misguided.
Ethical particularists are largely concerned with actions, with what it is
right to do in a given circumstance. But judgements about aesthetic value,
you may say, do not involve deciding to act; so a central issue in the
philosophy of morality, that of providing a bridge from reasoning to
action, has no analogue. But this view of the aesthetic as contemplative
is somewhat parochial, certainly as far as music is concerned. First, we
must recall that we are concerned with the artistic rather than the merely
aesthetic and that, for a good deal of its history, performance was central
to the artistic life in dance, recitation or the performing of music. The
player does have to decide how something is going to be played or the
actor how lines are to be spoken and even the non-participating member of
the audience will be better equipped if she is aware of the constraints. Up
to a point it is a matter of being right or wrong. If an inner part carries the
tune, the pianist had better be aware of it. But beyond that there are
matters on which there will, of course, be a difference of opinion among
the qualified and this may be, as we say, a matter of interpretation. But
active involvement is not entirely restricted to players; the reader of poetry
or prose, too, pays attention; this or that passage appears salient and he
may want to recommend or describe the poem or novel to somebody else
in the ways we have been discussing.
It is certainly more often the case that our appreciation of the aesthetic
is more completely contemplative and that the will is not involved but that
this is not universally the case is shown by considering examples such as
choosing a shirt, a tie to go with a suit or an arrangement of flowers in
gardening or internal decor. These are all matters aesthetic rather than
artistic in nature but action is called for.

155
PHILOSOPHY OF MUSIC

Particularism III

I have suggested that we characteristically first judge a piece to be grand

or mediocre and then try to find grounds for this being so. The grounds
tend to be ex post facto and particular to the case. Although this is often
the case in the arts, and my examples so far reflect this assumption, we
need again to remind ourselves that music is a performing art. This prior-
ity does not seem to hold in performance, for here we commonly have to
make decisions about how to play, speak lines or dance a particular
passage prior to embarking on it. We have first to find grounds to do this
dance, this speech or this piece of music in a particular way and we are, of
course, looking for the right or, at least, an illuminating way. More
importantly, these grounds may be quite general. Thus the pianist is told
by her teacher that it is permissible to pedal through staccatos when
playing Beethoven and Schubert; this is a general principle, although one
that may have exceptions. It is more of a negative than a positive principle.
Normally one does not pedal in staccato passages because, of course, the
staccato effect is lost thereby; nevertheless there are passages in some
composers where it is effective and proper. So, before she starts to
practise, she may read through the music, noting problematic passages as
far as fingering is concerned and, at this stage, make some preliminary
decisions about pedalling. Of course, these initial decisions may be
revised when she comes to play it and, ultimately, it is a question of this
sounds right here, although only the experienced, informed and aware
performer or listener is in a position to make this judgement.
I have stated these principles about pedalling in a fairly general way
and they could, of course, be relativized to the music of a particular
period. Should, then, the particularist acknowledge that there are gener-
alizations that are important in the arts? After all, if, as is generally agreed,
long experience and general familiarity are crucial for artistic judgement,
what form could this take if it is not general? If this is so, then it ought to
be capable of being summarized in the form of maxims. Anybody who has
had advanced lessons in creative writing or the performing arts or in
painting, knows that the good teacher alerts you to the crucial points. At
such points there may be something generalizable: you dont do that.
But it may well be that there are exceptional cases where it does work.
Take another example. Consider the common interpretative point that a
diminuendo is better than a ritardando at the end of a Mozart slow
movement whereas slowing down is legitimate at the end of a Beethoven or
Schubert slow movement or even, arguably, in the last bar or so of a Bach
slow movement. But we are told by all sides that rules have to be applied.

156
 VALUE

In this context this can only mean one of two things: only sometimes is it
permissible to end a Mozart slow movement with a ritardando; or we have
to decide just how much of a diminuendo is called for. So, as far as the first
is concerned, you might then ask which Mozart slow movements can be
ended this way: those that seem heavier and more romantic, more an
anticipation of Beethoven? Once we get to this stage is there any other
answer than Frank Sibleys, that we look and see?19 Our decision on this
might not be arbitrary but it might not be open to any further explanation
or grounding, if, indeed, explanation is the proper word here.
Suppose we say that there are generalizations but they are extremely
complex and have to be qualified. We have rules of thumb that we apply in
the broad range of central cases and these work. One problem about the
arts is the fact that rules of thumb might just be useful in the middle range
of more or less competent pieces but useless when we are dealing with
finer achievements. So the great pianist may get away with pedalling
where I might not or with a ritardando where it would be crazy for me to
try it, just as the great composer can overrule the conventions of harmony
and counterpoint that the student ignores at his peril. And isnt this just
our experience with great musicians? Wilhelm Furtwnglers wayward
way with tempi and his uniquely flexible ensemble produces, in Bruckner
or Beethoven, an effect of mystery and profundity. In the hands of ordin-
ary musicians, this is a recipe for disaster and they had best not attempt it.
Once again, our recognition of the particular majesty of Furtwnglers
musicianship is the particular judgement in the light of which talk of
general rules or maxims becomes subsidiary. Their range and applicabil-
ity is only to the middling competent.
But suppose, you will say, a pianist copies Solomon or a conductor
copies Furtwngler. This is possible. The teenage Glenn Gould copied
Artur Schnabels interpretation of Beethovens Fourth Piano Concerto in
one of his earliest concerts, only to have a local music critic snort Who
does the kid think he is, Schnabel?20 Here, of course, familiar considera-
tions enter. This is a token of somebody elses interpretation where the
interpretation is the type. The young Gould no more interpreted the music
than a copyist composes when he writes down the notes of an existing
work.

Particularism IV

Particularists may allow that there are, indeed, general considerations that
go to make a work of art good or bad, but these are multifarious; they may

157
PHILOSOPHY OF MUSIC

include unity and elegance, for example, but we cannot rank them in any
order. In any particular case, then, we have to judge how they apply in the
given situation and whether they are relevant considerations. There seems
much to be said for this, for after all unity may be often a virtue in a move-
ment and thematic integration is something we admire, but we would not
wish Mahlers symphonies, in which this is less conspicuous, to be more
like those of Sibelius, in which the logical development is so striking.

Particularism V

One strong form of ethical particularism maintains that the very same
factors may count in favour in some circumstances and against in others.
That it produces pleasure may normally be a reason in favour of an action;
that the pleasure is sadistic makes the pleasure a reason against another
action. Indeed, although I dont think huntsmen are necessarily sadistic,
that they take pleasure in hunting a fox to death is thought by the anti-hunt
lobby to be a reason against hunting; however that they take pleasure in it
is a reason for people playing rugby. Now, of course, the universalist who
believes that there are general rules that cover morality will make the
obvious reply; qualify pleasure as sadistic pleasure and we have no
problem in seeing that this is never a good reason for doing something.
Innocent pleasure is. The case for claiming that a factor may have one
polarity in one context and another in another is much stronger in the
arts. You cannot specify generally and in advance the situations in music
where a device works as opposed to those situations in which it does not
(and we return to the earlier and generally accepted point that there are no
criteria for merit in the arts). This claim is not controversial in the arts.
However, for the moral philosopher there are certainly problems in
rejecting what have been called switching arguments. You use a switch-
ing argument when you claim that a consideration that operates in one
case ought to operate in a like case. An essential part of moral reasoning
involves pointing out special pleading. For example, imagine that a
wealthy Western country was adamant that there should be no relaxation
of patent rights to favour African countries that could not afford anti-HIV
drugs; if they could not pay the rates demanded by drug companies, then
tough. If then, when it came to relaxing patent rights because the
inhabitants of this wealthy country had insufficient medicines against
anthrax, the drug companies were immediately persuaded by its
government that they should not stand upon their rights in the face of the
greater need of the people, this would be rightly damned as double

158
 VALUE

standards; morality sometimes demands fairness. To reject switching

arguments is to emasculate this type of moral objection.
But there is nothing like this in the arts. Special pleading is no
objection to recommending a work of art and this, surely, indicates some-
thing of very considerable importance. Praising and recommending a
work of art is always special pleading. So, whatever the case in ethics,
switching arguments evidently do not work in the arts. The very same
device may both explain why one poem works and why another fails to
work. Criticism in the arts is far more specific and more completely
holistic. We are interested in how this image works in this particular poem
or why this harmonic change is so effective here where it is banal
elsewhere. Indeed, it is true in the arts that pretty well every consideration
is capable of having its polarity reversed by a change in context. Playing
Schubert well requires an exact calculation of the moment of hesitation
that precedes a change of phrase. Many competent players seem to lack
this sense. It is akin to the slight hesitation required in playing the three-
in-a-bar rhythm of a Viennese waltz. Played in strict tempo the music
loses its lilt, but even where an eighteenth-century composer produces
music in 3/4 time, as Handel does in his Water Music, it would be a gross
solecism to attempt a Viennese lilt to the rhythm.

