Paris and Musique Concréte

Creación de objeto sonoro

In 1952, Schaeffer finally published a definitive syntax for musique concrete in the form of a
treatise entitled "Esquisse d'un solfege concret." This appeared as the last section of a book, A
La recherche d'une musique concrete,8 which outlined the events of the previous four years.
The treatise is divided into two main sections.

The first consists of a set of twenty-five provisional definitions for use in the description of objets
sonores, and the basic processes that might be applied to them, while the second is concerned
with the application of these definitions to create an operational language for the synthesis of
musique concrete.

The twenty-five provisional definitions may be summarized as follows:

1. Prelevement, concerned with the initial action of creating a sound and then recording it on
to disc or tape.

Faq

Any such sound event (objet sonore) is then classified in two ways, each associated with its
own set of definitions:

A
2. Classification materidle des objets sonores, the material classification of sounds prior to
any aesthetic or technical analysis. This classification is based on the temporal length of
each sound and its center of interest. Three classes are identified:
3. Echantillon, a sound lasting several seconds or more with no clearly defined center of
interest.
4. Fragment, a sound lasting one or perhaps a few seconds with a clearly defined center of
interest.
5. Elements, short extracts isolated from a sound, for example the attack, decay, or part of the
main body of the event.

B
6. Classification musicale des objets sonores, value judgments on the nature of sounds, in
particular their degree of complexity. Four classes are identified:
 7. Monophonie, concomitant elements isolated by the ear from an accompanying texture.
Schaeffer draws a parallel with the subjective ability to identify a melody within a
polyphonic texture.
8. Croupe, a monophonie of some significance lasting many seconds, which may be studied
for its internal development or repetitions. A groupe, by definition, is constructed from
cellules or notes complexes:
9. Cellule, thick sound complexes with no overall shape, involving rapid changes of rhythm,
timbre, or pitch, or complex combinations of notes that cannot easily be discerned.
10. Note complexe, any element of a monophonie that displays a sufficiently clear envelope
(attack, body, and decay) to be equated to a musical note. Schaeffer adds a rider to the
effect that the element also must be of a simple nature.
11. Grosse note, a note complexe in which the attack, the body, or the decay is of a significant
duration. Beyond certain limits, a grosse note must be treated as a groupe.
12. Structures, the ensemble of material with which the composer starts his examination. This
may consist not only of cellules or notes complexes but also of ordinary notes, prepared or
not, obtained from classical, exotic, or experimental instruments.
The next group of definitions identifies the operations involved in processing the sound
prior to the main task of composition:
13. Manipulations. Three types are identified:
14. Transmutation, any manipulation of the material that leaves the form essentially unaltered.
15. Transformation, any manipulation that alters the form of the material, rather than its
content.
16. Modulation, any manipulation that is not clearly a transmutation or a transformation, but a
variation selectively applied to one of the three attributes of pitch, intensity, or timbre.
17. Parametres caracterisant un son leads on from definition (16) to propose parameters for
the analysis of concret sounds. In place of the classical notions of pitch, intensity, and
duration, Schaeffer substitutes the idea of:
18. Three plans de reference, which describe the evolution of each of these quantities as a
function of one of the others: pitch/intensity, pitch/duration, and intensity/duration.

The importance of these plans merits a close examination of their characteristics, and these will
be returned to in due course.
The next group of definitions describes the primary processes involved in realizing a piece of
musique concrete:
19. Precedes d'execution. Six operations are identified, the last three being concerned with the
spatial organization of the material in its final realization:
20. Preparations, the use of classical, exotic, or modern musical instruments as sound sources,
without any restriction as to the mode of their performance.
21. Montage, the construction of objets sonores by simple juxtaposition of prerecorded
fragments.
22. Mixage, in contrast to montage, involves the superimposition of monophonies, to create
polyphonic textures.
23. Musique spatiale, all music that is concerned with the projection of objets sonores in space
during a public performance.
24. Spatialisation statique, the projection of clearly identifiable monopsonies from specific
locations. This feature arises from the use of different channels on the multitrack tape recorder
for the distribution of information at the time of mixage.
25. Spatialisation cinematique, the dynamic projection of objets sonores during performance
using the potentiometre d'espace.

