Ligeti Inaudible Structures

Inaudible Structures, Audible Music: Ligeti's Problem, and His Solution
Author(s): Jonathan W. Bernard

Source: Music Analysis, Vol. 6, No. 3 (Oct., 1987), pp. 207-236
Published by: Wiley
Stable URL: https://www.jstor.org/stable/854203
Accessed: 03-09-2018 01:15 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Wiley is collaborating with JSTOR to digitize, preserve and extend access to Music Analysis
This content downloaded from 129.74.250.206 on Mon, 03 Sep 2018 01:15:23 UTC
All use subject to https://about.jstor.org/terms
JONATHAN W. BERNARD
INAUDIBLE STRUCTURES, AUDIBLE MUSIC:

LIGETI'S PROBLEM, AND HIS SOLUTION
Not long after his emigration to the West in 1956, Gy6rgy Ligeti decided t
challenge a well-established compositional trend. Ligeti is known today as on
of a relatively small number of composers who in the late 1950s sought viable
alternatives to post-Webernian serialism, and it is no exaggeration to say that,
nowadays, when aside from a handful of undisputed masterpieces much of wha
was written during the serialist era seems hopelessly dated, Ligeti's music from
about the same time sounds as fresh and original as ever. His career affords th
music analyst the opportunity to study one composer's turning away from
serialism: the nature of his objections, their implications for the furthe
development of his technique, and the way in which the methods he arrived at
achieve a meaningful organization of musical materials. This article discusse
these issues and proposes analytic approaches designed to engage Ligeti's
solution to problems of musical composition in what turned out to be the post-
serialist era.
Ligeti's emigration brought him into contact with a thriving European
(especially German) avant garde. Perhaps because he came to it later than
others, having lived since the end of the war under conditions of provincial,
state-imposed isolation, Ligeti responded to the stimulus of this activity
differently from the way many of his contemporaries had done. Although he had
arrived with scarcely any knowledge of twelve-note technique, let alone the
extension of serial principles to aspects of musical sound other than pitch, not
even three years had passed before Ligeti was setting down his criticisms of
serial techniques as they had come to be applied to composition during the
1950s.' His previously published analysis of Boulez's Structures (Part Ia), which
reflected an extensive familiarity with serial methods, demonstrated that he
spoke from a well-informed position.2
Ligeti's difficulties with serialism, as expressed in the early Die Reihe article
'Metamorphoses of Musical Form' and in other, subsequent publications, can
be succinctly summarized. He found problematic 'the organization of all the
musical elements' - that is, pitch, duration, timbre, dynamics, mode of attack
- 'within a unified plan' because he 'detected within it a discrepancy:
MUSIC ANALYSIS 6:3, 1987 207
JONATHAN W. BERNARD
quantification applied equally within the various areas produced,

of view of our perception and understanding of musical pro
different results, so that there was no guarantee that a single b
produce analogous structures on the various levels of pe
understanding'. Unity, therefore, existed only on the le
description, 'clapped on the musical events from the outsid
generalization of serial procedures to engage more abstr
characteristics, such as types of motion, form, density and s
tighten the loose connection that had developed between compos
and the actual, resultant sound of the music. Pre-planning
important that it was the real compositional act.4
Even in the treatment of pitch alone, new music had apparent
up against certain limits:
The individual character of the various serial arrangements fades

of the superposition of several horizontal series, in which, w
possible, common notes occur at the same pitch. Such interw
obscures the single serial threads (especially when all the parts are
one instrument), and the resulting intervals have little or nothing
the original arrangement. Where such a procedure is coupled with
durations the composer can hardly even retain an influence
intervals that are to result, let alone determine them. They
automatically from the type of procedure. In this way the pitch
its last remnant of function, paralysed by the emerging complex
The relegation of resultant sound almost to the status of by-pro

'decreasing sensitivity to intervals' and permeability of structur
of different textures can run concurrently, penetrate each other
into one another completely . . . it is a matter of indifference w
coincide in the thick of the fray.' Small wonder, then, that com
adopted serial methods discovered that it was becoming 'increasi
achieve contrast' - their music suffering an inevitable flattenin
Ligeti did give certain composers, such as Boulez and Stockhau
having transcended the worst of these difficulties. Yet even the
completely unscathed: 'Although [their] works create the
abundant coherence, nonetheless this coherence, arising as it
the relationships established during the compositional process, i
a certain quality of "malgre lui".,7 What Ligeti heard in much of
written at that time was a preference for 'homogeneous sequenc
particularly the chromatic scale', with the result that 'the verti
this material results in a piling up of neighbouring tones.
primarily the intervals that constitute the structure but relatio
distribution of registers and various displacements in the b
breaking down of the vertical complexes.'
These observations seem to have brought Ligeti to a key re
208 MUSIC ANALYSIS 6:3, 1987
qualities he had noted were in fact the true determinants

music, why not engage them directly, instead of thr
methods that could not control such qualities, except more or less
serendipitously? More than a decade after first expressing his reservations,
Ligeti summed up his subsequent experience as a composer in a statement
which has almost the character of a manifesto: 'In working out a notional
compositional structure the decisive factor is the extent to which it can make its
effect directly on the sensory level of musical perception.'9
This statement, however, is less straightforward than it perhaps appears to
be. Certainly it should not be taken to mean that Ligeti felt he had learned
nothing from his exposure to serialism except what not to do. He has spoken, for
example, of having absorbed and applied to his own work such aspects of serial
thinking as 'the principle of selection and systemization of elements and
procedures, as well as the principle of consistency: postulates, once decided
upon, should be carried through logically'.'0 Further, by making this
declaration Ligeti does not categorically reject music that cannot be heard, in all
its particulars, according to the way it is composed. One may not actually hear
the row at every moment (or even in some cases, as a literal series, at all) in the
twelve-note compositions of Schoenberg and Webern, but this hardly negates
the row's structural importance or mitigates the worth of the music." Clearly
there are any number of ways in which a 'notional compositional structure'
might make its effect, even if restricted to doing so directly. Knowing that this is
what Ligeti is after does not make matters any easier for the analyst of his music.
For consider the following:
Technically speaking, I have always approached musical texture through

part-writing. Both Atmospheres and Lontano have a dense canonic structure.
But you cannot actually hear the polyphony, the canon. You hear a kind of
impenetrable texture, something like a very densely woven cobweb ....
The polyphonic structure does not come through, you cannot hear it, it
remains hidden in a microscopic, underwater world, to us inaudible.12
How can we come to terms with this apparent discrepancy between what is
written and what is heard? What is the point of composing strict canonic
structures that cannot be perceived as such? And if we do not hear this 'micro-
polyphony', as Ligeti terms it, then what do we hear? Before we can attempt to
answer these questions, Ligeti's ideas about his music must be exposed in some
detail.
One of the most striking general features of Ligeti's descriptions of his music,
both in his articles and in his interviews, is his frequent recourse to visual
analogies, especially ones having to do with space. Ligeti is under no illusions
about the ultimate significance of these analogies - he calls the space of his pieces
'imaginary' and is careful to distinguish the sense in which his music is spatial
from that in which Stockhausen's Gruppen is, for instance, or any other work
which involves literal dispersal of forces to different points within the
JONATHAN W. BERNARD
performance space - but still these expressions have an inher

