3.3 Words as meaning and words as sound


Many of the artists in Cathy Lane’s book describe their experiences of language as musical sound – especially when they do not understand the meaning of the words (Jaap Blonk, Oliver Brown, Michael Vincent, Leigh Landy)(see Lane 2008). Language as musical sound is something I recognise as being another musical parameter in my work (though it is strongly related to the play with zones). I observe the use of five main techniques when working with text or language as sound:


1. Natural speaking or reciting text.

2. Singing text.

3. Repeating text.

4. Using “text”; sounds without semantic meaning, but “sounding like text”.

5. Processing the sound of the spoken or sung text or “text”.


1. Speaking or reciting naturally can correspond to Bergsland’s central zone, although there are many degrees of intelligibility, depending on the text itself, how it is performed and what place it has in the musical whole (feature salience).

2. From my experience, singing the text involves an abstraction from meaning. It could also involve a change of meaning when compared to speech. The degree of abstraction varies with the way the text is sung. There is, for instance, a big gap between the recitative-like and text-near, and the more complex melodies.

3. Repeating text, both in real time and with sampling, can gradually reduce/transform it from meaning to sound. (I will return to this in Chapter 4.)

4. When describing my music I often use terms such as ”text” and ”word-like”. I speak or sing, it sounds like language, but it has no semantic meaning. I experience this type of expression as “having a hint of meaning”; it sounds as if I mean something concrete. This definitely adds something else when compared to working with pure sound-sculpting, vocalising or using text with semantic meaning.

5. Processing the sound of spoken or sung text can change it slightly (reverb, flanger) from the natural, or change it dramatically (granular synthesis, big changes in pitch) – and in the grades between. In my experience, the sounds’ quality as being meaningful can often be present even if the processing is strong.

I will give an example of how I use these techniques in an excerpt from a live performance with my trio, BOL:

Example III, 4: Excerpt from “Skylab Audiovision”, live at Verkstedhallen, Trondheim, September 2009. The text is from the poem Skylab (Rolf Jacobsen, translation by Roger Greenwald).


Here I register the main techniques listed above, except for no. 1, the natural speech:

– 0.03: Whispering/‟talking”/singing with Ring Modulatoreffect (Roland SP555) with low amplitude, blending with the other instruments – “word like”, a hint of meaning/real world, but towards the peripheral zone.

– 0.48: Sampling something “word-like”, creating a loop and processing in with G/F patch in MaxMSP (granular synthesis with variations made by frequency filter control, see Chapter 2); peripheral zone, but still with a hint of meaning/real world, even when heavily processed.

– 1.44: Singing with a full, and almost natural voice, with words. I think of this as being close to the central zone (natural sound and meaning, but also abstract because I am singing, not speaking).

– 2.04 Repeating a text phrase, sung with a natural voice, sampling parts and playing back the samples repeatedly, in layers. I experience this first as being close to the central zone, singing words with a “natural” voice. The repetition of words (samples in loops) gradually changes them from meaning into sound, moving further away from the central zone.

Next page  ➡

Collapse all Expand all