Performance Practice
of Electroacoustic Music

György Ligeti



Glissandi was realized in 1957 at the Cologne studio of WDR (Westdeutscher Rundfunk). It was Ligeti’s first electroacoustic piece. Ligeti himself called the piece an “exercise” (Ligeti, Gesammelte Schriften, Vol. 1, p. 246). While there is no realization score for the piece, some sketches exist (Paul Sacher Stiftung, Basel, Sammlung György Ligeti). The sound material consists of synthetically produced impulse sounds and glissandi presented in different registers, in different speeds and dynamics that are generally developed through repetition and variation. Doati (1991) suggests that the use of glissandi as musical material might on one hand already reflect Ligeti’s interest in the articulation of speech that was to be at the core of his next electronic composition Artikulation (1958), while also acting as a means to leave behind melodic serial formulations, thus taking on a similar function similar to that of the cluster (Doati (1991), pp. 81). Doati identifies five different qualities of glissandi with specific sonic and movement properties that are presented already in the first two minutes of the piece (op. cit. p. 82). The fact that the piece bears the name of its musical material might furthermore point to an experimental approach, the composer not yet having fully mastered the technique involved (op. cit. p. 83). Formally, two roughly symmetrical parts can be distinguished. In the first part, an increase in density leads to a climax at the connection point. This is marked by acute impulse sounds starting numerous short glissandi. The second part is a heavily filtered mirrored version of the first part with additional variations of materials and motifs. Doati’s analysis of the piece, taking into account the perceivable musical elements and events (and methodologically basing itself on a theory on organization processes in music listening by Stephen McAdams (op. cit. p. 80)), proposes a segmentation yielding a three-part structure. According to this, the first main part (0-3’45’’) is divided into four subsections of differing content, the second part (3’45’’-4’10’’) acts as a short transition to the third part (4’10-7’32’’), which takes up “formal elements” and fragments from the preceding two parts and has a long pause at one minute to the end, thus mirroring the pause at the beginning of the piece (however, Doati doesn’t talk about the overall mirroring of the first part) (Doati (1991), pp. 84).


Performance material

The performance material was received from Schott on an Audio CD. The source tape for this digitization is not identified. The audio file has two identical mono tracks at 44.1 Hz/16 Bit. The duration is 7’35:43. The recording is split into 2 tracks, probably due to IDs having been automatically set at pauses. A hum is heard at 50 Hz (+ harmonics); it is not constant but changes according to cuts. It is more present in the second part of the piece, using modern, full range loudspeaker systems with subwoofers it becomes critical. (Experts’ opinions point out that older loudspeaker systems without subwoofers didn’t reproduce lower frequencies accordingly). The material has many dropouts and “scratches”. However, sound quality is clearly superior to that of the commercial releases.


Archival material

-The Paul Sacher Stiftung, Sammlung György Ligeti, has some sketches for Glissandi, as well as a cassette tape that was used by Roberto Doati for analytical purposes.

-The Institute of Sonology (Den Haag) has three copies of Glissandi made on request at the WDR for the CD-release Cologne – WDR: Early Electronic Music (BV Haast Records CD 9106, 1992). They are marked “Original”, “Ersatzoriginal” (replacement original) and “Hauptkopie” (main copy). The latter is 15 ips, the others are 30 ips tapes. The “original” tape has presents alien drumming noises and dropouts in the beginning, however, it has the best sound quality of the three. Besides some sizzling noises before the music starts, the beginning of the “Ersatzoriginal” is better than that of the “Original” and its quality better than that of the “Hauptkopie”.  All three copies contain the 50 Hz hum mentioned before, which makes it obvious that it is a byproduct of the production process. The copies were made using the Telcom noise reduction system. A new digitization of these tapes was carried out by Kees Tazelaar within the PPEAM project.