one thing, Ligeti seems susceptible to visual and tactile para
phenomena to a degree that approaches synaesthetic sensitivity: 'The
involuntary translation of optical and tactile impressions into acoustic ones
occurs to me very often; I almost always associate sounds with colour, form and
consistency, and vice versa: form, colour and material quality with every
acoustic sensation.'13 However Ligeti's condition may have originated, it has
had an undeniable effect on his compositional development. There is, for
example, his famous childhood dream, the memory of which apparently
influenced the composition of Apparitions (1958-9) and probably some later
works as well. In this dream, Ligeti found himself entangled in a gigantic web,
along with various insects and inanimate objects, and became a captive audience
for the gradual transformation, through the insects' struggles to free
themselves, of this web 'universe'.14 It is tempting to speculate that Ligeti may
have been impelled by this intersensory facility to his particular choice of
compositional method, in which concentration primarily upon 'conditions of
the material' inevitably arouses 'associations with visual and tactile
sensations'. 15
Ligeti's tendency to 'spatialize', if that word can be used, also owes something
to the state of new music at the time of his arrival in Western Europe. In
'Metamorposes of Musical Form' he notes 'the seeming conversion of temporal
relations into spatial ones', as if a musical composition could in some ways be
analogous to a painting. Under these conditions, 'the succession of events is a
mere exposition of something that in its nature is simultaneous'.16 Actually,
what seems to happen, in Ligeti's view, is that the time of a composition evokes
space, and that the spatial analogy thus suggested allows the composer (and, by
implication, the analyst) to traverse the structure of the music as if it were
present all at once."7 Furthermore, spatial models of musical structure are of
particular interest in the study of form, since the idea of 'form' in music is
essentially an abstraction from spatial configurations, from the proportions of
objects extended in space. Musical form, then, can be termed the imaginary
spatialization of temporal processes.18
Not every spatial possibility is equally attractive to Ligeti. His antipathy
toward 'moment' form is revealing - for he feels that while mobility may be
inherent in musical form, the form itself cannot be mobile. Moment form, he
says, is based on a false analogy to visual art: for example, Calder's mobiles.9
'Musical moments have meaning only in that they point to other moments: not
the meanings themselves, but only the shifts and alterations of meaning, are
comprehensible.'20 Instead of writing pieces the order of whose parts was
variable, it seemed to Ligeti to be 'much more worthwhile to try and achieve a
compositional design of the process of change'.21 For this reason, perhaps, Ligeti
has adopted, like Varkse, the analogy to crystallisation in order to describe not
only the finished product of his compositional process22 but also, somewhat
inconsistently, the process of eternal 'becoming' exhibited by a piece as it
progresses:
The technical process of composition is like letting a cry

supersaturated solution. The crystal is potentially there in t
becomes visible only at the moment of crystallisation. In muc
you could say that there is [in my music] a state of sup
polyphony, with all the 'crystal culture' in it, but you cannot
aim was to arrest the process, to fix the supersaturated solu
moment before crystallisation.23
The analogy to crystallisation expresses something else as

unidirectional nature of the form-creating process.
characterizes his electronic piece Artikulation as 'a gradual,
from the heterogeneous disposition at the beginning to th
and interpenetration of the contrasted characters at the en
consistent with the atmosphere of his childhood dream, in
the web 'seemed like an irreversible process, never returni
again. An indescribable sadness hung over these shifting f
the hopelessness of passing time and the melancholy o
events.'25 The allusions to crystalline structure suggest
regular, even symmetrical patterns in the music, about wh
to say shortly.
Finally, among general aspects of Ligeti's musical outlook
in the history of Western compositional practice should
upon. This has obvious relevance to the elaborate c
mentioned earlier; and in fact Ligeti has allowed that
counterpoint' as a student.26 His educational backgroun
emphasis upon traditional instruction, and as a pedago
remains convinced that even if the old techniques canno
contemporary composers, the student can nevertheless lear
think logically in the musical sense. Keeping modern co
standards of the past cannot be a matter of indifference to
The specific consequences of Ligeti's attitudes and opi
structure are best examined in the context of analytic illust
first to his First String Quartet (1953-4), a work which, lik
wrote before leaving Hungary, bears significant marks of
Among these is a penchant for symmetrical constructi
registrally consistent symmetrical construction. Example 1
521-33, from roughly the midpoint of the work. Excluding
cello, the contents of each chord are mirror-symmetr
bracketed numbers below the music show the adja
arrangements. Furthermore, the passage as a whole has a sy
in two senses. First, temporally speaking, the series of fou
ends with the same vertical arrangement (though not the
two chords in the middle share a different vertical arrang
verticalized aggregate of all pitches in the passage - includi
JONATHAN W. BERNARD
the arrangement shown to the right of Ex. 1. This is not mirror-sy

all its details, but it does have an overall symmetry embedded in
The pitches that serve as the boundaries of intervals in this embedd
are encircled for clarity. They include the lowest fifth of the cello
pitch of the symmetrical portion of each of the four individual ch
Ex. 1 First String Quartet, bs 521-33
5 6
171
_ ,- ?I(p_ 181
LIIS
141 171 171 141 141
131
111 I 171
171 HI[11
P131 ?
131 111 111 131 P] HIl
141 171 171 141 0
2 f7 J
S C 18
@
(D? I~
Apparitions, Ligeti's next major work after the First String Quartet, was also
the first work he completed after his emigration, apart from two electronic
pieces. Having observed the loss of sensitivity to intervals in serial music, Ligeti
decided to see what could be done if this newly evolved condition were taken as
a given and, in fact, exaggerated by dispensing altogether with intervals as
structural components:
I composed sound webs of such density that the individual intervals within
them lost their identity and functioned simply as collective interval groups
... this meant that pitch function had also been eliminated .... Pitches and
intervals now had a purely global function as aspects of compass and note
density.29
This maximized density took the form of chromatically filled spaces: 'I inserted
so many minor seconds that even the minor seconds, the chromaticism,
disappeared in the harmonic sense.'30
Ligeti's own remarks about Apparitions emphasize the ide

from one sound group to the next through what he calls 'a
relationship between states and events'. He continues:
The states are broken up by suddenly emerging events and a

under their influence, and vice versa: the altered states also
effect upon the type of events, for these must be of ever n
order to be able further to transform the transformed state. In
an unceasing development: states and events, once they h
reciprocally exclude their repetition, thus are irretrievable.3
These formal ideas are evidently combined with fixed re

various domains of musical sound; Ligeti has discusse
durations in some detail, comparing his employment of
selection of letters from a type box, and has mentioned oth
he enumerated the contents, but it is possible to specula
what these are and, more important, about how they are
how the transformations take place. The following analy
interrelated domains of pitch, compass (vertical span)
explore the first part of the first movement.33
In Ex. 2 the score of bs 1-23 is transcribed into grid nota
a uniform semitonal calibration along the vertical axis. (
margin mark locations of C, with C4 corresponding to
along the top edge are bar numbers.) The grid may be an es
analytic tool in Ligeti's case, since it is known that in the i
stages he uses a kind of graphic notation.34
The opening bars show a development of pitch/registr
strictly controlled group of intervals, several of them dete
number of available parts in each of the divisions of string
initial minor second, (D# -E)1, is superseded locally by a pe
[7]) [13] above. The next two events incorporate all three of
[7] and [13] - and bring in new ones as direct resultants of
of the original group. The event (C# =E)1 (b.8) does not at f
related, but with the lower boundary of F 1 = G2 (bs 9-11)
of a symmetrical expansion from (D# -E)1.36 A by-product
expansion is the interval (C# -F# )1, or [5], which now appe
F# 1=G2, combined with [1] and [7]. Further, the interv
distance from upper boundary to upper boundary, C3
F4 1=G2 is also [13] in total vertical extent, which means
expansion [2] noted earlier is an interval of projection as wel
projection and filling-in of E1-F2. These two [13]'s taken
read either as [13] [2] or [2] [13], which is next immediately
one as the total extent of the cluster in b.13, E2=G3; the ot
from previous to new low boundary pitch, C# 1-E2. Note t
[15] is expressed as [7] [1] [7].
JONATHAN W. BERNARD
Ex. 2 Apparitions, I, bs 1-23

