Contemporary Music Review,

1994, Vol. 10, Part 2, pp. 35-48

Reprints available directly from the publisher
Photocopying permitted by license only
9 1994 Harwood Academic Publishers GmbH
Printed in Malaysia
Defining Ti mbre- Refining Timbre
De n i s S ma l l e y
Music Department, City University, London, UK
Timbre is defined as the attribution of spectromorphological identity. Electroacoustic music experience,
particularly acousmatic music, questions the viability of a notion of timbre. Of primary significance is
the traditional linking of timbre to the source and cause of a sound: the concept of source-cause texture
is introduced to define this link. The ambiguous relationship between pitch and timbre indicates that
timbre cannot be defined as that part of the sound which is not pitch. Problems in establishing the
existence of sonic identities within the musical context, and of maintaining the coherence of identities
are discussed, opening up questions relating t o musical discourse. Two types of discourse -
transformational and typological - are defined, leading to the notion of generic timbre. In a
spectromorphological music where defining identities becomes problematic it becomes impossible to
disentangle timbre from discourse: here the notion of timbre has a limited viability.
KEY WORDS timbre, sources and causes, identity, registration, discourse, typology, transformation.
Introduction
Wh a t is t i mbr e ? Wh e n is it me a n i n g f u l t o us e t he t e r m? Ho w us e f ul is a c onc e pt
o f t i mbr e i n t he c ont e xt o f el ect r oacous t i c mus i c ? De f i ni ng t i mbr e is a h a z a r d o u s
ope r a t i on. " Ta l ki ng a b o u t t i mbr e , " s a ys Phi l i ppe Ma n o u r y " i s as del i cat e as t al ki ng
a b o u t t a s t e " (1991). Mor e ove r , t i mbr e c a n n o t eas i l y be p i n n e d d o wn . Cl a ude
Ca d o z ( 1991) goes so f ar as t o st at e t ha t a n y di s c us s i on of t i mbr e p r o v o k e s de ba t e
o n t he v e r y n a t u r e of mus i c, whi l e J ean- Bapt i s t e Barri 6re ( 1991) f i nds i n t i mbr e
" t h e i nevi t abl e b r e a k i n g - p o i n t of e v e r y mus i c a l di s c us s i on a n d e v e r y c ompos i -
t i ona l c onf r ont a t i on. I t r e pr e s e nt s o p p o s e d n o t i o n s of mus i c, o p p o s e d aes t het i c
pos i t i ons . ' " I n d e e d , t h e b o o k wh i c h i nc l ude s t he s e wr i t er s is cal l ed Timbre,
mdtaphore pour la composition. You c a n n o t get mo r e a l l - e mbr a c i ng t h a n t hat .
On e t r oubl e wi t h t i mbr e is t ha t it exi s t ed b e f o r e el ect r oacous t i c mus i c, a n d we
t he r e f or e h a v e t o s p e n d mu c h t i me a n d i nt el l ect ual e n e r g y e x t e n d i n g or c o mb a t i n g
n o t i o n s wh i c h we r e n o t neces s ar i l y d e s i g n e d f or t he mu s i c we ma ke . A s e c o n d
t r oubl e is t h a t we n o w k n o w t oo mu c h a b o u t t i mbr e. I t is o n e of t h o s e s ubj ect s
wh e r e t he mo r e y o u r e a d a n d t he mo r e y o u h a v e h a n d s - o n c ompos i t i ona l
e xpe r i e nc e t h e mo r e y o u k n o w, b u t i n t he pr oces s y o u b e c o me l ess abl e t o gr as p
its es s ence.
Whose Timbre?
Th e t hi r d t r oubl e is t h a t t i mbr e me a n s di f f e r e nt t hi ngs t o di f f er ent peopl e. " Wh o s e
t i mb r e ? " we mi g h t ask. For ma n y t r adi t i onal mus i c i a ns t he o f t - q u o t e d def i ni t i on of
t h e Ame r i c a n Na t i o n a l St a n d a r d s I ns t i t ut e, i n s pi t e o f its cur i ous negat i vi t y, ha s a
cer t ai n val i di t y:
35
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8
36 D. Smalley
" t ha t at t ri but e of audi t or y sensat i on i n t erms of whi ch a l i st ener can j udge
t hat t wo sounds similarly pr esent ed and havi ng t he same l oudness and pi t ch
are di ssi mi l ar. "
At least it implies t he traditional, pervasi ve not i on ( however t echni cal l y incorrect
it ma y be) of t he i nst r ument al not e as compri si ng a pi t ch wi t h timbre, and it
evokes t he not i on of a source even if it is not specifically ment i oned.
For t he cont empor ar y i nst r ument al composer (I am t hi nki ng part i cul arl y of
spectral appr oaches to i nst r ument al composition) t i mbre is an ext ensi on of
har mony, or vice versa. The composer uses spectral anal ysi s as a basis for
concept ual i si ng t he rel at i onshi p bet ween pi t ch and s ound qualities, and at t empt s
to negot i at e fl uent border crossings bet ween t he t wo.
A t hi r d appr oach to t i mbre concerns t he details of its mul t i di mensi onal i t y.
Thr ough research publ i cat i ons and t hr ough electroacoustic composi t i onal experi-
ence we have become ver y awar e of t he mul t i pl e variables whi ch det er mi ne
t i mbral i dent i t y. And we have also become concer ned to di fferent i at e wha t is
acoust i cal l y pr esent i n sounds f r om what is psychoacoust i cal l y pert i nent . The
maps pr esent ed by such aids as spectral anal ysi s or sonogr ams are one t hi ng;
fi ndi ng a safe percept ual wa y t hr ough t hese territories is anot her. The mul t i di -
mensi onal compl ex is not wi t hout probl ems. It can reveal to us t he " wh a t " of
timbre, and expl ai n t he " h o w, " but t hat does not mean we can aural l y separat e out
t he di mensi ons i nvol ved nor focus our ears on t hei r details, even t hough we are
so sensitive to t hei r mi nor behavi our al changes. In appl yi ng t hei r awar eness of
mul t i di mensi onal t i mbral feat ures, composers, ai ded and abet t ed by t echnol ogy,
are of t en absor bed wi t h concepts, met hods and t echni ques: t he l i st ener' s appre-
hens i on of t i mbral val ues cannot si mpl y be equat ed wi t h t he l aunchi ng of
mul t i di mensi onal at t ri but es by t he composer.