A first attempt at restoring the audio file provided by the publisher proved unsatisfactory due to much tape noise. This was especially critical at soft attacks and decays of glissandi in piano and pianissimo at the transition to and from the noise floor. Restoration was then realized using the iZotope RX4 software. In a first step, the 50 Hz hum was eliminated using four harmonically tuned notch filters, Linear Phase, -48db, -40db, -35db, -30db. The goal was to find a setting that would not weaken the bass sounds (e.g. at 2:37 etc.). A noise profile was then established. The first de-noising had the side effect that the high frequencies in some glissandi were less brilliant (e.g. at 0:40). Also, decaying sounds and reverberation had changed too strongly in some places. The noise profile thus had to be modified listening to the critical passages. Clicks and dropouts were then removed. Low clicks (probably analogue errors) and dropouts presumably from an earlier DAT tape transfer had to be manually corrected, since the music is too “noise-like” for the algorithms to work properly. Finally, the signal level was normalized.

With respect to the available audio sources for Glissandi it would seem important to locate and digitize the first generation “Original” and “Ersatzoriginal” tapes. Judging from the quality of the copies of these tapes (s. above, archival material at the Institute of Sonology), it may be necessary to combine passages from the “Original” and “Ersatzoriginal” tapes in order to achieve the best possible quality This was done within the project with the digitized copies of the WDR tapes mentioned above (s. below “recording”).


Performance in concert

There is no documented performance practice of Glissandi by Ligeti himself or people associated with the WDR at the time. Although it was not common for many composers to realize a live or prepared spatialization of a monophonic tape composition, a dynamic sound projection through multiple loudspeakers is not to be ruled out. There are compositional features that can be highlighted and/or differentiated through such a performance: Movement of the glissandi (register, direction, speed); attacks of impulse sounds; specific motifs and their variations; distinguishing near and far with respect to strongly filtered sounds; dynamics. Such an approach was chosen for the concert on January 29, 2015. It was realized using more than 20 speakers on two levels but keeping a stable sound image in the center throughout the piece. The audio used was the restored publisher material.


Recording in surround format

The project’s CD release (Les espaces electroacoustiques / Col Legno, 2016) features a studio realization in 5.1. It follows a slightly different, more cautious approach: while the center loudspeaker builds a stable monophonic basis as well, that was broadened in more complex, polyphonic passages using side and surround speakers, only selected events like the impulses sequence in the beginning (0’49”-1’03”) or successions of single phrases (e.g. after 4’22”) were distributed asymmetrically in space or spatialized using movements (e.g. 3’48”-3’55”). The audio used was the combined and restored version from the WDR tapes.


Selected Bibliography

Doati, Roberto, György Ligeti’s Glissandi: An Analysis, in Interface, Vol. 20 (1991), p. 79-87.

Koenig, Gottfried Michael, Ligeti und die elektronische Musik, in György Ligeti: Personalstil – Avantgardismus – Popularität, hrsg. von Otto Kolleritsch, 1987, Wien, Universal Edition.

Levy, Benjamin Robert, The electronic works of György Ligeti and their influence on his later style, Ann Arbor, Michigan: UMI Dissertation Services, 2006.

Ligeti, György, 1968, Auswirkungen der elektronischen Musik auf mein kompositorisches Schaffen, in Gesammelte Schriften, ed. Monika Lichtenfeld, Mainz, Schott 2007, Vol. 2, p. 86-94.

–, 1980. Musik und Technik, in Gesammelte Schriften, ed. Monika Lichtenfeld, Mainz, Schott 2007, Vol. 1, p. 237-261.

– ,1983, György Ligeti in conversation with Péter Várnai, Josef Häusler, Claude Samuel and himself, London, Eulenburg.

Nordwall Ove, Ligeti-dokument: brev, skizzer, partitur, kommentarer, om musikteater, om musikalisk form, verklistor, Stockholm: Norstedt & Söners, 1968.

–, György Ligeti: eine Monographie, Mainz, Schott, 1971.

Toop, Richard, György Ligeti, London, Phaidon Press, 1999.


Schematic Overview

Piece information

Composer György Ligeti
Title Glissandi
Year 1957
Studio WDR Köln
Type Tape, 1-channel
Duration 7’36’’
Publisher Schott


Performance material

Renter Schott
Format CD
SR/Bittiefe 44.1 kHz/16 Bit
Duration 7’36’’


Archival material

Paul Sacher Stiftung Sketches, audio cassette copy


Commercial releases (selection)

György Ligeti (1998) Wergo 286 902-2
Acousmatrix 6: Cologne-WDR: Early Electronic Music (1991) BVHAAST CD 9106