Particularism VI

It has been claimed that if the particularist is right and there were no
patterns in evaluation, such that we can learn the application of good or
bad the way we do elsewhere, then the judgements of artistic value have
to be expressive, involving something about our responses. Some
logical positivists claimed that moral judgements were merely expres-
sions of their authors approval or disapproval (which has been called the
boohurrah theory of ethics). The suggestion in parallel then is that
particularists ought to acknowledge that artistic judgements are not
cognitive but expressive. They reflect our reaction to the work first and
foremost and not features about the work itself, save derivatively. Thus
one version of what I shall call an expressivist account of a judgement
of a piece of music as exhilarating would base the rightness of the use of
that description on the fact that the speaker is herself exhilarated by the
music. Likewise, if the music is described by her as moving, she is moved
by it. Both these are theories I considered in Chapter 3.
All this needs unwrapping. So far, I have expounded expressivist
accounts as though they were about first-person judgements, similar to I

159
PHILOSOPHY OF MUSIC

am in pain, perhaps. When Wittgenstein suggested that these judgements

might be like taking my hand away from a hot plate, he seemed to have
this analogue in mind. I observed earlier that there is a certain sort of
commonality in our judgements of works of art. There are patterns in
reactions in as much as there is art that appeals across societies, sub-
cultures and tastes and across generations. A child can see, perhaps with
puzzlement, that somebody is moved by a piece of music or a play or a
poem. She sees that many are moved at a performance of Messiah. She
learns thereby that terms like wonderful, moving, great or over-
whelming are used to apply to that work, but she may not necessarily be
moved herself. If a response is idiosyncratic, she will discover this in time.
Then, by listening to what adults say or by reading, she finds that there is a
canon of works of art that most people admire. It is not hard to see how
such words expressing valuations can be picked up and, once we see this,
the enquiry as to whether value-judgements in aesthetics are expressivist
or cognitive looks like a false antithesis. They may be both on different
occasions, and, indeed, their expressivist use is usually consequent to their
being picked up in the way I suggested, even though they might be learned
through the experience of the child herself. I can imagine a child being
flushed and excited and her father saying I can see that you found the
music wonderful; she thereby learns this application of the word won-
derful partly through her own experience but with the help of another
persons usage. But more often, I suspect, a child learns to use these words
initially in contexts where no experience of her own is involved. Since she
may begin by learning to use the words to describe a work of art that
moves adults but passes her by, when she uses these words it is not
expressivist, if by expressivism is meant that their use stands to the
childs experience just as a cry of pain or an expression of delight does.
Of course, once acquired, they may be used expressively. But the fact
that they can be used in early usages to record verdicts or other aesthetic
assessments that are not reactive shows that there is a non-expressivist
usage. The fact is that the use of these appraisals is complicated and it
takes some time to learn them thoroughly. They are certainly not generally
purely expressive, for if they were, the distinction, which I have insisted is
so crucial in the arts, between thinking something good and liking it,
would disappear. This strikes me as the crucial rejoinder to an expressivist
treatment. Perhaps it is the case that there is a primitive use of these words
as interjections as Wittgenstein thought, but the mastery of these words
goes well beyond this.21

160
 VALUE

The concept of interest

So far we have concentrated on the single judgement and, by implication,

the response to a single work. We need to redress this by bringing out an
aspect of our engagement with the arts that is often neglected. A passion
for the music of a composer, a performer or a period raises new issues.
Important here is the way that pleasure gives way to interest.22 Having
heard a piece by a particular musician and enjoyed it, you seek out others;
perhaps you buy an anthology or a disc of greatest hits, hoping that
other music will appeal. This gradually turns into something different: an
interest. For education in the arts is typically a progress from pleasure to
interest. Certainly ones first interest in the arts may come from the sheer
pleasure in wordplay, in a rhythm or melody or in a patch of colour. I shall
give an example of the first, Harry Champions music hall song, Im
Henery the Eighth, I Am, although I suspect that readers wont thank me.
(You probably wont be able to dislodge it from your mind for a week or
two.)

Im Henery the Eighth, I am,

Henery the Eighth I am, I am!
I got married to the widow next door,
Shes been married seven times before.
Every one was a Henery [Audience: Henery];
She wouldnt have a Willie or a Sam.
Im her eighth old man named Henery;
Im Henery the Eighth I am!

The neatness of the rhymes, of the scanning and the inversion in the
penultimate line, not to say the humour in the surprise denouement, make
this delectable, at least for an hour or two. Music offers many parallel
examples. The best I can do is to refer you to music by Joan Cererois,
Loyset Compre and Stefano Landi in the discography.
Now the first powerful impact of music may well be in the sheer sound
quality of an instrument or a voice, or, perhaps, a tune in a Tchaikovsky
ballet that lodges itself in the mind like the music hall song I have just
quoted. Proust memorably reflects on the little phrase of Vinteuil.23
Then you notice other things or have them drawn to your attention. You
decide to read about Tchaikovskys music or listen to talks on it. You want
to hear more of his music so you try something else. At first it will be
because you are anxious not to miss out on a new source of pleasure. But
later you begin to prize the individuality of his voice and from this point

161
PHILOSOPHY OF MUSIC

on what matters is that it grips your attention; even as this is happening

you might not realize that your enjoyment has gone beyond mere
pleasure. Then you try other composers. You even might find something
absorbing while wishing that it would end. It has become an ordeal yet
one you willingly put yourself through. At the same time this becomes a
process towards the development of an individual taste.
Philosophy of art has been locked into a sort of infantilism. Many
philosophers still think that the explanation and justification of an interest
in music, if not the other arts, lies in the pleasure we take. Either their own
interest in music is superficial or, more often, they do not pay enough
attention to the ways their own tastes have developed. A concentration on
pleasure misplaces the nature of our interest of music. Certainly there is
much in the arts that gives pleasure and to which pleasure is the proper
response. Nobody goes to see a Gilbert and Sullivan operetta or Guys and
Dolls to be exercised over profound issues. But Rembrandt portraits or
Shakespeare tragedies are different, and to say that you enjoyed them
sounds as though your response is pretty inadequate. Similarly with late
Beethoven. You can, if you like, expand the concept of pleasure or
enjoyment to cover such cases, but an implicit redefinition denatures the
original claim. What would be the point?
So we need to make distinctions, distinctions being the centre of
philosophy as they are the centre of art, as Iris Murdoch once observed. We
may say of something it gives me great pleasure or it was a pleasant
experience. But we do not often speak this way when talking of the arts.
More often we will say I enjoyed that. We delight in the combination of
Ira Gershwins lyrics and George Gershwins tunes. Do we enjoy Tosca?
Well, enjoy is certainly not le mot juste. Am I stirred by it? Unless I am
one of those who despises Puccinis shabby little shocker,24 I am. Would
it then be true to say that I enjoy being stirred? I think I do, although
pursuing this path leads into dangerous waters; it tempts us to say that we
go to see Tosca because we enjoy being stirred and that is not so. The work,
not the experience, is the proper focus of attention. Something is wrong if
you go to a concert or to the theatre in search of a particular experience,
much as there is something wrong if you give to a charity because it makes
you feel good and because you prize that feeling, rather than giving to the
charity because there is a need. Of course, just as the good person is one
who is pleased to give to charity so the musical person is one who is stirred
by music. That is the major part of what it is to be musical. But we must not
be confused about the motivation. That she is pleased to give to charity is
not why she gives and that she is stirred by the music may not be the reason
why she attends the concert.