This bri ngs me to a f our t h ans wer to t he quest i on, " whos e t i mbr e?" The
ever yday l anguage of qual i t at i ve descri pt i on is accessible to ever yone. It is closely
allied to t he " mat t er " of sound. Terms like br i ght / dul l , compact / s pr ead, hol l ow,
dense, ma y be vague, and vague t hey are dest i ned to r emai n since t hey are
qualities, but t hey have t he advant age of an i mmedi at e, compr ehensi bl e i dent i t y
and are t herefore not to be scoffed at. They are verbal signs t hat essential qualities
have been recogni sed.
Ti mb r e a n d So ur c e
Havi ng i nvoked f our at t i t udes to t i mbral percept i on I s houl d n o w hazar d a
pr el i mi nar y defi ni t i on. I have freel y adapt ed a defi ni t i on by Mi chel Chi on. 1
Ti mbre is a general , sonic phys i ognomy t hr ough whi ch we i dent i f y sounds
as emanat i ng f r om a source, whet her t hat source be actual, i nf er r ed or
i magi ned.
Prior to t he electroacoustic peri od musi c al ways i nvol ved i dent i fi abl e sources. The
l i st ener coul d spont aneousl y l i nk s ound bot h to a s oundi ng body and a human,
1 (Chion 1986: 7):."... le timbre, n'~tant rien d'autre, en effet, que la physionomie g~n~rale qui
nous fait identifier un son comme 4manant d'un instrument d~termin~ ou (plus g6n~ralement)
d'une source d6terminG qui peut ~tre imagin4e ou imaginaire."
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8
Defining Timbre - Refining Timbre 37
physical cause. Gesture not only activated the source but could, through breathing
and bow control, and techniques of touch, maintain and continue the sounding of
the vibratory system. Traditionally, therefore, there is an inherent, culturally
imbedded, stable source bondi ng i n music. Source bonding is the term I use to
encapsulate
the natural tendency to relate sounds to supposed sources and causes, and to
relate sounds to each other because they appear to have shared or associated
o r i g i n s .
All those attributes traditionally packaged under timbre - the nuancing and
articulation of note-objects and phrase-shapes, and the control and variation of
tone, including the roughness/smoothness of note-grain - could be traced to
source-cause interactivity.
In electroacoustic music, however, sources and causes are many, varied, evident
or ambiguous, actual or implied, unknown or unknowable: we can perhaps detect
traces of cause or source but realise that neither can exist in reality. In instrumental
and vocal music source and cause are known in advance of listening, but in
electroacoustic music they are unveiled by the composer and are discovered (or
not) by the listener in the course of the work, and within that single work there
may be a great variety of source-cause shadings and strategies. One of the great
interests in electroacoustic music is the adventure of bonding play, which I regard
as an inherent perceptual activity. Listeners may share source bondings when they
listen to electroacoustic music, but they may equally have different, personalised
bondings including those never intended or envisaged by the composer. Widely
ranging bondings are inevitable in musics which are not primarily weighted
towards fixed pitch-intervals and the instrumental note, and they can occur in
what might be considered the most abstract of works.
If source and cause are unstable, illusory or non-existent does this mean that my
definition of timbre crumbles? It is certainly too restrictive because apprehending
source and cause is not the only identification strategy. Source bonding is extrinsic -
it refers to sounding experiences outside the work. But extrinsic links also involve a
wide range of real and imagined non-sounding phenomena. Thus source bonding is
part of a more comprehensive representational fabric which I shall refer to as the
ext ri nsi c mat ri x. T h e word "bonding" evokes a binding engagement and kinship
between listener, musical context, and source-cause searching which i s a
spontaneous activity for most listeners. However, the wider extrinsic matrix is a less
concrete, more inventive and more perishable medium of exchange between work
and listener. I can only indicate very generally the content of the extrinsic matrix in
this paper. Timbre is concerned with the temporal unfolding and shaping of sound
spectra, in other words with spect romorphol ogy. One concern of spectromorphology
is motion and growth processes, which are not exclusively or even primarily sonic
phenomena. Energy, which is inherent in spectral motion, belongs both to sounding
and non-sounding experience; it is linked not only to motion in general but to
human gesture, whose sounding manifestations are implicated in the causality of
source bonding. Motion, growth and energy can be regarded as having a sonic
reality but they can also be interpreted metaphorically and symbolically.
Since the notion of source is too restrictive we must invoke a more abstract
concept of timbre which can include the source. My definition therefore becomes
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8
38 D. Smalley
a general, sonic phys i ognomy whos e spect romorphol ogi cal ensembl e permi t s
t he at t ri but i on of an i dent i t y.
We are goi ng to have to delve i nt o t hi s spect romorphol ogi cal ensembl e, but first
we mus t go mor e deepl y i nt o t he cont ext ual envi r onment whi ch est abl i shes t he
condi t i ons for identification. That means ret urni ng to t he nor ms est abl i shed
t hr ough i nst r ument al source-cause.
Source- cause Text ure
Robert Cogan (1984) sums up t he significance and scope of cont ext for i nst ru-
ment al i dent i t y:
" . . . t h e essential sonic feat ures of any musi cal i nst r ument are not finally to
be di scovered ei t her i n t he practice r oom or i n t he anechoi c chamber, or even
i n t he most sophi st i cat ed statistical anal yses of i nst r ument al s ound i sol at ed i n
t hose places. Rat her, t he essential sonic feat ures of a ny musi cal i nst r ument
are to be f ound i n t he s um total of its st ruct ural sonic cont ri but i ons to musi cal
contexts. These feat ures will be reveal ed by anal yzi ng t he f unct i ons and
relations of t hese contexts and of t he i nst r ument s in creat i ng t he m. " (p. 185)
The i dent i t y of t he i nst r ument is exposed t hr ough our experi ence of wha t I shall
call s o u r c e - c a u s e l e v e l s . In this i nst ance let us say t hat t he source is a violin, and
t hat t he cause is t herefore specific t ypes of gest ural action of a violinist. The l owest
source-cause level is t he i m m i n e n t l e v e l - t he ongoi ng, intrinsic musi cal cont ext
wher e we encount er t he i nst r ument . Thr ough t he unf ol di ng of t he musi cal style
we become awar e of r e g i s t r a t i o n - t he art i cul at i on and pl ay of not e-obj ect s and
t hei r chai ni ng i n phr ases over a cont i nuum of registers. The second level is t he
c u m u l a t i v e l e v e l , whi ch reaches back to i ncl ude our previ ous experiences of
violin-sources in t he hands of ot her violinist-causes who articulate bot h t he same
musi c and ot her genres and styles. I recognise t hat at bot h t he i mmi nent and
cumul at i ve levels t he violin ma y be associ at ed wi t h an accompanyi ng or collab-
orat i ng source-cause. The t hi r d level, t he e x t e n d e d l e v e l , i ncl udes ensembl e
contexts and ext ends t he source-cause base to i ncl ude t he i mmedi at e f ami l y of
st ri nged i nst r ument - sour ces. The final level, t he d i s p e r s e d l e v e l , spreads over t he
wi dest possi bl e r ange of source- cause to i ncl ude all bowed and pl ucked
i nst r ument s, for exampl e viols and rebecs and t he i nst r ument s of ot her cultures.