I 1 113 1 1 151 1N 9 ij 2 11 2
t 4
.. .. .. . d__L t- L . . . .
F1
-A Ii7sl
iit
171 A
4- Ft111
TTt
Total spa qo i teal

Toti(wer Teea
spi of nteak (herere~eanD
I I -
Intervals that are present in essentially unrealized p

are [11]: F# 1-F2 and [10]: F# 1-E2. The former is
(bassoons), in the chord Ab 1-G2-F# 3. Generally
doubling of interval size seems to proceed from the i
expressed as [7][7] - F2-(C-G)3, then as [7][1][7]
[15][15], C# 1-E2-G3. It is interesting, though not ne
[11] is also the average of [7] and [15]. The occurr
interval, in b.16 falls within the span of the lower [11
conjunction with the filled-in [10], D# 1 =C# 2, that
it also expresses [11], which can now be regarded
transposed down [5] and now completely filled ch
(D=G)2 in b.19 bears a relation to all three previo
replicates the [5] Al= D2 and in that, restoring as it d
[11] into the same spatial relationship that they occupied in

events in bs 15-16.
The cluster in the strings in b. 19 is the first event in the piece to enter before
the previous event has ended (discounting momentary overlaps, as for example
in b. 16). Here [13] is expressed as [5][1][7], as in bs 9-11; the overlap of(D= G)2
with the uppermost component, (D= A)2, expresses [7] as [5][2], as occurred in
the relationship of the cluster in bs 9-11 to events of earlier bars, but the means
in b. 19 is different, with the spans of [5] and [7] being given simultaneously.
Furthermore, the [5] between successive lower boundaries, (D# -Ab)1,
replicates the [5] (C~ -F#)1 between successive lower boundaries in bs 8-9,
which means that the relationship between the composite spans CQ 1-G2 and
F# 1-A2 is one of transposed replication, by [2].
The entrance of the celesta in b.21, on (C= B)4, brings in the filled-in interval
[11] at a distance of [15] above the immediately previous event. The following
chord in the strings (bs 22-3) is the largest simultaneously sounding
chromatically-filled span heard so far in the piece, and combines the structures
of two previous chords: [5][1][7] and [7][1][7] --- [5][1][7][1][7]. Just as [5][1][7]
in bs 9-11 was transposed up [2] to its position in b. 19, so [7][1][7] in b. 13 is now
transposed up [2] to its position as a component of the chord in bs 22-3. The total
span of [21] duplicates that of the composite of the first two events in the piece:
D 1-C3.
The last events in the graph are not analysed in full, for they seem to belon
more to the next part of the piece; but it is possible to find, in their overlap wit
the preceding [21], the intervals [28] and [27] now emerging as filled-in
composite spans after their more oblique statements as Ab 1-C4 (bs 19-21) and
F# 3-D# 1 (bs 15-17) respectively.
In retrospect, we can identify a series of phases through which all the interval
employed so far pass:
1) indirect or oblique statement;

2) successive lower or upper boundaries of clusters, or space between clusters,
or composite space;
3) composite space, filled chromatically;
4) simultaneous statement, filled chromatically;
5) incorporation as segment into larger cluster.
Next, we can detail the actual phases through which each interval participates
in this series:
JONATHAN W. BERNARD
[1]: (4); (5)

[2]: (2); (5)
[5]: (2); (5)
[7]: (4); (5)
[10]: (2); (4)
[11]: (1); (2); (3); (4)
[13]: (2); (4); (5)
[15]: (1); (2); (4); (5)
[21]: (2); (4); (5) (b.26)
It is evident from the above that no two trajectories are identical, and that none
of the intervals passes through all five phases. With the exception of the last
phase, the series seems to represent a process of gradual emphasis, as the
employment of each interval as a space-defining entity is made progressively
more explicit. The identification of the fifth phase - incorporation into larger
clusters - suggests that the development of spatial relations through various
interval sizes in this piece does not proceed simply by means of bringing each
interval into prominence and then maintaining it in that status. Once brought
to an explicitly space-defining role, an interval can subsequently be employed in
any of its various other roles and, as the piece goes on, can be absorbed into the
texture to become a component in larger sonorities, perhaps even suffering a
temporary or permanent cancellation as an explicit entity.
Conversely, as Ligeti's description of the form suggests, new intervals must
also arise. In bs 25-6, for example (not shown in the graph), the interval [8]
emerges explicitly from the lower two adjacent intervals in the chord [7][1][7]:
F# 2=A3. The boundary notes in the cello/viola chord in b.25 are F# 2, C# 3, D3
and A3; then, in b.26, F# 2 and D3 become the boundaries of a separate chord
in the winds. A little further on (b.29), [8] becomes a segment of a larger chord,
marking the contrabasses' portion, (D = B)1, of D 1 = Eb 3. Phases (2), (4) and
(5) are thus represented.
In Atmospheres (1961) Ligeti is still working with chromatically filled
complexes of sound, but the idea of a repertory - of durations or intervals, for
instance - has been discarded. 'Rhythm', as Ligeti has said of this work, 'is
completely eliminated, [and] the absorption of individual shapes into static
planes is accomplished to the greatest possible extent.'37 Atmospheres is widely
reputed, not without reason, as Ligeti's klangfarben piece, but Ligeti has
effectively cautioned the analyst by saying that 'it is a rather superficial view to
lay too much emphasis on timbre' in this work or other works of his, and that
in Atmospheres 'modifications of timbre and dynamics are obviously very
significant but the patterns emerging from them are even more important'.38 It
would seem, then, that the 'iridescence', as Ligeti calls it, caused by minute,
continual shifts in doubling and in the location of gaps in the chromatic filling,
and by dynamic changes, bowing changes and so on serves to characterize and
differentiate the various 'static planes' and impels movement from one plane to
the next. With this in mind, it is possible to analyse Atmospheres through the
same sort of grid notation employed for Apparitions. A gra

in Ex. 3.39
In b.1, the strings enter in three slightly separated groups: contrabasses,
(Eb =Bb )2; cellos/violas, C3=G4; violins, Bb 4=C= 7. Overlapping the gap
between violas and violins are the winds: Ab 3-C5, plus D2 in the contrabassoon.
In the course of bs 1-8 the winds' plane undergoes changes, as several pitches
drop out; the consecutive stages are displayed in the example. This reduction in
density, taking place as it does against a backdrop of chromatically filled space
in the strings, causes the winds almost to disappear, as if their plane had merged
with that of the strings. In particular, the departure of the flutes after b.4
changes the winds' upper boundary to B4 and prepares for the assumption of
this pitch as the cello/viola upper boundary in b.9. At this point, all previous
structures are superseded and the cellos and violas sound alone until the rest of
the ensemble rejoins them in b. 13. In the course of bs 9-13, a ripple of dynamic
change passes through the cluster of cellos and violas - or rather two ripples,
moving in opposite directions, as pairs of parts crescendo, one by one, to forte and
then drop back to the ensemble pianississimo. The succession of dynamic
emergences creates the symmetrical design shown in the graph.40
The strings reach their new location by transposition of all three groups,
although not in parallel: the contrabasses move up [5], the cellos/violas up [4],
the violins up [3]. A gap thus remains between the upper two groups, but the
lower two have meshed completely, and a new regularity has developed: the
segment encompassing low and-middle strings, Ab 2= B4, corresponds precisely
to the size of the violins' segment, Db 5=E7. During bs 13-22 a new sort of
dynamic fluctuation takes place in the strings, dividing the contents of the
cluster into two interlaced collections on separate crescendo-diminuendo
schedules. These momentarily emerging vertical arrangements exhibit the same
regularities and mutual complementarity as do the black and white notes on a
piano keyboard.
Meanwhile, something else has happened to the winds, which now occupy a
space (not completely chromatically filled) bounded by Gb 3 and C# 6, plus Eb 1
in the tuba, joined at b. 15 by El in the contrabassoon. From their original
location at D2-C5, they have undergone an expansion that adds two octaves to
their compass. This will be called an '[11]-[13] expansion', after the two
intervals that define the intervals of change at lower and upper boundaries
respectively. The winds' fluctuations at this point are more complex than those
of the strings and for the sake of legibility are shown in a separate example (see
Ex. 4). As in the strings, crescendi successively emphasize white-note and black-
note collections, but unlike the strings the winds project staggered dynamic
patterns; and besides dynamic change another form of 'disturbance' is
introduced into the texture: rearticulation of individual notes, shown by the
placement of crosses. As in bs 1-8, the winds in bs 18-22 die away by stages and
sound as though they have been absorbed into the strings.
From b.22 to b.23 the strings contract a total of two octaves in range, from
Ab 2-E7 to G3-EC 6. This can be called an '[ 1]-[13] contraction', complementing
JONATHAN W. BERNARD
Ex. 3 Atmospheres, bs 1-29
ii ' .-.: ' ' Si. ..