162
 VALUE

The pleasure of the connoisseur, in the view of many aestheticians,

implies and is implied by the recognition of merit. But we need to make a
distinction between getting pleasure from listening to a work and thinking
the music is good. It is important for you to be able to say that you love the
music of Richard Strauss but do not think it of the very highest quality. It
might be that you love it because it gives you pleasure but pleasure, as I
argue, is not the only thing that matters here; it is a mistake to assume that
it is integrally tied in with judgements of value. The music of Strauss
might please you even while you recognize the over-use of effective but
stock devices that makes Strauss a little suspect.
There are more extreme cases still than those we have been consider-
ing. A friend of Billie Holliday, who much admired her art, said that he
found her singing of Strange Fruit too disturbing: It was revolting
listening to it; I dont even like to think about it. (Strange Fruit is about
the lynching of blacks in the American South.) Now this is in no sense at
all a negative judgement. Many feel this way about King Lear, of course.
These are works neither liked nor loved. But they are admired.
If you are seriously interested in drama, you dont expect to enjoy
Wozzeck or for that matter, Saved or Trainspotting: you endure them. Even
some of Bachs cantatas, we are told, were intended to harrow the con-
gregation rather than to titillate their ears. Philosophers puzzle over this
and look for explanations in terms of education for life or some such. But
you only need such an explanation because of a residual and insistent
utilitarian account of human motive in the background: the supposition
that there is an ulterior motive such as pleasure or satisfaction. We see
these operas because we want to know what is there; these may be some of
mankinds greatest achievements and we do not want to miss them no
matter how harrowing they may be. We lose ourselves in them; we cease
to look at the time or at the other members of the audience, wonder at
what they are doing or where and how they live. The great work of music,
finely performed, absorbs us. To be interested is usually to have gone
beyond the stage where pleasure is a motive. Likewise some readers of
this book will be deeply interested in philosophy. If you are, then do you
enjoy doing philosophy? The answer will almost certainly be sometimes
but more often not. Somehow the question is inappropriate. Philosophy
is what you do. In the same way the pursuit of the arts is central to the life
many of us lead.

163
PHILOSOPHY OF MUSIC

The verdict on verdicts

So should we endorse particularism or not? It is sometimes difficult to

give straight answers in philosophy. President Johnson once sighed Oh
for a one-armed lawyer. Equally in philosophy we may be liable to end
with the conclusion, on the one hand this, and on the other hand that.
Although I shall recommend particularism in the end for some yet more
general reasons, considerations run both ways. These considerations com-
mend particularism:

There are no precise and general criteria for merit in music that do not
fall foul of exceptions. Those that have been offered tend to be either
too vague to be very helpful or positively or negatively evaluative
from the outset. They may, too, be overridden by other factors that
make an overall verdict different from the pro tanto judgement.
Switching arguments do not work in the arts. What can be a merit in
one work may be a fault in another.
Even the consensus is constructed from single judgements made by
experienced and qualified listeners. The single judgement is thus
epistemologically and ontologically prior and divergences of judge-
ment may remain.

This consideration runs against particularism. For run-of-the-mill, more

or less competent writing, there are general considerations that may rule
in favour of or against a particular work. These may, in modest work, give
way to a negative judgement in a case where the rules are defied; in
general, only the master is equipped to defy the rules. I should add that the
defiance is not total. Even the revolutionary is revolutionary in terms of
adding to or adjusting an existing tradition.
In as much as we are not required to offer a philosophical theory that
removes difficulties that are part of our life in the arts, we must recognize
that unremovable differences of opinion are part of the data on which we
reflect. I merely say that particularism is an account of value that sits
fairly happily with the existence of unresolvable differences of opinion in
the arts. But this need not require us to deny the existence of maxims or
rules providing that their location and function is clearly recognized.
In an earlier book on the aesthetics of music25 I argued that an agreed and
consistent analysis of what a work of music is could not be obtained because
the different ideologies with which different musicians and music-lovers
approach music ensured differences over its identity. I pointed out that some
take the performance of an interesting interpretation as basic while others

164
 VALUE

give allegiance to the score. My arguments in this chapter strongly suggest,

in addition, that it may be impossible to make our account of artistic value
internally consistent. The existence of irreconcilable disputes seems to rule
out realism. A non-realist account reduces claims about objective value
to claims about what the consensus says. But if, then, the claims made for
or against the work in question are reduced, they turn out to be no more
than expressions of personal taste. But then they have been stripped of their
force. The advocate of Berlioz is not merely saying I like Berliozs work
although I admit it isnt very good; she claims that his music really is
masterly and ought to be recognized as such.
Ideology raises its head again, for you may well argue that a culture in
which individual experience is rated as highly as it is in ours will certainly
allow that experience to play a more dominant role in determining value
than a culture in which general and public standards of goodness and
badness in the arts are assumed in advance and where individual judge-
ments are presumed to be subservient to these. Those cultures would have
no truck with particularism. The fact that we can neither subscribe to
realism nor relinquish it is likely to be a result of competing ideological
strains. That divergences of opinion have to be expressed realistically
reflects our reluctance to surrender reactive judgements and such reactive
judgements are privileged in an ideology that places a premium on the
personal reaction. We are stuck with a demotic metaphysics that cannot be
made consistent.
Do we just acknowledge that this is how it is and jettison any attempt at
philosophically motivated reform of our concepts? Certainly I lean this
way. Indeed, it is not surprising that this sphere should harbour contradic-
tions. After all, we certainly hold contradictory beliefs elsewhere so why
might not the rationale that seems to lie behind our judgement of the value
of music be internally inconsistent? Many systems of belief are. Theol-
ogies often are. Like theologies and ideologies, philosophies of the arts try
to encompass varying interests and considerations. We might, of course,
simply acknowledge that we are irrational in refusing to qualify some
reactive judgements much as we think that a parents blindness about his
or her childs faults is a rather lovable irrationality without which the
world would be a worse place. The difference, though, is that we do
acknowledge that the parents judgement is erroneous by reference to
some agreed facts. But there is no parallel to this in the arts, nothing
accessible beyond the judgements that we make, no independent facts of
the matter about value that can be inspected.
The indications of inconsistency or incoherence in our conceptualiza-
tion of the arts that I have collected are piecemeal. The analytic philosopher

165
PHILOSOPHY OF MUSIC

in me would like there to be more general arguments showing why we

cannot get a coherent picture of the arts and where strains are likely to
occur. But dull experience shows that our concepts arise as the results of
different pressures and different interests. Here and there inconsistencies
may poke through the surface. But I have no intuitions as to how this gives
a basis for predicting the eruptions of incoherence. What is important is to
recognize that our concepts of music and of the arts in general, are concepts
with a history, concepts which change with other changes in society, in its
values and its most general concepts. A result of this is that, at any one time,
a concept may contain, in its sediment, ideas of different ancestry that may
conflict.