The four levels create a pot ent i al l y enor mous reference field whos e t ot al i t y I
shall call s o u r c e - c a u s e t e x t u r e . Its cont ent encompasses our total, l ong- t er m
experi ence of i mmi nent registrations, possi bl y i n ma ny musics, across space and
time. The i dea of source-cause t ext ure is all t he mor e per t i nent t oday wh e n
t hr ough recordi ngs we have l i st eni ng access to a vast cross-cultural source-
cause repert ory. In passi ng we shoul d not e t hat bot h mi xed and acousmat i c
electroacoustic wor ks t ap existing di spersed levels in order to expand t radi t i onal
i nst r ument al source-cause.
In electroacoustic musi c as a whol e, however , we do not fi nd such a definable,
hierarchical basis for est abl i shi ng t he source-cause aspect of t i mbral i dent i t y. Take
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8
Defining Timbre - Refining Timbre 39
wat er as an exampl e of an i dent i fi abl e source. Ther e ma y be a cause - wat er
act i vi t y can appear t o be sel f-act i vat ed or can be act i vat ed by h u ma n or ot her t ypes
of i nt er vent i on, as i n t he splash, f or exampl e. In a musi cal wor k, wat er , like a ny
s oundi ng source, can exist on bot h t he i mmi nent and cumul at i ve levels. But t he
ext ended a nd di sper sed levels cannot be i dent i fi ed because we cannot est abl i sh
a ny musi cal r eper t or y be yond t he cumul at i ve level. Our wat er is not a sour ce-
cause in t he t radi t i onal , i nst r ument al sense - it does not , f or exampl e, f unct i on i n
st abi l i sed ensembl es. However , it can be ext ended or di sper sed be yond t he
cumul at i ve l evel by r ef er r i ng out si de musi cal wor ks t o t he ext ri nsi c mat ri x. In
i nst r ument al musi c sour ce- cause t ext ur e pr ovi des a solid, cul t ural base. I n
el ect roacoust i c musi c, as far as c o mmo n l y i dent i f i abl e non- i ns t r ument al sour ce-
causes are concer ned, we mi ght say t hat an equi val ent base exists because t he
s oundi ng area of t he ext ri nsi c mat r i x can pr ovi de a subst i t ut e f or t he ext ended a nd
di sper sed l evel s of sour ce- cause t ext ur e, even if we cannot pi n down specific l evel s
s har ed by al l non- i ns t r ument al source-causes. In ot her wor ds we ha ve ma n y
exper i ences of wat er s ounds be yond t he cumul at i ve l evel , i n nat ur e and i n cul t ure;
we can all cl earl y rel at e to expansi ve shi ft s i nt o t he extrinsic mat r i x i n t he case of
ma n y envi r onment al sounds, a nd also t o ot her source-causes such as hi t t i ng,
scr api ng or r ubbi ng sol i d materials. We mi ght say, t her ef or e, t hat i n t he case of real
source-causes, r egi st r at i on at t he i mmi ne nt l evel has a si gni fcant hi gher - l evel
cul t ural basis.
Wher e sour ce- causes are i nf er r ed, i magi ned or non- exi st ent t he shi ft t o t he
ext ri nsi c mat r i x also occurs be yond t he cumul at i ve level, but t he emphasi s will
l ean mor e i n f avour of t he n o n - s o u n d i n g area. Whe n t he non- s oundi ng area is
ent er ed we are no l onger on secure c ommon gr ound because sour ce bondi ng is no
l onger operabl e: we cannot i dent i f y real sources and causes. Spect r omor phol ogi cal
at t ri but es a nd i deas t he n evoke a nd suggest non- s oundi ng subst i t ut es f or t he
ext ended a nd di sper sed levels, whe t he r t hese be to do wi t h mot i on, t ypes of
behavi our , spat i al exper i ence, ener get i c phe nome na , psychol ogi cal t ensi ons, a nd
so on. I bel i eve t hat such ideas, i nt angi bl e as t hey mi ght be, are a si gni fi cant means
of art i cul at i ng t hat necessary, shar ed, hi gher - l evel cul t ural basi s for a musi c wi t h
non- exi st ent source-causes.
Lest a nyone shoul d t hi nk t hat t he ext ri nsi c mat r i x is i r r el evant to i nst r ument al
sour ce- cause - t hat i nst r ument al musi c refers onl y to itself - it needs to be
under l i ned t hat be hi nd t he causal i t y of i nst r ument al gest ur e lies bot h a br oader
exper i ence of t he physi cal i t y of gest ur e and its pr opr i ocept i ve t ensi ons, a nd a
deeper , psychol ogi cal exper i ence of gest ure. In i nst r ument al musi c h u ma n - b o n d e d
sour ce- cause t ext ur e represent s t hese pr i mal levels of gest ure f ound i n t he ext ri nsi c
mat ri x. I n el ect roacoust i c musi c whe r e sour ce- cause links are sever ed, access t 6
any deeper , pri mal , t ensi l e l evel is not medi at ed by sour ce- cause t ext ure. That is
wha t makes such t ypes of acousmat i c musi c difficult for ma n y t o grasp. I n a cert ai n
physi cal sense t her e is not hi ng t o gr asp - sour ce- cause t ext ur e has evapor at ed.