:+<. . .-'
. . . . . . .
' I ' : : : : : 7 + > + .. . . . . + . . . . . . . . . . . . .
-,++ }, +- 7 - ; ; : i . ; . + i ' : . . . . . . . . . . ? . . . . . . .-.. . . . .
:t~i:..........
. . . . . . . . . . . . + k . . . . . '. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .< . . . . . .
. . . . . . . . . . . . . ; , . . . . "-, ? + ; . . . . . . . . . . . ++ . . . . . .+ . . . . ., .+ . + I ; } L } .
. . . . . . . . . . ._ ._ . . . . . . . . . . . . . . . '- . -. . . . .. . . .
!L"u .- + . . . . . . . . . .. . . . . . .. . . ..
:. ;.. .+ +ii s ~ i. i ii !i i~i.. . . . . . . . . .

: : ; : l~ .". : ,i . . . . , . . . h,'L;. : . . . . -.
itt
A.... . . ...... ... --.. . .... ...
- i +-
_ ,+;. . . . . . . .- .+ .
Ex. 4 Atmospheres, bs 13-20, winds
131 1 115116 18 1 20
C# x x DL Db Db
B B
x x B6 Bb Bb
A
x x A6 A AA
G
x xI G1; G? G
SF F G
x Eb E6 E6
-. x DID
x Db6 D6 D
5 C x C C
EE x B6 B6 B6
x GG
x Ai A1 A:
x G?
x F F
x E
x Eb
x DD
I x C
x BB
V X. -A-
At
i-
--x
x A
x Ab
Bb
G~ x Gb
--E
I :, E
J,
E6E6
JONATHAN W. BERNARD
the previous expansion in the winds. The new span now serves as t
departure for a great downward sweep, from D6, the upper bo
plane from b.24 on, to F2. The textural change as this hap
dramatic: sul ponticello, molto vibrato playing in ever faster figurat
in each part between two pitches takes over from sul tasto sustain
pitches as the downward sweep progresses.4' Just before the botto
two more planes appear (b.25), flutes and clarinets nestled symme
the compass of the contrabasses plus three cellos. This smaller str
left sounding alone in b.29 after all other instruments suddenly e
The graph displays the striking symmetries outlined by the
planar states from b. 13, where the composite winds'/strings' range
the sweep of bs 23-9, which represents a symmetric contraction f
boundaries of that range, to the further symmetric contractio
string group left sounding in b.29.
The organization of volumes of sound of varying density a
schemes based upon vertical span and symmetrical considerations
a prominent feature of Ligeti's work since Atmospheres. Begin
second movement of Apparitions and continuing, with increasing
in subsequent works, Ligeti integrated this basic 'spatial consc
other compositional preoccupations, notably the high regard for t
contrapuntal procedures of older music that he had acquired as a s
relatively brief examples from works of the 1970s will serve to illu
techniques inspired by (if rather remotely related to) these
'Bewegung', the third of the Three Pieces for Two Pianos (197
section is based on a rigorous pitch symmetry, canonically unf
5a). Each strand of the canon considered separately is a pair of
note-against-note strictly in mirror fashion, so that the inter
vertically expand and contract symmetrically.43 In bs 49-52 t
constitute the entire texture, apart from rapid figuration (not
example) overlapping from the previous section and slowly fading
producing a double canon with the comes entering [9] below the d
of course, just as accurately be described as a double canon in inve
52-8 the double canon is itself doubled to become a quadruple cano
overall axis of symmetry, A4, preserved from the preceding
original pair of pairs becomes the centre of the entire structure as
are added, [6] above and [6] below respectively. In bs 57-8 the c
dissolve, as the two upper pairs enter together, followed one chor
lower two pairs. During the final two chords of the piece (bs
symmetrical layout is expressed entirely in simultaneities.
In the reduction of Ex. 5a, double barlines mark the points at wh
more comes 'catch up' with the dux. This happens once during the
(b.50), once just before the quadruple canon begins (b.52) and again
and 58. At these five points all voices resolve, as it were, int
symmetrical chord, to which the last two chords, also mirror-symm
as a kind of culmination. The catch-up chords can be heard as poin
Ex. 5a 'Bewegung' (3 Pieces for 2 Pianos, III), bs 49-63
6mt
A J ,, * ? x~ - .. ..
100 LIL p ti
mzmr
TT-
A a-I#-6 - v
- m ,, Ti~
S --"-- ---- -
-###
rwr u 'I (( "" ~b
,a"
i zzzz- t- ?----
~AL
Ag
TI
v U-
(V"I)
articulation, with the shift to the thicker texture of the quadruple canon and the
final abandonment of canonic projection (b.58) the principal points among the
five. In between it may well be impossible to hear the canonic design, since the
ensemble of two pianos is exploited, here as in much of the rest of the work, to
produce a unified (as opposed to stratified) texture. (What is not apparent from
JONATHAN W. BERNARD
Ex. 5a is that none of the four canonic pairs is confined to a singl

score; the two piano parts continually exchange pitches at irreg
intervals.) It should be possible, however, to hear something el
Here the contents of Ex. 5a have been transcribed into grid n
notation provides a more accurate idea of the spatial quality
enabling the listener/analyst to see it whole - as if, as Ligeti put i
passage were present all at once. From this graph it is evident tha
time lag between entrances of parts enforced by the canonic plan, o
perceive the general outline that emerges as a resultant of the can
despite the fact that the internal details are slightly askew,
dimensions stand out clearly.44
In the third movement of the Kammerkonzert (1969-70), w
another application of canonic procedure, in a setting radically dif
that of the previous example. Bars 1-12 present one of Ligeti
textures, a device which is one of his principal agents for 'get[ting]
as a concept altogether'.45 In this passage, an initial pitch is joined
by others in close spatial proximity, forming a slowly, steadily exp
sound. The resultant of the canon can be interpreted in terms
shifting symmetrical relationships. Example 6 displays the overlap
development. The first stage, at (a), surrounds initial pitch E4 sym
then by adding C# 4 leaves E4 no longer quite at the centre. In (b
group of instruments begins the canon over again; this time the e
the canonic series, C4, makes its appearance, pulling the centre
downward. With the entrance of the first sustained pitch, in b.7, t
begins: see (c). This eventually takes the form of a 'whole-tone' div
space opened up by the canon, as CQ 4, Eb 4 and F4 are completely
With respect to the original axis and initial pitch E4, space has bee
by [2] above and [4] below, a condition which is neatly summed up
three sustained pitches to enter - F# 4, E4, and C4 respectively
effectively 'predicts' what happens next, shown in (d): a new mecc
begins at b. 12 in octave Ab s. The two central Ab s, in octaves 3 and
by motion [4] below C4 and [2] above F# 4 respectively; the outer
simply expand occupied space outward from this central arrangem
Two works of the 1960s, Lux aeterna (1966) for sixteen-part a ca
and Lontano (1967) for orchestra without percussion, are especi
subjects for study of Ligeti's canonic technique, principally b
represent a deliberate attempt on the composer's part to exert mor
previously over the spectrum of relative clarity (transparency) to r
of texture. Ligeti regards Lux aeterna, in retrospect, as a turning po
comparable importance to Apparitions.46 In a brief essay entitled
zu Lux aeterna' he notes that the work, which he composed d
finishing the Requiem but which was conceived as a separate work
its text is taken from the traditional Requiem mass, was deliberatel
have more limited possibilities for opacity; its ensemble has only s
parts to the Requiem's twenty and omits orchestral forces. Lig
Ex. 5b 'Bewegung', bs 49-63
11
- It
t 1 ~ t I 1
I I i i l
t t I
- .:j . 4
JONATHAN W. BERNARD
Ex. 6 Kammerkonzert, III, bs 1-12
(call llit Stranld)
a) (b) (c) (d)
121 21
141 41 .
Ex. 7 Lux aeterna, bs 1-37
, I. . . . .... 7.. . . . . .. . 0
3 5. , I. . . . .7 8 I. .I 7 1 1. .
2 1 I I1 7 9 I, I ' ;1 15 1i7 12 0 '
gradual transformation as a principle of harmonic constru