166
Coda

I began this book with a list of the philosophical problems most likely to
strike thoughtful musicians and music-lovers: questions about musics
power, whether it can have meaning, what counts as art music and ques-
tions about the divergence of taste. I said then that these are questions that
relate to the value that music has for us. My approach has been problem-
orientated and the nature of those problems made it natural that I should
conclude with a discussion of value. Other writers have viewed music
primarily as a branch of metaphysics,1 raising questions about sounds,
tones and musical movement. My approach has been to introduce only such
ontology as is requisite to help with the problems at hand and I think it fair
to say that most recent writers on music share this approach.
My conclusions favour particularism and I shall close by making a more
general case for it. In my previous book on the philosophy of music, Music
and Humanism, I suggested that for four centuries language has been a
model in terms of which we understand music, a model which we use,
consciously or unconsciously, in approaching Western classical music. We
use metaphors drawn from language, we speak of musical paragraphs and,
most significantly, music can be followed and it can be followed over a long
period. As we saw, for a long time it was a clich that music is a language
of the emotions. Western music has performed a sort of arc from being a
decorative art to being what is sometimes called a fine art and back to a
decorative art again; for it is a familiar difficulty with most advanced
Western music, beginning with the second Viennese school, that it cannot
be followed in the way that music of the past can, even after we have
familiarized ourselves with the style. When music was constructed and
understood on linguistic analogies, with rhetoric as a model, it was an art
that I would describe as a humanist art. Three characteristics are essential

167
PHILOSOPHY OF MUSIC

here. First, it can be described in expressive terms as grave, exuberant

and the like. Secondly, it can be interpreted, not just in the sense that
performers interpret the music from a score, but in the sense that it can be
given a meaning. Thirdly, it can be followed. I have discussed these claims
in the previous chapters. Not all music that we classify as fine art offers
much to those who approach it in these terms but the very finest music of
the Western classical era does, and it is this that gives us the essence of the
art of that place and time. In this respect great art is stereotypical. It provides
a paradigm and the paradigm defines the art no matter how small a
proportion of the music rewards the effort of being thought of in these terms.
Part of what I have called a humanist element is the individuality we
prize in a composer. For many listeners the music of Tallis speaks across
the centuries as, more than any other composer of his generation, the
voice of an individual and for that reason we love his music above that of
his contemporaries. Thus in one respect, and at a very general level,
particularism seems to accord with the way judgements are made. Yes, we
look for unity and diversity, for technical accomplishment in the layout
for instruments and voices, but we also look for imaginative ideas and
imagination in their treatment and in this respect the crucial features of
our engagement with music (and, pari passu, with other arts) begin to
emerge. For what we prize, all things being equal, and what makes the
difference between the great master and the smaller talent, is the
individuality of voice. In the same way, when your friends die it comes as
a shock to realize that you miss more those whose wit and intelligence
took you by surprise. The death of a Mercutio distresses us in the way that
the death of a Polonius does not. Distinctiveness is the mark of the master.
Whatever role is played by generalizations fades into the background.
And the roles of beauty or grace or the giving of pleasure matter less. Of
course, we could hardly dispute that there is a generalization that we
prize originality nor would I dispute that originality is merely one virtue
among many and one that could be counter-balanced by defects. But I do
suggest that individuality is the most prized of artistic virtues in our
culture. It informs our current concept of art and it explains a great deal.
For we cannot prize what is individual without prizing what makes it
different from others; it is this that, at the summits of the art of music,
makes the search for critical generalities fruitless.
I suggested that in some ways this is akin to the way we value our
friends. Earlier I quoted Hume:

We choose our favourite author as we do our friend, from a con-

formity of humour and disposition. Mirth or passion, sentiment or

168
 CODA

reflection; whichever of these most predominates in our temper, it

gives us a peculiar sympathy with the writer who resembles us.2

But in some ways friendship is not an apt parallel. It matters to me that I

do not miss the opportunity of hearing a piece of music or a performance
that is said to be outstanding and, in the pursuit of this, I am prepared to
encounter some boredom. But I do not seek out people who might interest
me because, of course, I cannot make such demands of others. In fact, to
prosecute an acquaintance on the assumption that I might find her inter-
esting would give her a reason not to regard me as a friend. For friendship
should have no such ulterior motive. I cannot make a claim on her because
I think, in advance, that she might interest me. Those who nurture
acquaintances in this way, like Mme. Verdurin in Prousts Remembrance
of Things Past, are suspect. For there to be anything like an analogy with
music there would need to be something like the operation of a consensus
advising me which people to pursue or advice from somebody who knows
more people than I do as to who to endeavour to meet. But friendships are
far more the product of accident and it is in the nature of friendship that
this should be so.
For all this, friendship is not so bad an analogy. Charles Lamb wrote
about the death of a friend:

One sees a picture, reads an anecdote, starts a casual fancy, and thinks
to tell it to this person in preference to every other the person is
gone whom it would have peculiarly suited ... Common natures do
not suffice me. Good people, as they are called, wont serve. I want
individuals.

As with friends, so with music.

169
Notes

Introduction

1. Charles Rosen, Piano Notes (London: Allen Lane, 2003).

2. Critical discussions of these two thinkers will be found in Malcolm Budd, Music and the
Emotions (London: Routledge & Kegan Paul, 1985). Roger Scruton, who is a sym-
pathetic critic, discusses Schopenhauer in Death Devoted Heart (Oxford: Oxford
University Press, 2004).

Chapter 1: Overture and beginnings

1. W. Wordsworth, Lyrical Ballads, Preface to second edition.

2. See Lydia Goehr, The Imaginary Museum of Musical Works (Oxford: Clarendon Press,
1992).
3. Enrico Fubini, A History of Musical Aesthetics, M. Hatwell (trans.) (London:
Macmillan, 1990) gives an excellent account of the subject up to about 1975. It does not
cover recent work.
4. E. Hanslick, On the Musically Beautiful, G. Payzant (trans.) (Indianapolis, IN: Hackett,
1986).
5. See ibid., 3.
6. See ibid., xxiii, 3.
7. Ibid., 32.
8. Ibid., 9.
9. See ibid., 89.
10. Ibid., 22 and see p. 14.
11. Ibid., 67.
12. Ibid., 19.
13. Ibid., 910.
14. Ibid., 11.
15. Ibid., 80.
16. Ibid., 64.
17. Ibid., 59.
18. Ibid., 49.

171
PHILOSOPHY OF MUSIC

19. Ibid., 1.
20. Edmund Gurney, The Power of Sound (New York: Basic Books, 1966 [1880]), 202.
21. Ibid., 226n.
22. Ibid., 296.
23. Ibid., 246n.
24. Ibid., 43.
25. Ibid., 369.
26. Ibid., 91.
27. Ibid., 204ff. This aspect of his thought is defended by Jerrold Levinson, Music in the
Moment (Ithaca, NY: Cornell University Press, 1997).
28. Ibid., 165.
29. Ibid., 375.
30. Ibid., 339, 342.
31. Ibid., 340.
32. Ibid., 345.
33. Ibid., 3245.
34. Ibid., 140.
35. Ibid., 177.
36. Ibid., 344.

Chapter 2: The work of music

1. The quotation from John Cage comes from Noel Carroll, Cage and Philosophy, in
Musical Worlds, Philip Alperson (ed.) (University Park, PA: Pennsylvania State
University Press, 1998). There is an excellent discussion of Cages Silent Music in
Stephen Davies, Themes in the Philosophy of Music (Oxford: Oxford University Press,
2003).
2. Stanley Cavell, Must We Mean What We Say? (Cambridge: Cambridge University Press,
1976).
3. See Calvin Tomkins, The Bride and the Bachelors (London: Weidenfield & Nicolson,
1965) on this and much else that is relevant.
4. W. B. Gallie was responsible for the idea of essentially contested concepts. See his
Philosophy and the Historical Understanding (London: Chatto & Windus, 1964), Ch. 8.
5. Wittgenstein, Philosophical Investigations (Oxford: Blackwell, 1953), 142.
6. For discussions of procedural definitions, and so on, see Stephen Davies, Definitions of
Art (Ithaca, NY: Cornell University Press, 1991). See also Berys Gaut, Art as a Cluster
Concept, in Theories of Art Today, Noel Carroll (ed.), 2544 (Madison, WI: University
of Wisconsin Press, 2000).
7. See Robert Stecker, Artworks (University Park, PA: Pennsylvania State University Press,
1997).
8. Wittgenstein, Lectures and Conversations on Aesthetics, Psychology and Religious
Belief, Cyril Barrett (ed.) (Oxford: Blackwell, 1966), 34.
9. On historical accounts, the locus classicus is Jerrold Levinson, Music, Art and
Metaphysics (Ithaca, NY: Cornell University Press, 1990).
10. T. J. Diffey, The Republic of Art, British Journal of Aesthetics 9 (1969), 14556.
11. Kenneth Clark, Civilization (London: The Folio Society, 1999), 1524.
12. R. A. Sharpe, Contemporary Aesthetics (Hassocks: Harvester, 1983, reprinted Alder-
shot: Gregg Revivals, 1990).
13. Love-spoons are a Welsh tradition; a love-spoon was carved for a partner and the various
devices on the spoon symbolized such marital virtues as care and fidelity.