These definitions by their very nature could only serve as generalizations of the various
processes involved in the earlier stages of musique concrete. The whole solfege was subjected
to significant change as Schaeffer's work continued, and ultimately consolidated in a formidable
work, Traite des objets musicaux, which appeared in 1966.10 This synopsis nevertheless gives
a useful insight into the philosophical principles applied during the period of its gestation.
Indeed, the three plans de reference have a more lasting significance that extends well beyond
the limited sphere of concret composition, for they are germane to any psychoacoustic study or
synthesis of sound material.
Schaeffer defined his plans as follows:

1. Plan melodique ou des tessitures, the evolution of pitch parameters with respect to time.
2. Plan dynamique ou des formes, evolution of intensity parameters with respect to time.
3. Plan harmonique ou des timbres, the reciprocal relationship between the parameters of
pitch and intensity represented as a spectrum analysis.
These three plans may be combined as follows:

The result highlights the problems encountered in creating a visual representation of sonic
events. Although several attempts have been made to improve on this model over the years,
the impossibility of reducing such detail to a single two-dimensional graph has proved a major
stumbling block. The model cannot be expanded to embrace multiple objets sonores, and is in
itself only capable of limited accuracy. The plan harmonique, for example, only provides an
accurate indication of timbre at one selected instant during the course of the event. If this
spectrum changes significantly several extra plans harmoniques might be required to represent
the timbre at different stages in the evolution of the objet sonore.

Schaeffer was only too well aware of these difficulties. His solution was to construct a syntax
that was based on a limited number of descriptive criteria for each plan. This involved a
rationalization of the seemingly infinite range of sonic possibilities into categories that were
neither too specific nor too vague. His approach was based on the following line of reasoning:
In the strictest sense, it is impossible to give a simple description of the evolution of pitch with
respect to time unless the sound under examination is exceptionally pure. As noted earlier, a
thorough description would demand the superimposition of the plan harmonique both on the
plan melodique, to obtain a frequency/time graph of partials, and also on the plan dynamique,
to obtain an amplitude/time graph of partials. Such an outcome destroys the whole purpose of
the simplified analytical model.

Fortunately, the psychology of perception offers a viable compromise, for the brain, when
evaluating the quality of a sound at a selected instant, takes into account the acoustic
phenomena that immediately precede it. Indeed, there is a minimum sampling time necessary
for the comprehension of any sonic event. Experiments have shown that sufficient information
is contained in extracts of the order of about one twentieth of a second for the brain to identify
any center or centers of pitch interest with some degree of certainty. Lengthening the analysis
interval permits the ear to study the changes in these centers with respect to time.

Schaeffer's approach is of considerable significance, for it focuses attention on aspects of

psychoacoustics that are an essential part of any study or manipulation of sound material,
whether natural or electronic in origin. He made an important distinction between two very
different elements regularly encountered in objets sonores: (1) the complex spectrum
associated with a sharp attack or an abrupt change in content, and (2) the more ordered, slowly
changing spectrum usually associated with the body and the decay. The latter characteristic is
particularly clear if the objet is a note with a definite pitch center. The former characteristic is
often described as a transient response, an important feature in many natural musical sounds.
One of the major problems of all-electronic synthesis even today is the difficulty encountered in
creating satisfactory transients, and this key aspect will be returned to in due course.

During attack transients the spectrum table is extremely complex, so much so that a plan
harmonique drawn during this particular stage of a sound would be most misleading, for its
content will be undergoing rapid changes only partially comprehended by the ear. The spectral
elements are in many instances so disordered that the result is a semicontinuous spectrum of
noise, indicated on the plan melodique by a wide shaded band or bands of frequencies. The
body and the decay, by contrast, are often sufficiently stable for a much narrower band or
bands to be drawn, in particularly clear cases reducing to a line. Schaeffer thus proposed that a
single plan harmonique for an objet sonore should be drawn during the body of a note, at the
point where the spectrum reaches its greatest state of development. The preceding diagram
illustrates the use of the three plans to identify the salient features of a sound of moderate
density, displaying three predominant areas of partials after an initial transient. It also reveals
that the decay is characterized by a more rapid attenuation of higher partials relative to their
lower counterparts.