However , nei t her t he i mmi nent not t he cumul at i ve levels i n el ect roacoust i c
musi c can be consi der ed equi val ent s of i nst r ument al music. At t he cumul at i ve
l evel of el ect roacoust i c wor ks, t he di ver si t y of sour ce- cause a nd t he vari et i es of
spect r omor phol ogi cal phys i ognomy are vast. I nst r ument al musi c cannot be
consi der ed compar abl e. Even t he i mmi nent l evel of i ndi vi dual el ect roacoust i c
wor ks is concer ned wi t h t he regi st rat i on of mul t i pl e spect r omor phol ogi es ma n y of
whi ch ma y be uni que t o a single wor k. It is to t he i mmi nent l evel , to t he i nt ri nsi c
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8
40 D. Smalley
spect romorphol ogy of the musical wor k t hat we must now turn in the search for
t he identity of timbre, but before confronting this probl em I must discuss pitch in
electroacoustic music.
The Ti mbre of Pi t ch - t he Pi t ch of Ti mbr e
The basis for the traditional notion of timbre relies on the conceptual possibility of
separating pitch from timbre. The heritage of music writing has t ended to
encourage abstract approaches to pitch whi ch, at worst, becomes alienated from
the spectromorphologies responsible for creating and shaping it. Once tonality and
intervallic pitch are no longer regarded as the predomi nant carriers of musical
messages, pitch and timbre can cohabit in a spectromorphological music wher e t he
ear has opportunities for shifting in and out of pitch values. In electroacoustic music
we are concerned not onl y wi t h the received not i on of t he timbre of pitch but also
the pitch wi t hi n timbre, and moreover, t i mbr e wi t hout pitch. Pitch is present even
when not perceived. Perhaps it is resting, hi dden deep in a spectromorphology,
awaiting possible attention, a moment when, for example, the context might change
so that perceptual focus becomes directed t owards what was a sleeping attribute.
The salience of pitch therefore becomes contextual in electroacoustic music.
Timbre rather t han being that part of t he sound whi ch is not pitch, encompasses
the inherent qualities of the whol e sound. Perceived pitch is one of these i nherent
qualities, but the t erm "t i mbre" cannot alone be expected to shoulder the bur den
for all t he others. Once split open, the timbre complex is not a package of distinctly
separable parameters, but a field of interactive behaviour, pl ayed out in the
relations bet ween spectral space and temporal change. Under such conditions does
not the concept of timbre become so general as to be meaningless?
Regi s t rat i on
If timbral identity in electroacoustic music is so heavily reliant on the single
i mmi nent level t hen registration wi t hi n t he wor k is the key. It is registration whi ch
enables us to connect to source-cause texture and to t he extrinsic matrix.
Registration in instrumental music I defined as "t he articulation and pl ay of not e
objects and their chaining in phrases over a cont i nuum of registers." Registration
in electroacoustic music cannot be so systematic because there is no consistent
low-level note-object, no consistent phraseological practice, nor is there t he same
registral continuum. If there is no underlying, pitch-base syst em shared by
spectromorphologies t hen registration can no longer be controlled by the exposi-
tion of pitch registers, but instead becomes concerned wi t h variable spectromor-
phological attributes. With such an all-embracing openness we are in danger of
losing our way, as registration starts to per meat e the very nat ure of the musical
discourse perhaps to the point of becomi ng synonymous wi t h it. Robert Cogan
(1984) outlines t he general requirements of registration:
"'To be musically and sonically significant, a sonic element must reveal itself
in a musical context. Indeed it requires not onl y a presence but, usually,
reiteration and amplification (used here in t he structural sense). Some phoni c
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8
D e f i n i n g T i m b r e - R e f i n i n g T i m b r e 41
propert i es are so weakl y (or accidentally) pr esent i n an i nst r ument al sound
and so unampl i f i ed i n a gi ven cont ext t hat t hey di sappear i n t he cont ext or
do not , at any rate characterize it. They do not f u n c t i o n i n t he part i cul ar
cont ext . " (p. 146)
We have t herefore to di scri mi nat e t he i nci dent al f r om t he funct i onal , to decide
whi ch spect romorphol ogci al at t ri but es mat t er, and we have to discover t hi s anew
i n t he i mmi nent level of each electroacoustic work. A pai r of rel at ed variables
under pi ns our at t empt s at det er mi ni ng identities:
1. t he coherence and st r engt h of spect romorphol ogi cal i dent i t y;
2. t he dur at i on needed to establish existence and expose registration.
Existence
There is no st andar d dur at i on for est abl i shi ng existence. In electroacoustic musi c
quite l ong evol ut i on t i mes ma y be needed to establish existence and i mpose
i dent i t y. Take an active wat er t ext ure as an example. Al t hough we ma y recognise
i mmedi at el y t hat it is wat er it takes l onger to det er mi ne what we mi ght refer to as
its i nner "t i mbr al " detail because it is not t he odd globule but t ext ural behavi our
whi ch establishes its i mmi nent i dent i t y, for exampl e its resonances, noise cont ent ,
and pi t ch-st reams or contours. I dent i f yi ng a crude source-cause is one thing;
penet r at i ng its behavi our al detail is anot her. If such a t ext ure wer e not a
recognisable source-cause t hen ext ended dur at i on coul d be even more i mport ant .
Wi t h this wat er t ext ure t he est abl i shi ng of existence and t he evol ut i on of its
registration are i nseparabl e, and t he at t ri but es whi ch we discover ma y prove to be
significant or not as t he i mmi nent level proceeds.
Textural behavi our can t herefore be an exampl e of what I shall call a regi st rat i on
ge ne r at or - an ongoi ng cont ext whose cont i nui t y is necessary to est abl i sh existence
and registration. Cont i nuous evol ut i on and cont i nuous t r ansf or mat i on can also be
consi dered regi st rat i on generat ors. They can be ongoi ng contexts wher e t here is no
border bet ween existing and t ransformi ng, existing and evolving. In ot her words,
change becomes f undament al to existence and registration: we appr ehend i dent i t y
as t he consequence of change.
There are t ypes of spect r omor phol ogy whos e existence can onl y be est abl i shed
aft er a certain evol ut i on t i me because t he compl et i on or partial compl et i on of a
pat t ern is i nt egral to i dent i t y. Mot i ons based on rot at i on are exampl es of
spect romorphol ogi es whos e t i mbre (if t hat is t he ri ght word) is embodi ed in t he
spectral changes over at least one rotation-cycle. This is an i nst ance of a
registration pat t er n wher e it becomes impossible, or at least percept ual l y i mpl au-
sible, to di st i ngui sh bet ween what mi ght be consi dered its t i mbral mat t er on t he
one ha nd and its shor t - t er m evol ut i on on t he ot her.