172
 NOTES

14. See Stan Godlovitch, Musical Performance (London: Routledge, 1998).

15. On the distinctive nature of the work of rock see John Andrew Fisher, Rock n
Recording: The Ontological Complexity of Rock Music, in Alperson (ed.), Musical
Worlds, 10923. Glenn Goulds ideas on classical recording are relevant here. See The
Glenn Gould Reader, Tim Page (ed.) (London: Faber and Faber, 1987). Andy Hamilton
has a well-informed article on these issues: The Art of Recording and the Aesthetics of
Perfection, British Journal of Aesthetics 43(4) (2003), 34562. See also Stephen
Davies, So You Want to Sing with the Beatles? Too late!, in Themes in the Philosophy
of Music, 94107 (Oxford: Oxford University Press, 2003).
16. See Goehr, The Imaginary Museum of Musical Works.
17. Complete Pianoforte Sonatas, Volume III: Beethoven, H. Craxton (ed.), D. F. Tovey
(ann.) (London: ABRSM Publishing), 217.
18. A version that distinguishes the composers instructions from editorial additions.
19. See Nicholas Wolterstorff, Worlds and Works of Art (Oxford: Clarendon Press, 1980).
20. I have not acknowledged many articles in these notes, preferring to keep references to
those that are easily obtainable and readable by beginners, but I have learned from the
lively debate initiated by Julian Dodds Musical Works as Eternal Types, British
Journal of Aesthetics 40 (2000), 42440. But there is by now a considerable literature on
this. For a start consult Stephen Davies, Musical Works and Performances (Oxford:
Oxford University Press, 2001). Much that is relevant to the question of discovery is to
be found in Peter Kivy, The Fine Art of Repetition (Cambridge: Cambridge University
Press, 1993), Essays II, III and IV.
21. The debate on constructivism has been lively. See Michael Krausz, Rightness and
Reasons: Interpretations in Cultural Practices (Ithaca, NY: Cornell University Press,
1993). There was a discussion in the late 1990s conducted through the pages of the
Journal of Aesthetics and Art Criticism. See Robert Stecker, Interpretation and
Construction (Oxford: Blackwell, 2003) for a far more detailed account of the
arguments than I have space for here.
22. See Peter Kivy, Authenticities (Ithaca, NY: Cornell University Press, 1995), Randall
Dipert, The Composers Intentions: An Examination of their Relevance for Perform-
ance, Musical Quarterly 66 (1980) and T. S. Kuhn, The Structure of Scientific
Revolutions (Chicago, IL: University of Chicago Press, 1962).
23. See Peter Kivy, The Possessor and the Possessed (New Haven, CT: Yale University
Press, 2001).
24. On Janeks Jen fa see John Tyrell, Janeks Operas (London: Faber and Faber,
1992), 1007.
25. My information on Toscanini and on the orchestration of Prokofievs Romeo and Juliet
comes from Erich Leinsdorf, The Composers Advocate (New Haven, CT: Yale
University Press, 1981), 200.

Chapter 3: Meaning

1. Igor Stravinsky, Autobiography (London: Calder and Boyars, 1975), 534.

2. The Best of the Two Ronnies (BBC Video: M and S 2224633).
3. Deryck Cooke, The Language of Music (Oxford: Oxford University Press, 1959) gives a
famous glossary linking phrases and expressive descriptions within a basically
expressionist aesthetics of music.
4. See Gregory Karl and Jenefer Robinson, Shostakovichs Tenth Symphony and the
Musical Expression of Cognitively Complex Emotions, in Music and Meaning, Jenefer
Robinson (ed.) (Ithaca, NY: Cornell University Press, 1997).

173
PHILOSOPHY OF MUSIC

5. Leo Tolstoy, What is Art and Essays on Art, A. Maude (trans.) (Oxford: Oxford
University Press, 1930).
6. See Colin Radford, Emotions and Music: A Reply to the Cognitivists, Journal of
Aesthetics and Art Criticism 47 (1989), 6976 and Muddy Waters, Journal of
Aesthetics and Art Criticism 49 (1991), 24752, Aaron Ridley, Music, Value and the
Passions (Ithaca, NY: Cornell University Press, 1995) and Derek Matravers, Art and
Emotion (Oxford: Clarendon Press, 1998). There are, naturally, some differences
between them.
7. Martha Nussbaum, Upheavals of Thought (Cambridge: Cambridge University Press,
2001), Ch. 5 and see esp. 249, 251.
8. Wittgensteins thoughts about the relevance of gesture are best investigated via Zettel
(Oxford: Blackwell, 1967) para. 15775, rather than the more commonly used Lectures
and Conversations, Cyril Barrett (ed.) (Oxford: Blackwell, 1966), which are based on
student notes. I find it hard to think that Wittgenstein would have said that the expression
of emotion is a gesture (see Lectures and Conversations, 37).
9. In a famous paper by O. K. Bouwsma, The Expression Theory of Art, in Aesthetics
and Language, William Elton (ed.) (Oxford: Blackwell, 1959).
10. D. F. Tovey, Essays in Musical Analysis III: The Concerto (Oxford: Oxford University
Press, 1946), 143.
11. Lionel Pike, Beethoven, Sibelius and the Profound Logic (London: Athlone, 1978).
12. Peter Cooke and Dudley Moore, The Dagenham Dialogues (London: Methuen, 1971).
13. R. Graves, The Challenge, in Collected Poems (London: Cassell, 1938).
14. See Miranda Seymour, Robert Graves: Life on the Edge (London: Doubleday, 1996),
235.
15. E. T. Cone, Schuberts Promissory Note; an Exercise in Musical Hermeneutics,
Nineteenth Century Music 5(3) (1982), 23341.
16. See Lawrence Kramer, Musical Meaning (Berkeley, CA: University of California Press,
2002).
17. See Jerrold Levinson, Performative vs. Critical Interpretation in Music, in The
Interpretation of Music, Michael Krausz (ed.) (Oxford: Oxford University Press, 1993).
18. On the new musicology, see Peter Kivy, New Essays on Musical Understanding
(Oxford: Clarendon Press, 2001), Ch. 8.
19. Edward J. Dent, Mozarts Operas (Oxford: Oxford University Press, 1970), 177.
20. See E. T. Cone, The Composers Voice (Berkeley, CA: University of California Press,
1974), Ch. 7 and Nicholas Cook, Analysing Musical Multi-Media (Oxford: Oxford
University Press, 1998).
21. Samuel Johnson, The Idler no. 51, in Selected Essays, W. J. Bate (ed.) (New Haven,
CT: Yale University Press, 1968), 319.

Chapter 4: Value

1. John Mackie, Ethics: Inventing Right and Wrong (Harmondsworth: Penguin, 1977).
2. On these topics see Frank Sibley, Approach to Aesthetics (Oxford: Oxford University
Press, 2001).
3. Arnold Schoenberg, Theory of Harmony, R. Carter (trans.) (London: Faber and Faber,
1978), 60ff.
4. On the ideal observer theory see Charles Taliaferro, The Ideal Aesthetic Observer
Revisited, British Journal of Aesthetics 30(1) (1990), 113 and John Hospers, The
Ideal Aesthetic Observer, British Journal of Aesthetics 2(2) (1962), 99111.
5. Oscar Wilde, The Critic as Artist, in Intentions (London: Methuen, 1927), 189.