Five principal criteria were proposed for evaluating the plan melodique, to be associated
specifically with the pitch characteristics displayed during the body of the objet sonore. These
were: (1) stable, displaying a fixed pitch characteristic; (2) cyclic, displaying a pitch vibrato of
about 5 to 6 Hertz (Hz); (3) continuous ascent; (4) continuous descent; and (5) discontinuous,
where the pitch flickers in a complex fashion. Suggestions for subsidiary criteria included a
variation on (2), spinning, to describe sounds that fluctuate more rapidly about a central pitch,
and a variation on (5), indistinct, to describe the pitchless quality of uniform noise.

The principal criteria for the plan dynamique were divided into four groups, one for the attack,
two for the body, and one for the decay. Three principal criteria are specified for the attack: (1)
plucked, (2) percussive, and (3) aeolian.
Two subsidiary criteria were suggested for use in describing the artificial types of attack
encountered in the use of concret techniques: (I) stepped, to describe an attack that develops
as a succession of terraced levels, and (2) pulsed, to describe an attack that develops in
successive waves.
The decision to provide two complementary sets of principal criteria for the body of the sound
requires some explanation. Schaeffer clearly felt it desirable to classify not only the nature of
the body itself but also the way in which it develops out of the attack. Six principal criteria were
proposed under the latter heading. These were: (1) shock, no sustaining into a body at all; (2)
natural resonance, the sound sustained by a smooth natural reverberation; (3) artificial
resonance, the same effect created by artificial overlaying; (4) drubbing, a beating continuation
of the attack impetus; (5) pulsation, sustaining by repetition of the attack either sequentially or
by partial overlaying; and (6) artificial, a synthetic sustaining characteristic produced by a
montage of various elements.

Five principal criteria were proposed for the body itself, with the intention that these should be
treated as complementary to the criteria for the plan melodique, discussed above. These were:
(1) stable, steady intensity; (2) cyclic, continuous amplitude vibrato of about 1 to 5 percent; (3)
continuous crescendo; (4) continuous decrescendo; and (5) discontinuous, for example,
stepped or pulsing.
The decay of a sound, concerned with the gradual dissipation of the accumulated energy, was
accorded five principal criteria. These were: (1) cut dead, rapid decay with almost no
reverberation; (2) normal reverberation, a natural exponential decay; (3) artificially extended
reverberation, generally involving a subsidiary peak of reverberant energy; (4) artificially
discontinuous reverberation, sharp inter-ruptions to the natural decay characteristic; and (5)
artificially cyclic reverberation, superimposition of an amplitude vibrato onto the decay.
The plan harmonique, as already observed, provided an analysis of the timbre spectrum of an
objet sonore, most suitably at the peak of its development. Schaeffer's approach was slightly
different for this plan, for he divided his principal criteria into three complementary groups,
concerned respectively with density, richness, and coloration. Four principal criteria of density
were proposed: (1) pure, a single fundamental tone; (2) blurred, a less distinct fundamental; (3)
thick, an identifiable primary area of frequency, but with no clear fundamental; and (4) white, no
distinct frequency area.
Two principal criteria of richness were identified: (1) rich timbre, displaying many partials; and
(2) poor timbre, displaying few partials.

Three principal criteria of coloration were identified, intended to provide a qualitative

assessment, as a counterpart to the more quantitative assessment provided by the criteria of
richness. These were: (1) dark, few partials, rapidly falling in intensity up the spectrum; (2)
clear, few partials, but with a more evenly distributed energy spectrum; and (3) brilliant, a similar
energy distribution to that shown in a clear sound, but with a greater number of upper partials
forming more concentrated groups.
These thirty-three criteria, for all their generalities, provide for some 54,000 different
combinations of sonic characteristics, a daunting indication of the scale of the task facing any
prospective morphologist. The "Esquisse d'un solfege concret" was only intended as a
preliminary treatise, designed to serve the needs of a particular school evolving around the
philosophy of musique concrete. It is pertinent, nevertheless, to note that this pioneering
attempt to study the nature of sound in relation to the electronic studio provided a major point of
reference for those who have subsequently sought to pursue this important area of
psychoacoustic research.