Coherence
The not i ons of existence and regi st rat i on i mpl y a certain coherence, ot herwi se
i dent i t y woul d not be feasible. Coher ence i n t he case of i nst r ument s is usual l y
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8
42 D. Smalley
associ at ed wi t h spectral fusi on, whi ch i n t ur n is associated wi t h harmoni ci t y.
Jean- Cl aude Risset (1991) considers fusi on a necessar y t i mbral propert y: 2
" The not i on of t i mbre i mpl i es fusi on; it corresponds to t he s ound qual i t y of
an ensembl e of component s i nt egr at ed i n an audi t i ve ent i t y whi ch is
assi gnabl e to a single source, whet her real or vi rt ual . " (p. 257)
Since f usi on is so oft en closely al i gned wi t h har moni ci t y, and since I fi nd it t oo
rigid a word, I prefer t he t erm "' i nt egrat i on. " That also al l ows us to talk of an
i nt egrat i on-di si nt egrat i on cont i nuum.
Fr om a t heoret i cal poi nt - of - vi ew i nt egr at i on/ di si nt egr at i on possesses a spectral
and a morphol ogi cal di mensi on - spectral space and its i nt ernal di st ri but i on on t he
one hand, and shapi ng over t i me on t he ot her. Int egrat i on means t hat wi t hi n a
sonic phys i ognomy t he di st ri but i on of spectral component s in spectral space, and
t hei r behavi our over t i me shoul d not be such t hat a component or sub-group of
component s can be percei ved as an i ndependent ent i t y. Cert ai nl y a fairly hi gh
degree of i nt egrat i on is necessary if s omet hi ng called t i mbral i dent i t y is to be
est abl i shed, but wher e are t he borders bet ween t i ght fusi on, a looser i nt er depen-
dence, and a di ssol ut i on t owar ds i ndependence? When is a t i mbre a t i mbre and
wh e n is it a collection of t i mbres? Whe n does t he phys i ognomy crack?
The course of even a relatively short spect r omor phol ogy can travel bet ween
i nt egr at i on and di si nt egrat i on. For exampl e, usi ng t he uni f yi ng weapon of t he
at t ack phase I can i mpose a causal coherence on t he onset of a spect r omor phol ogy;
as it cont i nues I can i nt er vene i n t he t empor al evol ut i on of t he spectral
component s, per haps refocusi ng t he spectral space by wei ght i ng at t ent i on t owar ds
certain component s, per haps separat i ng some out, per haps clearing gaps i n
spectral space, per haps dr awi ng at t ent i on to intervallic pitch, per haps secreting i n
ne w component s. In ot her words ne w objects, st rands or gestures can emerge i n
t he mul t i -l evel pl ay of hierarchies as t hey expand and contract i n t he course of
electroacoustic musi cal discourse. We mus t also not forget t hat aural acui t y,
at t ent i on, and t he volition of focus are crucial prot agoni st s in t hi s unvei l i ng of
i dent i t es, part i cul arl y i n cont ext s of fl eet i ng ephemer a.
It seems t hat as l ong as a spect r omor phol ogy is percei ved as a coher ent ent i t y,
t hen " t i mbr e" is a wor d we mi ght use to encapsul at e its global phys i ognomy. Thi s
is t he equi val ent of sayi ng it has an i dent i t y, or t hat it has a source-cause since
source-cause i mpl i es coherence. But t he pr obl em is to i dent i f y coherence! Pl ayi ng
wi t h i nt egrat i on and di si nt egrat i on is at t he ver y hear t of electroacoustic musi cal
discourse, a di scourse whi ch becomes spect r omor phol ogy itself once t he t i mbre
compl ex is split open, a discourse wher e t he not i on of t i mbre can at one mome nt
per haps be grasped, but at t he next it evaporat es.
Di scourse St abi l i t y and Vari abi l i t y
This supposedl y coher ent i dent i t y whi ch mi ght be called t i mbre is not necessari l y
an ent i t y whi ch can be pri sed apart f r om t he soundi ng context. A t i mbre ma y be
2 "La notion de timbre implique la fusion; eUe correspond ~ la qualit~ sonore d'un ensemble de
composantes int~gr~es en une entit~ auditive et assign6es ~ une rhyme source sonore r~eUe ou
virtuelle."
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8
Defining Timbre - Refining Timbre 43
a discrete object whi ch can be separat ed out f r om its context, but it can also be a
cont i nui t y whi ch is i nt er t wi ned or bl urred wi t h ot her cont i nui t i es or whos e st art
and end cannot be discerned. We can t herefore talk about t he concept of t i mbr al
l evel , whi ch concerns t he rel at i onshi p bet ween t wo cont i nua.
1. durat i on: shor t - t er m ent i t y . . . . . . . . . . . . . . . . . l onger t er m evol ut i on
2. s epar abi l i t y: discrete object . . . . . . . . . . . . . . . . . . . . . cont i nui ng cont ext
The cont i nui ng cont ext coul d even encompass t he whol e wor k i f t he wor k is itself
a cont i nui t y, or if we coul d ascribe a general t i mbral et hos to t he work as a whol e
- a t i mbre of t i mbres if you like.
Timbral level i n t radi t i onal not e- based musi c is quite simple. The not e is t he
l owest level and it is art i cul at ed by an i nst r ument al source. Form devel ops f r om
not e articulations. In electroacoustic musi c cont i nui ng cont ext s resist and deny
low-level segment at i on. Thus once t i mbral level ceases to be clearcut we cannot
separat e t i mbre and discourse: tirnbral at t ri but es become woven i nt o t he spectro-
morphol ogi cal fabric.
But let us assume t hat in t he openi ng stages of t he i mmi nent level coher ent
existences have been est abl i shed, whet her t hr ough t he persi st ence of existence,
t hr ough separability, t hr ough source-cause coherence, or t hr ough l i mi t ed registra-
tion. Let us also assume t hat discourse is about mai nt ai ni ng and devel opi ng some
of t hese identities, and t hat we wi sh our t r ansf or med identities to s how t hei r roots.
What next?