the counterpoint has the function of destroying old struct
ones.47 To track and interpret this process, graphic n
different variety from that of previous examples will
duration scale (horizontal axis) is consistent: one squar
common time, at a tempo of approximately crotchet =
been dispensed with; instead, the shaded areas corresp
individual pitches, projected and prolonged by overlapp
parts. In this, as in many subsequent works, Ligeti has exp
performers to make their entrances as gently, even imper
during sustained passages. The listener is evidently not int
attack points; thus the graph simply shows where each pitc
one part.
The order of entrances of the first four notes, (F-E-G-F#
as two interlocking three-note groups with vertical int
[1][2] and [2][1] (a and b); the symmetrical arrangement
Ex. 7 cont.
. . .. . . .... I I.
. . . . . . . t . . . . . . . . . . . . . . . ... . . ..
. . . . . . . . . . . . . .... . . .
.' . . . ." . . ... . . . i .. . . . . ..... ..
r . . . . . ..? ? -
i . . . . . . ... . . . ... ..... ..
! . . : . . . . . . . . . uli
. . , . . . . .. . .
' ? i . . . . . . . .
t ' ' l ' ' ! ' ! ' ' ' i *~ ~~ r .r . . . ; ' t ' t ' ' ' , ' ' f
t
J tit T till. llll?I?l~?..:r~l

2 2 21 2624 1251 16 271 281 29 30 31 13 13 14 ' 5 1)6 1)7 1)8 1)9
21 221 2 :A 1 241 2S 112"3313 128-F-t4 1 13 1 9 1
JONATHAN W. BERNARD
available semitonal space is completely filled. As in the previous example, from

the Kammerkonzert, the semitonal web spreading out from an initial pitch has
grown slightly asymmetrically with respect to that pitch. By and large this is
typical of Ligeti's practice: here, as the dimensions of occupied space continue
to spread outward, the general idea of a symmetrical centre is effectively
conveyed, but the actual, specific location of that centre is constantly in flux.
Pitch E4 drops out in b.7, leaving the cluster (F = G)4, [2] in extent (c). As
if in response to this, Eb 4 enters (b.7), then Db 4 (b.9), marking off striations of
the semitonally calibrated space [2] apart (d). (Henceforth, in this analysis and
the one of the opening of Lontano that follows, such formations will be referred
to as '[2]-striations'.) Meanwhile, in b.8, Ab 4 enters (e), at first seeming to
function as an extension of the semitonal cluster (F = G)4. However, this effect
is only momentary, for F# 4 soon exits, leaving (F-G-Ab )4, or [2][1], as the top
portion of the structure (f). Pitch F# 4 reenters in b. 11, again seeming to reassert
chromatically filled space (g); but almost immediately G4 drops out, leaving
[1][2]: (F-FO -Ab )4 at the top of occupied space (h). The effect is of F 4
replacing G4, with a bit of overlap; the resultant is a duplication of the [1][2]/
[2][1] relationship that opened the piece, one semitone higher and with greater
temporal separation between the components.
The entrance of Bb 4 in b.12 (i) comes in apparent reflection of the [2]-
striations below. In fact, this impression is strengthened by the subsequent C5
in b.13 (j); in between, however, the appearance of A4 (bs 12-13) briefly
reasserts chromatically filled space as well as forming [2][1]: (Ft -Ab -A)4 and
[1][2]: (A-Bb )4-C5 (shown interlocked at k). With the entrance of C5 and the
nearly simultaneous exit of A4, [2]-striations flank a [1][2] centre, (F-F4 -Ab )4.
Further developments emphasize symmetry, but through a constantly changing
web structure. First, just as Db 4 disappears, a thickening develops at what
would be the centre if Db 4 were still present (1). The formation is, literally
speaking, asymmetrical (m). The 'lump' is gradually dissipated with the exit of
G4 (b. 16) and of F# 4 (b. 18); while this is happening, A4 reenters (b. 16), then
G4, just as F# 4 disappears in b.18, yielding another lump in reflective
symmetry to the first one (n).
As A4 exits in b.20, the pace of events accelerates somewhat. The graph
shows clearly how a 'ripple' of semitonal filling (o) moves up through the space
in use from the bottom, starting with E4 above the already present Eb 4. For a
split second, in b.22, the interval (Eb -Ab )4 is completely filled. Behind this
ripple comes another, less well defined (p). The graph shows momentary gaps
in the continuity of individual pitches, as well as the fact that the entire space
traversed is only (Eb -G)4 this time and is never filled all at the same instant. The
ripple analogy is particularly apposite, since the change in the texture at this
moment sounds as though a disturbance of some sort has been introduced - or
perhaps the ripple is simply a premonition of some larger effect about to break
out upon the surface of the music. In any case, just as the second ripple is
making its way up the semitonal spectrum, A5 enters dramatically, along with
its octave A4, which functions as a continuation of the first ripple's ascent (q).
226 music ANALYSIS 6:3, 1987
The pitch A5 is quite a bit higher than any other yet heard
location of this seemingly deliberate discontinuity is, ho
determined by the dimensions of the space previously oc
precisely, what these dimensions have come to be. The graph
the exit of D6 4 in b. 15, the upper and lower boundaries rem
Eb 4 respectively, until the entrance of A5. This space, [9] in
by the entrance of A5, for A5 is [9] above C5. At that momen
two spatial regions: one rather densely packed, the other c
Immediately thereafter the lower, dense region begins to t
exits; then a kind of descending ripple occurs across bs 26-
intervallic interstices in its wake (r). The effect here bring
description of a process that takes place in San Francisco Polyph
texture 'gets less dense, as if someone went through it with a
out'.48 Already in bs 25-6 the [2]-striations have reappeared
These are momentarily obscured by the descending rippl
partially in b.29 (t) as (F-G-A)4, then finally are presented
fashion by the successive exits that attenuate occupied space f
(b.32), F4 (b.34) and G4 (b.35), leaving A4 (u). Meanwhile,
successive exits of C5 in b.24 and Bb 4 in b.34, together with
present the by now familiar [1][2] pattern (v). Other
configurations are also present as the process of attenuation c
At b.35 the only pitches left sounding comprise the empt
This event marks a return to the maximal clarity of the single
cannot be considered exactly equivalent to it, since the oct
define a region of space, however equivocally. Octave-boun
considerable role later in Lux aeterna; here, the octave, beside
of arrival and as an obvious sectional division, has other, long
The chord that enters in b.37 (basses falsetto, F# -A-B) inc
intervallic structure, [3][2], is identified by Ligeti as the
harmony of the piece.4 Here it serves as a spatial pivot (see Ex
distance [10] from A5 down to B4 (successive upper boundaries
the distance from lower boundary F# 4 (b.37) down to Ab 3, th
of the next stable configuration in the piece, first entering in
Ex. 8
Ilio
10
19
1
6-6im -?
JONATHAN W. BERNARD
As for A4's role, it is of even longer-range import: in one sense a