174
 NOTES

6. David Hume, Of the Standard of Taste, Selected Essays (Oxford: Oxford University
Press, 1993 [1741]), 150.
7. Sibley, Approach to Aesthetics, 77.
8. Samuel Alexander, Beauty and other Forms of Value (London: Macmillan, 1933), 176.
9. See Alan Goldman, Aesthetic Value (Boulder, CO: Westview, 1995), Ch. 2.
10. Isaiah Berlin, The Proper Study of Mankind (London: Chatto & Windus, 1997).
11. The storm and stress movement of German literature in the 1760s and 1770s, notable
in the work of Goethe and Schiller, had a musical parallel in the dramatic style of Haydn
and J. C. Bach in the early 1770s.
12. R. Vaughan Williams, Some Thoughts on Beethovens Choral Symphony, in National
Music and Other Essays (Oxford: Oxford University Press, 1963).
13. The influence of Peter Winchs The Universalizability of Moral Judgements, in Ethics
and Action, 15170 (London: Routledge & Kegan Paul, 1972) will be obvious to
philosophers.
14. Hume, Of the Standard of Taste, 141.
15. When Wilde described a sunset as a third-rate Turner, the joke is, of course, that concepts
of taste are normally inapplicable in these cases. We do develop tastes in landscape but
this is a sophistication which strikes us as a little pretentious and it was exactly in this
way that Wilde was, of course, sending himself up.
16. Wittgenstein, Lectures and Conversations, 14.
17. I have found Brad Hooker and Margaret Little, Moral Particularism (Oxford: Oxford
University Press, 2000) helpful, especially the essays by Jonathan Dancy and Margaret
Little. Many of the forms of particularism I note here are discussed in their moral guise
in the various essays that make up this collection.
18. In R. A. Sharpe, Contemporary Aesthetics (Hassocks: Harvester, 1983).
19. Sibley, Approach to Aesthetics, 76.
20. A Desert Island Discography, in The Glenn Gould Reader, Tim Page (ed.), 43740
(London: Faber and Faber, 1987).
21. Wittgenstein, Lectures and Conversations, 3.
22. See Stephen Davies, Why Listen to Sad Music if it Makes One Feel Sad?, in Music
and Meaning, Jenefer Robinson (ed.), 24253 (Ithaca, NY: Cornell University Press,
1997).
23. Marcel Proust, Swanns Way (Harmondsworth: Penguin, 1999), 37884.
24. Joseph Kerman, Opera as Drama (New York: Vintage Books, 1956), 254.
25. R. A. Sharpe, Music and Humanism (Oxford: Oxford University Press, 2000).

Coda

1. For example, see Roger Scruton, The Aesthetics of Music (Oxford: Clarendon Press,
1997).
2. Hume, Of the Standard of Taste, 150.

175
Bibliography and discography

Further reading

If, as I hope, you are encouraged to delve deeper into the extensive con-
temporary literature on the aesthetics of music, you may find this bibli-
ography useful. Over the past decade, publications in the area of the
philosophy of music have increased dramatically; significant contribu-
tions are largely to be found in the two major journals in aesthetics, The
British Journal of Aesthetics and the Journal of Aesthetics and Art
Criticism. The books I have recommended share one thing in common;
they are well written, indeed some of the authors have a very considerable
feeling for the shape and rhythms of English prose and they are accessible
to the general reader. If you have approached this subject from, say, liter-
ary theory or from some other branch of philosophy such as the philoso-
phy of language or philosophy of mind, the aesthetics of music will prove,
in that respect, a pleasant surprise.
Alperson, P. (ed.) 1998. Musical Worlds. University Park, PA: Pennsylvania State University
Press.
Cook, N. 1990. Music, Imagination and Culture. Oxford: Oxford University Press.
Davies, S. 1994. Musical Meaning and Expression. Ithaca, NY: Cornell University Press.
Davies, S. 2001. Musical Works and Performance. Oxford: Oxford University Press.
[Davies offers exhaustive surveys of the recent literature but these books are not just a
matter of who says what. There are many acute criticisms and keen observations here.]
Fubini, E. 1990. A History of Musical Aesthetics, M. Hatwell (trans.). London: Macmillan.
Godlovich, S. 1998. Musical Performance. London: Routledge.
Goehr, L. 1992. The Imaginary Museum of Musical Works. Oxford: Clarendon Press.
Kivy, P. 1989. Sound Sentiment. Philadelphia, PA: Temple University Press.
Kivy, P. 1990. Music Alone. Ithaca, NY: Cornell University Press.
Kivy, P. 1993. The Fine Art of Repetition. Cambridge: Cambridge University Press.
Kivy, P. 1995. Authenticities. Ithaca, NY: Cornell University Press.

177
PHILOSOPHY OF MUSIC

Kivy, P. 2001. New Essays on Musical Understanding. Oxford: Clarendon Press.

Kivy, P. 2002. Introduction to a Philosophy of Music. Oxford: Oxford University Press. [A
useful general account of Kivys views.]
Ridley, A. 1995. Music, Value and the Passions. Ithaca, NY: Cornell University Press.
Sharpe, R. A. 2000. Music and Humanism. Oxford: Oxford University Press.

References

Alexander, S. 1933. Beauty and other Forms of Value. London: Macmillan.

Alperson, P. (ed.) 1998. Musical Worlds. University Park, PA: Pennsylvania State University
Press.
Berlin, I. 1997. The Proper Study of Mankind. London: Chatto & Windus.
Bouswma, O. K. 1959. The Expression Theory of Art. In Aesthetics and Language, W.
Elton (ed.), 7399. Oxford: Blackwell.
Budd, M. 1985. Music and the Emotions. London: Routledge & Kegan Paul.
Carroll, N. 1998. Cage and Philosophy. In Musical Worlds, P. Alperson (ed.), 12530.
University Park, PA: Pennsylvania State University Press.
Cavell, S. 1976. Must We Mean What We Say? Cambridge: Cambridge University Press.
Cone, E. T. 1974. The Composers Voice. Berkeley, CA: University of California Press.
Cone, E. T. 1982. Schuberts Promissory Note: An Exercise in Musical Hermeneutics,
Nineteenth Century Music 5(3), 23341.
Cook, N. 1998. Analysing Musical Multi-Media. Oxford: Oxford University Press.
Cooke, D. 1959. The Language of Music. Oxford: Oxford University Press.
Davies, S. 1991. Definitions of Art. Ithaca, NY: Cornell University Press.
Davies, S. 1997. Why Listen to Sad Music if it Makes One Feel Sad?. In Music and
Meaning, J. Robinson (ed.), 24253. Ithaca, NY: Cornell University Press.
Davies, S. 2001. Musical Works and Performance. Oxford: Oxford University Press.
Davies, S. 2003. Themes in the Philosophy of Music. Oxford: Oxford University Press.
Dent, E. J. 1970. Mozarts Operas. Oxford: Oxford University Press.
Diffey, T. J. 1969. The Republic of Art, British Journal of Aesthetics 9, 14556.
Dipert, R. 1980. The Composers Intentions: An Examination of their Relevance for
Performance, Musical Quarterly 66, 20518.
Dodd, J. 2000. Musical Works as Eternal Types, British Journal of Aesthetics 40, 42440.
Fisher, J. A. 1998. Rock n Recording: The Ontological Complexity of Rock Music. In
Musical Worlds, P. Alperson (ed.), 10923. University Park, PA: Pennsylvania State
University Press.
Fubini, E. 1990. A History of Musical Aesthetics, M. Hatwell (trans.). London: Macmillan.
Gallie, W. B. 1964. Philosophy and the Historical Understanding. London: Chatto &
Windus.
Gaut, B. 2000. Art as a Cluster Concept. In Theories of Art Today, N. Carroll (ed.), 25
44. Madison, WI: University of Wisconsin Press.
Godlovich, S. 1998. Musical Performance, London: Routledge.
Goehr, L. 1992. The Imaginary Museum of Musical Works. Oxford: Clarendon Press.
Goldman, A. 1995. Aesthetic Value. Boulder, CO: Westview.
Gould, G. 1987. The Glenn Gould Reader, T. Page (ed.). London: Faber and Faber.
Gurney, E. 1966. The Power of Sound. New York: Basic Books.
Hamilton, A. 2003. The Art of Recording and the Aesthetics of Perfection, British Journal
of Aesthetics 43(4), 34562.
Hanslick, E. 1986. On the Musically Beautiful, G. Payzant (trans.). Indianapolis, IN:
Hackett.