Transformational discourse
I recognise t wo approaches to musi cal discourse whi ch concern t he appl i cat i on of
t hose at t ri but es associated wi t h timbre. The first is t ransf ormat i onal di scourse,
wher e an i dent i t y is t r ansf or med wh{le ret ai ni ng significant vestiges of its roots (i.e.
an i dent i t y-base i n t he i mmi nent level, or a st rong extrinsic identity). To achi eve
this, certain at t ri but es mus t r emai n stable whi l e ot hers vary. Digital synt hesi s and
t r eat ment t echni ques per mi t det ai l ed and subt l e spect romorphol ogi cal i nt er vent i on
al l owi ng us to create and t r ansf or m wha t wer e t i mbre' s mul t i di mensi onal
attributes. Spect romorphol ogi es t herefore become mal l eabl e, and t he composer has
to decide how to ret ai n stability. The st abi l i t y/ var i abi l i t y rel at i onshi p is a difficult
area of der i si on- maki ng related to t he coherence of discourse. There is so muc h
choice, and in t ransformat i onal discourse t here are ma n y ways of achi evi ng and
losing this equi l i bri um. Here are a f ew exampl es of pr eval ent t echni ques:
1. st ret chi ng or cont ract i ng i n t i me whi ch does not t hr eat en t oo much t he aural,
evol ut i onar y i nt egrat i on of spectral component s. Compress too much and you
dest roy i dent i t y; stretch too much and you dest roy, for example, at t ack
i dent i t y.
2. changes i n spectral densi t y by t hi ckeni ng, or by spreadi ng in spectral space,
whi ch are percei ved as addi t i on - a gr owt h process.
3. changes i n spectral wei ght i ng (e.g. brighter, or emphasi zi ng i nt ernal intervallic
pitch) wi t hout al t eri ng t he essential cont ent (i.e. f r equency spacing) of t he
spectral envel ope.
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8
44 D. Smalley
4. variation in or reshapi ng of a morphol ogy (for example, the attack impact) in
such a way that it does not affect t he identity of t he continuing body of a
spectromorphology. This implies either that
(a) t he sound is a strong spectromorphological t ype well known t hrough
extrinsic experience (e.g. a bell, t he voice); or
(b) that t he base-spect romorphol ogy has been previ ousl y announced and
that if this was some time previ ousl y t hen it was striking enough in
context for relevant features to be memorable; if the base-
spect romorphol ogy is less memorabl e it woul d need to be relatively
contiguous to t he present event.
Of course if a morphological feature (e.g. t he attack) is t he dominant, identifying
det ermi nant t hen interference can mean a change of identity: source-cause itself is
perceived to be transformed, and bondi ng pl ay is involved.
Al t hough in t hese four exampl es I am assuming the retention of identity, and
although the spectromorphological features ment i oned are commonl y attributed to
timbral multidimensionality that does not mean that we woul d necessarily choose
t he wor d "t i mbr e" to describe t hese t ypes of relations in transformational
discourse.
I not e in passing that memorability and contiguity are t wo i mport ant tools of
discourse. It is undoubt edl y true that pitch relations and source-causes are t he
most easily memorabl e sonic phenomena, and that the probl em wi t h multiple
spectromorphological attributes is t hat we do not know whi ch ones are to be
relevant in t he discourse. This is wh y electroacoustic music whi ch avoids pitch
phenomena and strong source-cause references will be most frequently concerned
wi t h contiguous relationships. Cont i nui ng textural contexts and cont i nuous
transformation are prime examples of contiguous devel opment . Non-cont i guous
discourse can onl y be highly devel oped wi t hi n the precise and detailed memora-
bility of culturally i mbedded pitch and rhyt hmi c systems. In my opinion, expecting
to appl y the criteria of pi t ch/ r hyt hm- based discourse to discourses whi ch i nst ead
exploit the disparate attributes of spectromorphological multidimensionality, is
misguided.
Typological discourse
The second t ype of discourse is typological discourse. Identities are recognised as
sharing timbral qualities but are not regarded as being descendant s of the same
i mmi nent identity - t hey do not possess a common identity-base. Typological
discourse is associative. While each i mmi nent level will present different mani-
festations of a timbral type, there is also a cumulative level whi ch reaches across
works. The larger groupings of the timbres of typological discourse can be called
generic timbres. Electroacoustic music has made possible the expansion and
devel opment of certain generic timbres whi ch have become its idiomatic property.
The timbres of noise are a particularly striking example. While noise is present
in instrumental and vocal timbre it onl y really takes off as a timbral t ype once it
meets up wi t h t he electroacoustic medi um. One preval ent t ype is granular noise.
Granular noise is textured impulses of varying consistencies and resolutions
ranging from roughness t hrough granularity to grit, wi t h an internal behavi our
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8
Defining Timbre - Refining Timbre 45
which, while showi ng varying degrees of regularity or irregularity, remains a
coherent identity. Beyond roughness granular noise smoot hs out, and somewher e
beyond grit the texture starts to fragment. Granular noise has strong source-cause
bondl ngs whi ch are extremely wi de- ranging - sea, wat er in general, wind, foliage
(living or dried), fracturing materials (wood, stone), motions of friction, unvoi ced
vocal behaviours, certain t ypes of breat hi ng and fluid congestion, t ypes of
mechanical noise, static interference. What category of sonic phenomenon is not
implicated? It is undoubt edl y true that these source-cause bondings, whet her
blatant or subliminal, account for t he popul ari t y of granular noise in that t hey can
initiate prolific bondi ng pl ay in musical discourse particularly in their inherent
ability to dr aw together humani t y, nat ure and the elements. Not e that there is a
powerful spatial dimension involved, at once personal and environmental. Not e
also granular noise' s tensile properties: it can be extremely relaxing or penetrat-
ingly tense. While it is commonl y regarded as non-pi t ched we can say that the
pitch of timbre can be activated in typological discourse. We shoul d also not forget
that resonance can be part of granular noise. Resonance embodi es the latent pitch
inside or behi nd granular noise, whi ch can become a partner in the surroundi ng
discourse. The most significant aspect of noise typologies is that t hey satisfy t he
most important criterion of all - t hey already lie within our cultural experience.
The discourse function of granular noise can be a decorative or subordinate
strand or trace injected into a spectromorphology, or it can be at centre stage. As
far as timbral level is concerned granular noise is not a discrete phenomenon but
a texture wher e higher level patternings are t he structuring aspects of its
physi ognomy. No doubt granular noise, if st udi ed in more detail woul d break
down into sub-categories - it is t oo unwi el dy as it is.