the falsetto chord at b.37, but in another, in the bars immediately
entrance of that chord, the true 'replacement' of the initial F4 of
other words, A4-A5 can be heard as intervallic space, but it can al
a single pitch with a prominently audible overtone. The interval d
and A4 is duplicated, [9] lower, in the boundaries of the stable harm
isolation in bs 58-61, Ab 3-C4. The interval [9] is thus connected t
A5: the latter on the local level, the former over a longer span of t
Ligeti calls Lontano a 'sister work'51 to Lux aeterna, and in
relationship between the two pieces is evident from the outset - not
general similarity produced by the use of canonic procedures in b
specifically in that the pitch series which, used in canon, 'generat
section of Lontano (bs 1-41) is nearly identical to that of the first
aetema (bs 1-37), [3] higher.52 Example 9 presents a comparison. A
obvious difference of chosen medium, however, the treatment of
strand in Lontano produces considerable textural differen
predecessor. For one thing, the strand in Lontano becomes doubled
above beginning in b. 14, with the twelfth note of the series; for
canonic pedal pitches extend certain canonic pitches beyond the du
they would have if they were projected solely by the canon; f
Lontano continual shifts occur in the orchestration of the canon,
choir of instruments to begin a segment of the canonic line as an
finishing and producing thereby a greater degree of overlap than
the first section of Lux aeterna, with its consistent setting of fo
four alto parts.53
The opening ten bars (see Ex. 10) have the same basic shape as do
of Lux aeterna (a), and after that the [2]-striations open up bel
fashion: F# 4 in b.11, E4 in b.13 (b). Already, however, t
development has begun to diverge from that of the earlier work.
pitch A4, once introduced in b.9 (c), continues to sound until towa
the section, whereas F# 4, its counterpart in Lux aeterna, exits fro
for several bars as the [2]-striations are formed. In Lux, F# 4's ab
Ex. 9
Lux aeterna
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Lontano- -------
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
effect of extending the striation through much of the rest

point; in Lontano the retention of A4 contributes to a great
comparison with the other work.
The imitation of [2]-striation in the subsequent entries of
Lux also occurs in Lontano, but by different means: the int
doubling, as E, F# and G# 5 enter in bs 14-15 (d). The expl
the already present E, F# and G# 4 begins to establish a tw
with events in two separate registers moving in parallel. A
separate entrances in the two registers serve to suggest that
yet quite established: A5 enters in b. 15, to complement th
(e); CQ 5 and B# 4, also in b.15, are followed closely by C# 6
f and g); then the apex of the canonic strand, D# 5, preced
h and i).
The sudden ceasing of (E-F# -G# -A)5 in b. 19 leaves a hole in the texture (j).
The break is short-lived, but the momentary separation actually serves to
emphasize a parallel development in the descent of the canonic strand from D# ,
through CQ to B (k and 1). The [2]-striation now becomes a feature of the upper
region of the texture in the two registers, as it did in the one register of Lux
aeterna's opening passage. Already in b.20 we hear G# 5 reentering, followed by
A# 5 and F# 5 in bs 21 and 22 respectively (m). The hole is thus filled again, but
not completely, and not quite in the same way. In fact, these three pitches - G# ,
A# and F# - are the next segment of the canonic strand, after D# , C and B, and
they project another [2]-striation.
The graph shows, from about b.22 onwards, a distinctly more pronounced
spatial separation of the two registers, as they continue to move ever more closely
in parallel. In fact, all pitches in one octave now have octave-correspondents in
the other, except for E4, the lowest of all. The graph suggests one possible
reason for the omission of E5: precisely to provide some measure of separation
between the two registers. The [2]-striation D# -C# -B is maintained on both
tiers, but otherwise, and especially from b.24, the texture enters its most opaque
phase, with the space F# -B in both registers completely filled chromatically (n).
The orchestration here, except for one sustained clarinet part, is all strings, thus
Ex. 9 cont.
18 2( 21 22 23 24 25 26 27 28 29 30( 31
t- - -+---8-'-, 15,
if-u-~t -. .1 ~
18 19 20 21 22 23 24 25 26 27 28 29 30( 31 32 33
JONATHAN W. BERNARD
Ex. 10 Lontano, bs 1-41
i ~ ~ " ! t' ' i : ' i " " +

S, i . . . . .|., . .
i i I !i l [i ! i tiiri
i !I !+!ti( !t! :i t: l: i: !: ![!: :i i: : : : : : :
. . . .. . . . . . i . . . . . . . . . .
. . . . . . . . . . .. . , . . . !. . . . . . . . . . . . . .
. . . . . . ... ..... . ............. ............

. . .. . .. .. . .... . . . . . , . .. . .
t
. . . . . . ... ..++ . , : . . , . + . . ... . ..+++ . I. . , .
. . . . . . . . . . . . . . . . . . . . . . . . .
' ' "1 ? ? ? ? ?~ t . . . . . . . ' . . . . . . . . .; ' '? ' +. . . . . , . . . . . . . '
1 ??r i ?)l tl i i ?i l . . . . . ?. ,
: . . . . . . . i . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ! . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I . . . . . . . . . . . . . .
. . . . . i . . . . . . . . .. . . . . . . . . . . . . . . . . . . . I . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I . . . . . . . . . . . . . . .
, . . i . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . I . . . . . . . . . . . . . . . . . . . . .
2 . . .I. . 1 . . . 14
. . . . .. 15. . . . .I . . .16
. . . . .? . . .I . . . .'. .i . . II
. . . . !I
. . . .:. . I . . . . . . . . . .
. . . ! ! . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... . . . . . . . . . . . .
4it 11
230 MUSIC ANALYSIs 6:3, 1987
Ex. 10 cont.
)t~ ~ ~~i ttt-t t ttS~i I tl-[ L:i Lt -Itti-:l)!! tS'

t tt
~ti tI I t
, ttiil . . . ! t . . . . !i . . . . . . ,t !!i! ! ! i i ! ; i ! i i
t 1 t
t . . . .. .... .. . .. .. ...
. .. .
. .._
. . . . . . . . . . .
m ........... .. ._
.".".. . .."." ........, . . .. . . t........... .... ~ ii i ii i. i. . .
. .. . . . .. .... . . ... ... ...r ... .... .... .... ...
_.:.~ , 1. :: .. ... .... . i. .-. ~~~-t
. . . . . . -t-
. . .
.... .. . . , .. ..
1 1 r 1
. . . . . . . . . . . . . . . . . .
, . .i . . .. . . .. .
. . ~ ~ ~. . . . . . . . . . . . . . . . . . . . . . . . . . . . k . . . . . . . . . . . .
? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
N o*.... .. ... . . ...t ... .....