178
 BIBLIOGRAPHY AND DISCOGRAPHY

Hooker, B. & M. Little (eds) 2000. Moral Particularism. Oxford: Oxford University Press.
Hospers, J. 1962. The Ideal Aesthetic Observer, British Journal of Aesthetics 2(2), 99
111.
Hume, D. Of the Standard of Taste, in Selected Essays, 13353. Oxford: Oxford
University Press, 1993 [1741].
Karl, G. & J. Robinson 1997. Shostakovitchs Tenth Symphony and the Musical
Expression of Cognitively Complex Emotions. In Music and Meaning, J. Robinson
(ed.), 15478. Ithaca, NY: Cornell University Press.
Kivy, P. 1993. The Fine Art of Repetition. Cambridge: Cambridge University Press.
Kivy, P. 1995. Authenticities. Ithaca, NY: Cornell University Press.
Kivy, P. 2001. New Essays on Musical Understanding. Oxford: Clarendon Press.
Kivy, P. 2001. The Possessor and the Possessed. New Haven, CT: Yale University Press.
Kramer, L. 2002. Musical Meaning. Berkeley, CA: University of California Press.
Krausz, M. 1993. Rightness and Reasons: Interpretations in Cultural Practices. Ithaca, NY:
Cornell University Press.
Kuhn, T. S. 1962. The Structure of Scientific Revolutions. Chicago, IL: University of
Chicago Press.
Leinsdorf, E. 1981. The Composers Advocate. New Haven, CT: Yale University Press.
Levinson, J. 1990. Music, Art and Metaphysics. Ithaca, NY: Cornell University Press.
Levinson, J. 1993. Performative vs. Critical Interpretations in Music. In The
Interpretation of Music, M. Krausz (ed.), 3360. Oxford: Clarendon Press.
Levinson, J. 1997. Music in the Moment. Ithaca, NY: Cornell University Press.
Mackie, J. 1977. Ethics: Inventing Right and Wrong. Harmondsworth: Penguin.
Matravers, D. 1998. Art and Emotion. Oxford: Clarendon Press.
Nussbaum, M. 2001. Upheavals of Thought. Cambridge: Cambridge University Press.
Pike, L. 1978. Beethoven, Sibelius and the Profound Logic. London: Athlone.
Radford, C. 1989. Emotions and Music: A Reply to the Cognitivists, Journal of Aesthetics
and Art Criticism 47, 6976.
Radford, C. 1991. Muddy Waters, Journal of Aesthetics and Art Criticism 49, 24752.
Ridley, A. 1995. Music, Value and the Passions. Ithaca, NY: Cornell University Press.
Rosen, C. 2003. Piano Notes. London: Allen Lane.
Schoenberg, A. 1978. Theory of Harmony, R. Carter (trans.). London: Faber and Faber.
Scruton, R. 1997. The Aesthetics of Music. Oxford: Clarendon Press.
Scruton, R. 2004. Death Devoted Heart. Oxford: Oxford University Press.
Seymour, M. 1996. Robert Graves: Life on the Edge. London: Doubleday.
Sharpe, R. A. 1983. Contemporary Aesthetics. Hassocks: Harvester.
Sharpe, R. A. 2000. Music and Humanism. Oxford: Oxford University Press.
Sibley, F. 2001. Approach to Aesthetics. Oxford: Oxford University Press.
Stecker, R. 1997. Artworks. University Park, PA: Pennsylvania State University Press.
Stecker, R. 2003. Interpretation and Construction. Oxford: Blackwell.
Stravinsky, I. 1975. Autobiography. London: Calder and Boyars.
Taliaferro, C. 1990. The Ideal Aesthetic Observer Revisited, British Journal of Aesthetics
30(1), 113.
Tolstoy, L. 1930. What is Art and Essays on Art, A. Maude (trans.). Oxford: Oxford
University Press.
Tomkins, C. 1965. The Bride and the Bachelors. London: Weidenfeld & Nicolson.
Tovey, D. F. 1946. Essays in Musical Analysis III: The Concerto. Oxford: Oxford University
Press.
Tyrell, J. 1992. Jan eks Operas. London: Faber and Faber.
Vaughan Williams, R. 1963. Some Thoughts on Beethovens Choral Symphony. In
National Music and Other Essays, 83120. Oxford: Oxford University Press.
Wilde, O. 1927. The Critic as Artist. In Intentions, 95217. London: Methuen.

179
PHILOSOPHY OF MUSIC

Winch, P. 1972. The Universalizability of Moral Judgements. In Ethics and Action, 151
70. London: Routledge & Kegan Paul.
Wittgenstein, 1953. Philosophical Investigations, E. Anscombe (trans.). Oxford: Blackwell.
Wittgenstein, 1966. Lectures and Conversations on Aesthetics, Psychology and Religious
Belief, Cyril Barrett (ed.). Oxford: Blackwell.
Wittgenstein, 1967. Zettel, E. Anscombe (trans.). Oxford: Blackwell.
Wolterstorff, N. 1980. Worlds and Works of Art. Oxford: Clarendon Press.

Discography

I list recordings that may need some tracking down or where certain
versions seem to me very much to be preferred. There are several
excellent versions of, say, the symphonies of Shostakovich or Prokofiev
and advice is readily available.
Count Basie, Lester Leaps In, on Count Basie (Bluenite BN005, 1996). [Unfortunately
the CD has no details of the musicians and date save the information that it is, remarkably,
a live recording.]
Beethoven, Piano Sonatas Nos. 23, 28, 30, 31, Solomon (Testament SBT 1192, 2000).
Benjamin Britten, Sacred and Profane, The Sixteen, Harry Christophers (cond.) (Collins
Classics 13432, 1993).
Holst, Hymns from the Rig Veda, on a CD including Two Eastern Pictures and Hymn to
Dionysus, Royal Philharmonic Orchestra and Royal College of Music Chamber Choir,
Sir David Willcocks (cond.) (Unicorn-Kanchana DKP(CD) 9046, 1985). [Unfortunately,
the CD is not helpfully banded.]
Joan Cererois, Serfn, que con dulce harmon, sung by Emily Van Evera and Timothy
Wilson [the incomparable counter-tenor], on The Christmas Album, Taverner Consort,
and Choir Players, Andrew Parrott (cond.) (EMI CDC 7 54529 2, 1992; reissued Veritas
VC5451552, 2002). [There is another version of this on a more recent, much lauded,
disc: Missa Mexicana, The Harp Consort, Andrew Lawrence-King (cond.) (Harmonia
Mundi HMU 907293, 2002). I rather prefer the first.]
Loyset Compre, Le grant Desir, sung by Catherine Bott and Catherine King, on
Renaissance Love Songs (BBC MM54, 1997). [This is a disc that was issued to
accompany the BBC Music Magazine V(6). You will need to track it down through
secondhand dealers.]
Dvok. Slavonic Dances Series I op. 46, Series II op. 72, Czech Philharmonic Orchestra,
Karel Sejna (cond.) (Supraphon SU1916-2-011 (1995) [recorded 1959]).
B. Goodman, L. Hampton, C. Christian, AC-DC Current, Benny Goodman Sextet: Benny
Goodman (clarinet), Lionel Hampton (vibes), Fletcher Henderson (piano), Charlie
Christian (guitar), Artie Bernstein (bass), Nick Fatool (drums). [Recorded live in 1939
and available on various albums.]
Jimmy Guiffre Trio, The Train and the River, on Sound of Jazz (Columbia CK
45234).
Janek. Jen fa, Chorus and Orchestra of the National Theatre Prague, Bohumil Gregor
(cond.) (EMI CMS 7 243 5 65476 29, 1995). [This performance embodies the Kovarovic
alterations.]
Stefano Landi. Homo Fugit velut umbra . . . LArpeggiata, Christina Pluhar (cond.) (Alpha
020, 2002). [This is very much a version for twenty-first century ears, I suspect.]
Richard Strauss, Gesang der Apollonpriesterin op. 33 no. 2, on Orchestral Songs vol. 2,

180
 BIBLIOGRAPHY AND DISCOGRAPHY

Felicity Lott with Royal Scottish National Orchestra, Neeme Jrvi (cond.) (Chandos
CHAN 9159, 1993).
Tallis, Spem in alium. Of the many recordings, two stand out. For a cathedral acoustic find
Tallis: Sacred Choral Works, with Winchester Cathedral Choir, Winchester College
Quiristers and Vocal Arts, David Hill (cond.) (Hyperion CDA66400, 1990). For a less
reverberant acoustic but perhaps a more authentic one, if it is true that it was first
performed in the Long Gallery at Arundel House, listen to the recording by the Tallis
Scholars, Peter Phillips (cond.) (Gimell 454906-2, 1985; reissued 1994). Imagine dying
before you had a chance to hear this music.