I n h a r mo n i c i t y is anot her generic timbre, and a vexing one because of its
disintegrative tendencies. This makes it difficult to pin down. At the ver y least
inharmonicity has initiated a different attitude t owards intervallic pitch, provoki ng
spectral spaces wher e we hear fixed pitches but wher e the intervals created are not
necessarily bearers of intervallic pitch functions. Inharmonicity has different
i mmi nent physi ognomi es related more to higher-level groupings in musical
contexts t han to individual objects. Who, for example, woul d confuse the style of
inharmonicity of Chowni ng wi t h t hat of Risset? But I do not yet know of any
secure met hod of categorising inharmonicities in such a way as to overcome the
unhel pful negativity of the term.
I hope I can safely say that wi t h generic timbres the term "t i mbre" is
appropriate. We can intuitively appr ehend t he contextual conditions determining
existence, registration, coherence, and source-cause texture at the imminent and
cumulative levels, spreading to t he extrinsic matrix at t he ext ended and dispersed
levels.
Transformational and typological discourse are not mut ual l y exclusive, and
moreover, tipping the scales either way can be a question of listening choice and
listening experience - listening experience is variable and subjective. The
difference bet ween t hem is really a question of whet her identities are more or less
specific - the less specific the identities, the more typological the discourse
becomes. And t hey can exist together. For example, a specific event subject to
transformations can also be a bearer of generic timbre. In fact I recognise six
i nt e r ac t i v e t ypes of eIect roacoust i c di scourse. The first three are primarily concerned
wi t h identities and how we interpret them. They are
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8
46 D. Smalley
source-cause discourse, which is concerned with the bonding play of specific or
inferred sounding identities.
transformational discourse
typological discourse
The second three discourses are concerned with relations among identities. They
are:
behavioural discourse - the changing states of identifies' cohabition/conflict and
dominance/subordination;
motion discourse - the relations of types of motion and growth and their directional
tendencies;
tensile discourse - how the previous five discourses together create formal tensions.
Two I ndi genous Ti mbres of Ti mbres
Before concluding I need to mention two types of timbre indigenous to
electroacoustic music. The first is processing timbre. Synthesis methods and
treatment processes often impose their own timbral flavour on musical contexts,
and indeed whole works. The sounds of what might normally be regarded as
different source-causes and spectromorphologies appear to be coated in the same
timbral veneer. At a lower level the detailed diagnosis of particular processing
timbres is likely to be the pastime of specialists - composers, researchers,
technologists - rather than other listeners. Individual processing timbres hover
between innovation, when they are exciting to behold, fashion, when usage
spreads, and clichG when they have been flagellated to the point of collapse. Who
can predict and delineate the border between excitement and torpor? The detection
of technological source-cause can be an undesirable interference in the musical
exchange between composer and listener. The composer's inventive skill is the
only panacea. Higher level veneers are often unnoticed by the composer, who has
been so immersed in spectromorphological detail, but they are easily picked up by
other listeners, whose aural approach is always more global.
Processing timbres are important in that they impart a characteristic sound to
electroacoustic music. Electroacoustic music has its timbres, which can have a
neutralising effect on sonic communication.
The second indigenous timbre of timbres is the loudspeaker. Our dependency is
total. Loudspeakers possess an inherent spectral weighting which affects the
hearing and the composition of musical timbre, and is itself a timbre. While on the
one hand we rely on timbral invariance for studio monitoring or home listening,
we have to remember that the loudspeaker is a variant. We harness comparative
fimbral variance to excellent effect in the multi-speaker diffusion systems. Boulez
(1987) called the loudspeaker "'a great anonymous pulveriser of sound that does
not measure up to the means which have been developed to create a new sonic
world." (p. 170) Even if this was true in 1985 (when he spoke these words) it is
no longer the case . . . . if you have enough money. We still need constantly to be
aware of how crucial the loudspeaker is for the electroacoustic arts and how its
fimbral collusion with global processing timbres can indeed rob the real timbres of
their individual physiognomy.
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8
Defining Timbre - Refining Timbre 47
Concluding remarks
I have been concerned not wi t h t he acoustic nat ur e of timbre, nor even, t r ut hf ul l y
speaki ng, wi t h its psychoacoust i c propert i es, but wi t h its appr ehensi on, i dent i t y
and f unct i ons i n musi cal contexts f r om t he l i st ener' s poi nt of vi ew. Above all, t he
use of t he label " t i mbr e" is a signifier t hat we have i dent i fi ed spect romorphol og-
ical, physi ognomi c at t ri but es to whi ch we can ascribe meani ng. We not onl y
recognise and i dent i f y qualities but we i dent i f y wi t h t hem. The concept of timbre,
t hen, evokes bot h real and symbol i c links bet ween our i dent i t y and sonic i dent i t y,
and i n electroacoustic musi c bet ween our i dent i t y and t hat of t he wi de- open
soundi ng and non- s oundi ng worl d out si de t he music.
Whe n source-cause slips i nt o hi di ng, i dent i t y becomes probl emat i c, even
impossible. Timbral qualities become phant oms and phant asms. Personal i sed
visions dur i ng t he i mmi nent level suggest circuitous, l abyr i nt hi ne trails to wi der
generi c identities and to t he extrinsic matrix. The spect romorphol ogi cal t races
whi ch inspire such trail-finding can no l onger be parcelled up i n general i sed
physi ognomi c packages. Ti mbre becomes t he t i mbre complex, and t he t i mbre
compl ex splits open to reveal di sparat e lives: t he t i mbral nucl eus dissolves, seepi ng
t hr ough musi cal discourse. In st ruggl i ng to mai nt ai n a t i mbral essence we are
i ndeed obl i ged to penet r at e t he ver y nat ur e of musi cal discourse.
Composi ng wi t h timbre, composi ng wi t hi n timbre, means conf r ont i ng and
enj oyi ng its dissolution. This can onl y real l y be ful l y pur s ued i n an a c o u s m a t i c
electroacoustic music. In contrast, advent ur ous cont empor ar y i nst r ument al music,
and works whi ch mix i nst r ument s wi t h an acousmat i c el ement , are root ed i n t he
umbilical securi t y of i nst r ument al source-cause coherence and directly appre-
he nde d sound- maki ng gesture. Thi s equat es not wi t h a bur ni ng desire to explore
timbre, but wi t h a hesi t ant reserve about cut t i ng loose i n order t o pur sue a freer
exploration. There is no reason wh y t he t radi t i onal not i on of t i mbre shoul d f ade
away. A not i on of musi cal t i mbre will al ways exist al ongsi de its di ssol ved
attributes. In keepi ng wi t h this ambi val ence I can summar i se t hi s discourse in six
words:
Ti mbre is dead. Long live timbre.