. .. .. . .. .. . .. .. . . . : . .. ... .. . . . . . . . . . .
. . . . .. . .. . . .. . . ? . . . . . . -. . . . . . . . . . . . t t
. . : . . .t. . . . .,.,.,. .... .. .. - -- . -t

. . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . t . . . . . . . . . . . . . . . . . . . t t
t . . . . . . . . . . . . . . i .
tt
t
+ I ;+
tT + t
t, tt+ii
f
f t
JONATHAN W. BERNARD
emphasizing homogeneity of texture by comparison with what

This passage has the effect of cancelling most previous
relationships and paving the way for the shape of the section's
the [2]-striation preserved above, it is not disrupted, appro
until the entrance of the pitch C (b.31), which in several octav
function here as do the octave As in Lux.
After the entrance of C(5, 6, 7), the texture begins to thin noticeably. Pitch
D# drops out in b.32, then E4 and B (4,5) in b.33. The graph (at o and p) reveals
that the thinning process systematically exposes [2]-striation in both registers,
and that the pitches forming these striations then proceed to drop out,
beginning with the lowest and ending with the highest, thus attenuating the
texture until only C is left. Combined with this reemergence of the [2]-striation
is the persistence of C# (5, 6) until b.38. As in Lux, the descent of the upper
boundary (q and r) produces the intervallic arrangement [1][2]; also as in the
earlier piece, the sounding of the last three notes of the canonic strand alone
(Db -Bb -C) from b.37 on constitute a form of [2][1].
Looking back, we find that the fluctuations in the density of the texture, as
revealed by the graph, are to a certain extent reflected in the series of pitches in
the canonic strand. Already noted are the [2]-striations at the outer edges of the
range of the strand: E-F# -G# and D# -C# -B, which occur as distinct segments of
the canonic series (twelfth to fourteenth and nineteenth to twenty-third pitches
respectively). Also worth mentioning are, for one, the very opening, which as in
Lux produces a narrow, compact band of pitches from two symmetrically
arranged forms of [1][2], and from numerous repetitions of Ab among the first
nine pitches in the series; and for another, the segment extending from the
twenty-fourth pitch (G#) to the thirtieth (A ), in which the constricted range
and constant changes in direction produce the climax of opacity noted earlier in
bs 25-32.
These canonic 'resultants' are also noticeable as features of the Lux graph. At
this point, a side-by-side comparison of Exs 7 and 10 is instructive. The
similarities between the two pieces show up more in general shape and contour
than in specific details. Internal relationships, in particular, are quite different.
It is well to remember that owing to the octave doubling in Lontano the visual
comparison with Lux is more accurately made (especially from about b.21 in
Lontano) between Lux and half the texture of Lontano. Even so, however, the
realisation of notes 22-8 of the canon in Lux (o and p, Ex. 7) differs considerably
from that of notes 24-30 in Lontano (n, Ex. 10) because of the larger number of
parts in the latter. Furthermore, the motion toward the bare octaves at the ends
of the sections under analysis is managed, in Lux, partly through an internal
'emptying' of occupied space which does not take place in Lontano. (See the
fissure that appears in the wake of the descending ripple r in Ex. 7.) This, as
noted in the analysis above, comes about in apparent response to the region of
empty space, equal in size to the space already densely filled below, opened up
by A5. But in Lontano the entrance of C7 is a less dramatic event than that of A5
in Lux simply because octave doubling is already a fact of the texture in the
orchestral piece. The progress from this point to the co

in Lontano sounds less like an emptying of space than a f
are already present.
Ligeti's problem could be summarized in the follow
one arrive at a compositional practice that satisfies the i
The worst excesses of the serialist era indulged the form
latter, but particularly for those who shunned the seri
what it led to, the experience of serialism made it effec
alternatives in which conscious method did not play a p
said that when he composes he always begins by imagin
piece to sound, in great detail, from beginning to end; t
produce that sound. This might seem to be excess at th
it not that the act of writing the piece, for Ligeti, invar
imagined plan.54 This must mean that the imagined
audible if it is based on consistent principles. The inaud
justify the audible music, but without a structure it wil
precisely what the music should sound like. Here is
contrapuntal structure in Ligeti's music becomes evi
though not secret, that can be applied flexibly and
temporal and spatial dimensions. Because it is notic
procedure, it does not automatically dictate in all re
emergent music, but neither is it simply a convenie
composer may hang his preconceived notions. There
Ligeti derives a deep satisfaction, both aesthetic and tec
canonic procedures as behind-the-scenes devices that
reciprocal relationship with the sounding surface, shapi
as they are shaped to meet the exigencies of the work.
readily communicable to the listener and analyst.
NOTES
1. Gyorgy Ligeti, 'Metamorphoses of Musical Form' (1958), in Die Re

(Form - Space), English edn (Bryn Mawr: Presser, 1965), pp.5-19.
2. Ligeti, 'Decision and Automatism in Structure Ia' (probably 1957), i
Vol. 4 (Young Composers), English edn (Bryn Mawr: Presser, 1960), p
3. Ligeti, 'Fragen und Antworten mit mir selbst' (1971), trans. Geoffrey S
Ligeti in Conversation (London: Eulenberg, 1983), pp. 124-37.
4. Ligeti, 'Uber Form in der neuen Musik,' Darmstadter Beitrage zur n
Vol. 10 (1966), pp.23-35.
5. 'Metamorphoses,' pp.5-6.
6. Ibid., pp.8, 10.
7. 'Uber Form', p.31: my translation.
8. 'Metamorphoses', p.6. Specifically, in Structures Ia he noted that 'W
this composition a complex network unfolds - of coarser or finer w
JONATHAN W. BERNARD
consisting of a significantly ordered flock of sounding "points": th

to form threads of varied thickness, which now stand out plastic
less distinct. The threads, for their part, are woven together wit
density' ('Decision and Automatism', p.61).
9. 'Fragen und Antworten', p.131.
10. 'Fragen und Antworten', p. 131.
11. Ligeti, 'On Music and Politics', trans. Wes Bloomster, Perspect
Vol. 16, No. 2 (1978), pp. 19-24
12. Peter Vairnai, 'Besz6lget6sek Ligeti Gyorggyel' (1978), trans. Gabo
Ligeti in Conversation, pp. 13-82.
13. Ligeti, 'Zustande, Ereignisse, Wandlungen', Melos, Vol. 34 (5 M
9: my translation.
14. Ibid. An English translation of Ligeti's account of this dream app
Conversation, p.25n.
15. 'Metamorphoses', p.15.
16. Ibid. The latter phrase is Ligeti's own quotation from Adorno
Adorno, Philosophy of Modern Music, trans. Anne G. Mitche
Bloomster (New York: Seabury, 1980), p.188.
17. 'Uber Form', p.24.
18. Ibid., pp.23-4, 34.
19. Ibid., p.34.
20. Ibid., p.26: my translation.
21. 'Metamorphoses', p. 19; emphasis in original.
22. 'Fragen und Antworten', p. 125.
23. V~rnai, p.15.
24. 'Metamorphoses', p.15.
25. V~rnai, p.25n.
26. Paul Griffiths, Gy6rgy Ligeti (London: Robson, 1983), p.20.
27. Ligeti, 'Uber neue Wege im Kompositionsunterricht: Ein Bericht',
of New Music (Stockholm: Nordiska, 1968), pp.9-44.
28. It is surely significant, in this connection, that in his own analytic
noted instances of pitch symmetry in Bart6k; see 'UTber die Harm
erster Kantate', Darmstadter Beitrage zur neuen Musik, Vol. 3
Registrally based symmetry, in particular, plays more of a role i
than has generally been recognised; see my article 'Space and Sym
Journal of Music Theory, Vol. 30 (1986), pp. 185-201.
29. 'Fragen und Antworten', p.128.
30. Josef Hiusler, 'Zwei Interviews mit Gy6rgy Ligeti' (1968, 196
Soulsby, in Ligeti in Conversation, pp.83-110. It should be noted t
in the resultant texture of clusters the traces of Bart6kian s
vanished, still the original idea of clusters predates Ligeti's e
Hungary and is attributed by him to his familiarity with Bart6k's
p.26). See, for instance, the third movement of Music for Strings
Celesta, bs 20ff. The first version of the first movement of Appar
entitled Vizi6k, completed in Budapest in 1956.
31. 'Zustinde, Ereignisse, Wandlungen', p. 169: my translation