181
Index

Alexander, S. 137 Cage, John 314, 48

Alkan, C.-V. 46 Cavell, Stanley 33
Altamira cave art 37 Champion, Harry 161
Aristotle 14 Clark, Kenneth 45
arousalism 2021, 96101 cognitivism 96, 99
Auden, W. H. 39 Collingwood, R. G. 42
authenticity 7784, 143 Cone, E. 11213, 115
autonomy 267, 94 constructivism 7077, 78

Bach, J. S. 19, 39 Davies, Stephen vii

Mass in B minor 81 Davis, Miles 212
French Suite No. 5 52 Descartes, R. 10, 15
Well-Tempered Clavier, Book 1, no. 18 Diffey, T. J. 44
119 Dvok, A. 119
Basie, Count, Lester Leaps In 545, Slavonic Dances, op. 46 no. 5 108
56
Beethoven, L. 22 Elgar, E. 67
Hammerklavier Sonata op. 106 119, The Dream of Gerontius 11112
120 First Symphony 81, 115
Moonlight Sonata, op. 27 no. 2 756, emotion and feeling 1720, 256, 92,
113, 116 945, 96102
Piano Sonata op. 109 879 error theory 9091
Piano Sonata op. 110 61, 70 expression 51, 767, 84, 85121; see
String Quartet op. 135 86 also Hanslick; Gurney
Symphony no. 5 11416
Berlioz, H. 142, 151 formalism 3, 1622; see also music,
Romo et Juliette 117 absolute music
Symphonie Fantastique 117 Freudian interpretation 39, 113
Les Troyens 115, 142 Furtwngler, W. 157
Brahms, J. 78
Double Concerto 1034 Goodman, Benny, AC-DC Current 57,
Britten, B. 8 1545
The Turn of the Screw 3940 Gould, Glenn 157
Bruckner, A. 834 Graves, Robert 112

183
PHILOSOPHY OF MUSIC

Greek philosophy of music 1314 Mozart, W. A. 1718, 20, 33, 46, 141,
Gurney, E. 227 1567
The Marriage of Figaro 117
Hanslick, E. vii, 1622, 25, 26, 27, 51, Symphony in G minor, no. 40 77, 143
95, 101, 120 music
Hardy T., Afterwards 103 absolute music 3, 10, 85, 107
Haydn, J. 19, 20, 24 as art 2954
Surprise Symphony, no. 94 83 and the canon 7, 1256
Military Symphony, no. 100 143 and gesture 9092, 101, 174 (n. 8)
Helmoltz, H. 23 and historically informed performance
Hobbes, T. 10 7980; see also authenticity
Hockney, D. 33 and interpretation 56, 57, 5960, 623,
Hogarth, William 33 7077, 7884, 10821, 140
Holliday, Billie, Strange Fruit 87, 163 and language 10, 15, 1622, 257, 92
Holst G., Hymns from the Rig Veda, 4, 1678
(second series) op. 26 109 and score 613
Hume, D. and text 15
Dialogues Concerning Natural and value 97, 12369; see also value
Religion 8, 10 work of 29, 5477
Of the Standard of Taste 132, 135,
1378, 149, 1689 negative emotions 23
Nietzsche, F. 14
ideal observer 1326, 148
ideology 3, 7, 10, 17, 43, 1646 Parker, Charlie, She Rote 55
improvisation 5460 particularism 15260, 1649
interest 1614 Peirce, C. S. 61
performance 21, 5961, 7784, 1557
Janek, L., Jenfa 84, 87 philosophical analysis 7, 913
Jane Eyre 30 Platonism 58, 639
Johnson, S. 118, 119 pleasure 14, 212, 23, 38, 81
Review of Soame Jenyns 8 compared with interest 1614
profundity 11821
knowledge, its analysis 35 Prokofiev, S. 3, 23
Kuhn, T. 11, 79 Fifth Symphony 3
Proust, M. 161
Lamb, Charles 169 Puccini, G., Tosca 162
Langer, S., futility of studying 8
Leinsdorf, E. 84 Rachmaninov, S., Third Piano Concerto
Liszt, F., transcriptions of Schubert songs 57
82 Rameau, J.-P. 13, 15, 83
Rauschenberg, Robert 34
Mahler, G. 75, 79 realism 5870, 778, 12566
Fifth Symphony 108 reference 8990
meaning 85121 Rembrandt, The Jewish Bride 45
medieval philosophy of music 14 reductionism 149, 151
Mendelssohn, F. 20 Renaissance philosophy of music 15
Messiaen, O. 146 Renoir, A. 116
metaphor 1028 representation 19
metaphysics 4950 Rti, R. 24
Milton, John, Paradise Lost 39 Rosen, Charles 4
Mingus, Charles, Thrice Upon a
Theme 119 Schenker, H. 22, 24

184
 INDEX

Schoenberg, A. 130, 139, 141 and criteria 41, 12932, 148

Schopenhauer, A. 8, 16 and divergence of judgement 13851;
Schubert, F. 159 see also particularism
Duet Fantasie in F minor 100101 as expressive 15960
Moment Musical in A flat D.780 and objectivity 1278, 12366 passim
11213, 114, 115 and pleasure 38, 1614
Schumann, R. 20, 116, 143 and special pleading 1589
Scenes from Goethes Faust 2 and test of time 128, 137
Shakespeare, W. Vaughan Williams, R. 14, 1423
Hamlet 10, 118 A London Symphony 1445, 147
King Lear 10910, 118 Serenade to Music 86
Macbeth 102103
Othello 106 Wagner, R. 97, 116
Shostakovich, D. 3 The Mastersingers of Nuremberg 39
Second Ballet Suite 69 The Valkyrie 2, 119, 143
Fifth Symphony 133 Walford, David vii
Sibelius, J. Walton, William 68
Fourth Symphony 110 Wilde, Oscar 134, 175 (n. 15)
Fifth Symphony 108 Wittgenstein, L. 21, 36, 41, 45, 51, 101,
Sibley, Frank 137, 157 138, 150, 160, 174 (n. 10)
Strauss, R. 15, 106, 1467 works of art 2954, 1679
Gesang der Apollopriesterin, op. 33 no. analysis 7, 356
2 106 analysis and understanding 4954
Metamorphosen 115 disposable art 5
Stravinsky, I. 14, 91 experience of art 412
and expression 29, 45, 513, 92,
Tallis, Thomas 168 1679
Spem in Alium 46 functional theories of art 24, 35, 3743,
Toscanini, A. 84 45, 478
Tovey, D. F. 103104, 119, 140 hard cases 3035, 49, 54
types and tokens 613, 70, 71 historical accounts of the nature of
art 35, 434, 46, 48
underlabourer, the task of a philosopher of institutionalist theories of art 36, 44,
music 8 54; see also proceduralist theories
uniqueness in art 412 and interpretation 513
and meaning 29
value 1, 212, 12369 proceduralist theories of art 35, 367,
aesthetic judgement compared with 40, 45, 478; see also institutionalist
value-judgements in ethics 15260 theories of art
and the canon 125 recognitional accounts of the nature of
and causality 14950 art 35, 447, 48
and consensus 136, 1378, 148 understanding and analysis 4954

185