R e f e r e n c e s
Barri6re, J.-B. (1991). Introduction. In Le timbre, m~taphore pour la composition, J.-B. Barri~re, editor,
pp. 11-13. Paris: IRCAM/Christian Bourgois.
Bayle, F. (1993). Musique acousmatique: proposi t i ons. . , positions. Paris: I NA- GRM/ Bu c h e t / Ch a s t e l .
Boulez, P. (1987) Timbre and Composition - Timbre and Language. Contemporary Musi c Review, 2(1),
pp. 161-171.
Bregman, A. (1991). Timbre, orchestration, dissonance et organisation auditive. In Le timbre, mdtaphore
pour la composition, J.-B. Barri6re, editor, pp. 204-215. Paris IRCAM/Christian Bourgois.
Cadoz, C. (1991). Timbre et causalitY. In Le timbre, mdtaphore pour la composition, J.-B. Barri~re, editor,
pp. 17-46. Paris: IRCAM/Christian Bourgois.
Chion, M. (1983). Guide des Objets Sonores. Paris: Buchet/Chastel/INA-GRM.
Chion, M. (1986). La dissolution de la notion de timbre. Anal yse Musicale 3, 2e trimestre, avril 1986,
pp. 7-8.
Chion, M. (1991). L'art des sons fixds ou la musique concrdtement. Fontaine: Editions Metamkine/Nota-
Bene/Sono-Concept.
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8
48 D. Smalley
Cogan, R. (1984). New Images of Musical Sound. Cambridge: Harvard University Press.
Duchez, M.-E. (1991) L' ~volution scientifique de la notion de mat~riau musical. In Le timbre, mdtaphore
pour la composition, J.-B. Barri~re, editor, pp. 47-81. Paris IRCAM/ Chri st i an Bourgois.
Emmerson, S. (1986). The Relation of Language to Materials. In The Language of Electroacoustic Music,
S. Emmerson, editor, pp. 17-39. Basingstoke: Macmillan Press.
Erickson, R. (1975). Sound Structure in Music. Berkeley University of California Press.
Grab6z, M. (1991). Narrativit~ et musique electroacoustique. In Musicworks 51, Aut umn 1991,
pp. 47-50.
Grisey, G. (1991). Structuration des timbres dans la musi que instrumentale. In Le timbre, mdtaphore pour
la composition, J.-B. Barri~re, editor, pp. 352-385. Paris : IRCAM/ Chri st i an Bourgois.
Krumhansl, C. (1989). Who is Musical Timbre So Hard to Understand? In Structure and Perception of
Electroacoustic Music, S. Nielzdn & O. Olsson editors, pp. 43-53. Amsterdam: Excerpta Medica.
Lerdahl, F. (1987). Timbral Hierarchies. Contemporary Music Review, 2(1), pp. 135-160.
M~che, F.-B. (1991) Musique, Mythe, Nature, ou Ies dauphins d'Arion. Paris: M~ridiens Klincksieck.
Manoury, P. (1991). Les limites de la notion de "t i mbre. " In Le timbre, mdtaphore pour la composition,
J.-B. Barri~re, editor, pp. 293-300. Paris: IRCAM/ Chri st i an Bourgois.
McAdams, S. & Saariaho, K. (1985). The Qualities and Functions of Musical Timbre. Proceedings of the
International Computer Music Conference 1985, pp. 367-374. San Francisco: Computer Music
Association.
McAdams, S. (1993). Recognition of Sound Sources and Events. In Thinking in Sound: The Cognitive
Psychology of Human Audition, S. McAdams & E. Bigand, editors, pp. 146-198. Oxford: Clarendon
Press.
Murail, T. (1980). La r~volution des sons complexes. In Darmst~dter Beitr~ge zur Neuen Musik, XVIII,
pp. 77-92.
Nattiez, J.-J. (1990) Music and Discourse: Towards a Semiology of Music. Princeton: Princeton University
Press.
Risset, J.-C. (1978). Haut eur et timbre des sons. In IRCAM reports, 11/ 78.
Risset, J.-C. (1991). Timbre et synth~se des sons. In Le timbre, mdtaphore pour la composition, J.-B.
Barri~re, editor, pp. 239-260. Paris: IRCAM/ Chri st i an Bourgois.
Saariaho, K. (1987). Timbre and Harmony: Interpolations of TimbraI Structures. Contemporary Music
Review, 2(1), pp. 93-133.
Schaeffer, P. (1966). Traitd des objets musicaux. Paris: Seuil.
Slawson, W. (1985). Sound Color. Berkeley: University of California Press.
Slawson, W. (1989). Sound Structure and Musical Structure. In Structure and Perception of Electroacous-
tic Music, S. Nielzdn & O. Olsson, editors, pp. 13-23. Amsterdam: Excerpta Medica.
Smalley, D. (1986) Spectromorphotogy and Structuring Processes. In The Language of Etectroacoustic
Music, S. Emmerson, editor, pp. 61-93. Basingstoke: Macmillan Press.
Smalley, D. (1991) Spatial experience in Electro-acoustic Music. In L'Espace du son 2, Lien, revue
d' esth~tique musicale, F. Dhomont, editor, pp. 121-124. Ohain: Musiques et Recherches.
Smalley, D. (1991) Acousmatic Music - Does It Exist? In Vous Avez Dit Acousmatique? Lien, revue
d' esthdtique musicale, A. Vande Gorne, editor, pp. 21-22. Ohain: Musiques et Recherches.
Smalley, D. (1992) The Listening Imagination: Listening in the Electroacoustic Era. In Companion to
Contemporary Musical Thought, Volume 1, J. Paynter, T. Howell, R. Orton, P. Seymour, editors, pp.
514-554. London: Routledge.
Ten Hoopen, C. (1994) PhD thesis, in progress, University of Amsterdam.
Truax, B. (1984) Acoustic Communication. Norwood NJ: Ablex Publishing Corporation.
Wishart, T. (1985). On Sonic Art. York: Imagineering Press.
Wishart, T. (1986). Sound Symbols and Landscapes. In The Language of Electroacoustic Music,
S. Emmerson, editor, pp. 41-60. Basingstoke: Macmillan Press.
D
o
w
n
l
o
a
d
e
d

B
y
:

[
M
c
G
i
l
l

U
n
i
v
e
r
s
i
t
y
]

A
t
:

2
1
:
3
9

2
2

A
u
g
u
s
t

2
0
0
8