32. 'Metamorphoses', p. 14; 'Fragen und Antworten', pp. 131-3
33. This plan of attack leaves other aspects of Ligeti's sound struc
reader should bear in mind, however, that these other aspects
importance according to the composer, are organised along
(again according to Ligeti) and thus are presumably separable f
analysis.
34. Ursula Stiurzbecher, 'Gy6rgy Ligeti' [interview, c. 1970], in W
Komponisten (Cologne: Gerig, 1971), pp.32-45. Obviously, h
claim to having recapitulated, in any substantive sense, Li
process. As he himself said of his analysis of Webern's Op.29,
more than merely conjure up the shadow of the actual creativ
Harmonik in Weberns erster Kantate,' p.53n: my translation).
35. Throughout this article the word 'interval' is used exclusi
distance, whether chromatically filled or not. Thus 'interval' r
compass' in Ligeti's terminology. Despite the potential conf
to use 'interval' in a sense not meant by Ligeti's use of the
compass' or even 'compass' is syntactically awkward.
The double vertical bars in the graph indicate clusters: c
filling. The fact that highest and lowest points, and bounda
groups within clusters, are marked with pitch names is not m
these boundaries are more 'pitch-like' than the notes within th
are shown in this fashion only to make it easier for the reader
info mation concerning interval sizes contained in the graph.
36. The short dash or hyphen is used here and subsequently to se
or to refer only to the boundaries of filled spaces; the long dou
the contents of clusters: chromatically filled spaces bounded by
37. Stiurzbecher, p.39: my translation.
38. Va'rnai, p.39.
39. The horizontal bracket in bs 1, 13 and 25 signifies that the ve
are attacked simultaneously.
40. The pitch names in the graph, here and in bs 18-20 (and al
individual points of maximum loudness.
41. Pitch names in bs 23-5 show each separate set of entries verti
42. The quasi-canonic passage of the second movement of App
which 46 string parts (24 + 8 + 8 + 6) playing four different l
but then immediately get out of phase, seems to be the fi
integration. For extensive critical commentary of widely v
Ligeti's work up to 1965, see Erkki Salmenhaara, Das musik
seine Behandlung in den Werken 'Apparitions,' 'Atmospher
'Requiem' von Gyirgy Ligeti, trans. from the Finnish
Forschungsbeitrage zur Musikwissenschaft, Vol. 19 (Regensbur
Nordwall, Gy6rgy Ligeti: Eine Monographie, trans. from t
Eppstein (Mainz: Schott, 1971), pp.9-77.
43. The canonic structure of this passage is also discussed, incomp
JONATHAN W. BERNARD
Febel in 'Gyorgy Ligeti: Monument - Selbstportrait - Bewegun

Klaviere)', Zeitschriftfiir Musiktheorie, Vol. 9, No. 1 (1978), pp.35
2 (1978), pp.4-13.
44. For another analysis that explores large-scale symmetry in Liget
Escot, "'Charm'd Magic Casements": Gyorgy Ligeti's Harmonie
Lines: Issues and Ideas in the Music of the '60's and '70's, ed.
(Lanham: University Press of America, 1985), pp.31-56.
45. Hausler, p.108.
46. Stiurzbecher, p.44.
47. Ligeti, 'Auf dem Weg zu Lux aeterna', Osterreichische Musik
(1969), pp.80-8.
48. Varnai, p.44.
49. Ibid., p.29. Ligeti calls this harmony a 'typical Ligeti signal' a
fourth made up of a minor third and a major second or the other
is, [3][2] or [2][3], which formulation seems to confirm that the i
equivalence is valid in general for analysis of his music.
50. The reader's attention is directed to two other analyses of L
Gottwald, 'Lux aeterna: Zur Kompositionstechnik Gy6rgy Lig
25 (1971), pp.12-17; Robert Cogan, New Images of Musical So
Mass.: Harvard, 1984), pp.39-43. Cogan's spectrum photos seem
coincide with my graph, although detail in the photos is difficult
certain points.
51. 'Auf dem Weg zu Lux aeterna', p.86.
52. There are close relationships between the canonic strands of othe
although they vary as to order within the pieces and as to level of
some strands and/or parts of strands in each are unique.
53. Bruce Reiprich, in 'Transformation of Coloration and Density
Lontano', Perspectives of New Music, Vol. 16, No. 2 (1978
discussed this orchestrational segmentation. However, Reipr
matters unnecessarily by referring to each of the segments -
separate canon.
54. 'Ligeti Talks to Adrian Jack', Music and Musicians, Vol. 22, No. 11 (1974), pp.24-
30.

Ligeti Inaudible Structures

Uploaded by

Copyright:

Available Formats

Ligeti Inaudible Structures

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ligeti Inaudible Structures

Uploaded by

Copyright:

Available Formats

Inaudible Structures, Audible Music: Ligeti's Problem, and His Solution

Author(s): Jonathan W. Bernard

INAUDIBLE STRUCTURES, AUDIBLE MUSIC:

MUSIC ANALYSIS 6:3, 1987 207

quantification applied equally within the various areas produced,

The individual character of the various serial arrangements fades

The relegation of resultant sound almost to the status of by-pro

208 MUSIC ANALYSIS 6:3, 1987

qualities he had noted were in fact the true determinants

Technically speaking, I have always approached musical texture through

MUSIC ANALYSIS 6:3, 1987 209

performance space - but still these expressions have an inher

210 MUSIC ANALYSIS 6:3, 1987

The technical process of composition is like letting a cry

The analogy to crystallisation expresses something else as

MUSIC ANALYSIS 6:3, 1987 211

the arrangement shown to the right of Ex. 1. This is not mirror-sy

Ex. 1 First String Quartet, bs 521-33

141 171 171 141 141

212 MUSIC ANALYSIS 6:3, 1987

Ligeti's own remarks about Apparitions emphasize the ide

The states are broken up by suddenly emerging events and a

These formal ideas are evidently combined with fixed re

MUSIC ANALYSIS 6:3, 1987 213

Ex. 2 Apparitions, I, bs 1-23

Total spa qo i teal

Intervals that are present in essentially unrealized p

214 MUSIC ANALYSIS 6:3, 1987

[11] into the same spatial relationship that they occupied in

1) indirect or oblique statement;

MUSIC ANALYSIS 6:3, 1987 215

[1]: (4); (5)

216 MUSIC ANALYSIS 6:3, 1987

same sort of grid notation employed for Apparitions. A gra

MUSIC ANALYSIS 6:3, 1987 217

Ex. 3 Atmospheres, bs 1-29

ii ' .-.: ' ' Si. ..

:. ;.. .+ +ii s ~ i. i ii !i i~i.. . . . . . . . . .

A.... . . ...... ... --.. . .... ...

218 MUSIC ANALYSIS 6:3, 1987

Ex. 4 Atmospheres, bs 13-20, winds

MUSIC ANALYSIS 6:3, 1987 219

220 MUSIC ANALYSIS 6:3, 1987

Ex. 5a 'Bewegung' (3 Pieces for 2 Pianos, III), bs 49-63

MUSIC ANALYSIS 6:3, 1987 221

Ex. 5a is that none of the four canonic pairs is confined to a singl

222 MUSIC ANALYSIS 6:3, 1987

Ex. 5b 'Bewegung', bs 49-63

MUSIC ANALYSIS 6:3, 1987 223

Ex. 6 Kammerkonzert, III, bs 1-12

(call llit Stranld)

a) (b) (c) (d)

Ex. 7 Lux aeterna, bs 1-37

2 1 I I1 7 9 I, I ' ;1 15 1i7 12 0 '

224 MUSIC ANALYSIS 6:3, 1987

gradual transformation as a principle of harmonic constru

i . . . . . . ... . . . ... ..... ..

J tit T till. llll?I?l~?..:r~l

MUSIC ANALYSIS 6:3, 1987 225

available semitonal space is completely filled. As in the previous example, from