Recording Sessions

Several hundred individuals aged from six to sixty were recorded over several weeks playing 'chinese whispers' in groups of sixteen. Each individual wore a preposterous contraption on their head comprised of a modified set of headphones that was used to mount a lapel microphone. The microphones were located near the ear of each participant. In this configuration, rather than the microphone being located near the mouth, it was instead located at the ear of ones neighbor to optimally pick up the whisper. Each microphone was recorded discreetly to a sixteen channel multitrack. The microphone located in the center of the room (in photographic documentation) was merely for reference.

The number of whispers undertaken per cycle varied but generally 4 were started at the beginning of each round and were passed between individuals until they returned to the initial instigator (For example, In a group of sixteen people every fourth person would start a new message that would be passed to the person on their right and so on...) Each group recording session lasted approximately an hour which was usually the stage that brain melting fatigue overtook the participants, caused by to much stimuli (and sustained laughter). Usually at the end of each group session, 8 whispers were started simultaneously (every second person in the circle) and passed around. With such a mass of information streaming between the participants, many of the transmission outcomes became morphed aggregates of several discreet messages.

Individuals were able to either start their own sentences or were given pre selected statements on small slices of paper as shown below.

As the work to some degree was commenting on the veracity of the media that surrounds us, these statements were gleaned from a variety of sources ranging from current newspaper articles to extracts from the bible. The image below links to some photos of three group recording sessions that took place in the UTS sound facility. In addition to the studio sessions, recordings of over 100 primary and secondary school children were undertaken on location within the schools with a mobile recoding rig. This proved to be infinitely more chaotic than the controlled environment of the studio facility however many exceptional sequences were recorded in this manner. Unfortunately I am unable to provided any documentation of the children to maintain their anonymity...

Composition

A sixteen channel audio system was setup in the Gunnery Studio's for development. The majority of the production process for Flutter only three weeks or so. At this stage of production thousands of whispers within hours of multitrack recordings were edited and arranged. Underlying the whispering sequences was a sixteen channel music composition. Various experiments were undertaken regarding the presentation of several streams of whispers at once within the multi speaker array. The combination of whispers and their intelligibility interestingly relates to what Walter Murch describes as encoded and embodied sound in relation to film sound design. I recommend reading the article as it is a elegant investigation into mixing and auditory perception. The crux as it relates to Flutter, is that evidently in film, two discreet streams of dialogue should be coherently perceived by an audience at a given time. In Flutter however, when two different streams of whispers moved around the speaker array this was not the case. It became difficult to precisely follow a particular message and its subsequent mutations. I believe this may be more pronounced in the spatial audio setup evident in Flutter (as opposed to film sound in which all dialogue radiates from a central mono sound source) where the two different streams of messages where physically separated by considerable space (i.e., one starting at speaker 1 and the other at speaker 8 and then traveling clockwise). One would imagine that this separation would make each stream more defined, but this, in combination with the movement of the sound made it quite difficult to maintain a clear sense of two discreet message sequences as you were both trying to locate and decode two discreet sound sources simultaneously. Like listening to someone having a conversation behind you and in front of you at the same time. No fun at all really...It was practically impossible with tests of four messages simultaneously. Murch states that in the absence of any other sound elements 2.5 tracks of dialogue generally should be intelligible. This was perhaps reduced in Flutter due to the nonsensical nature of the content. In conventional dialogue, we are able to grasp the idea of a sequence (or concept), or get a sense of what is being conveyed, as there is generally a recognizable progression taking place (Much like being able to guess where a conversation is going before it gets there). In Flutter however, each new sequence in the progression is somewhat random and therefore potentially more difficult to comprehend as a discreet entity (when interwoven with several other streams). As we are so used to decoding conversations, when we are presented with dialogue that does not follow these rules things become blurred.

Below is a photo of me looking suitably enthusiastic at this stage of development..

Credits

Thanks to the following people for their assistance in the realisation of this project. Artspace, VACB, Brendan Lloyd, Ali Crosby, Michaela Davies, Phoebe Torzillo, Kathryn Davies, Bel Macedone, the students and staff of Newtown Performing Arts High School and Newtown Primary School, as well as the hundreds of whisperers who were kind enough to momentarily loan me their synapses for the project...