I’m trying to find a satisfying set of words and sounds to use as cues for my orchestra piece at the moment. Each cue needs to be the name of something that makes or describes a sound, and then I also need a sample for each sound. For the past month I’ve been thinking through using objects as well, but the orchestra players will only respond to sound (spoken text or the sounds themselves), so there seems little purpose in using objects in addition. I’m using multimodal cues, so the noun representing the sound sources (dog, gun etc) and samples of the sounds themselves. Although objects would add a third mode, given most of the players won’t see them they add little to the samples and make the associations and practicality of the piece unnecessarily complicated. In a previous version of the piece I was exploring projected images too, but it’s going to be on the radio and again the logistics made this less workable. That might be in future piece though.
My problem at the moment is finding rich associations between sound sources. I’m focusing on words that combine well, such as car, door, bell and alarm. This allows the cueing players to set up chains of association, such as car-door, alarm-bell, car-alarm, door-bell. This will, I hope, add to the playfulness of the patterns that emerge from the cues. The difficulty is finding words that relate to more than one other word in the group and have immediately recognisable sounds. So some sounds, like ‘chicken’, are really clear, but have fewer associations with other words. Conversely, some words, such as ‘wheel’, link to lots of other words but are harder to represent clearly through short samples. I’ve found a very useful website to help find word pairs which has made the job a lot easier, but I’m coming to the conclusion that across 24 cue words/sounds, it’s going to be difficult to create a very rich network. It might also be good to have a few outliers in any case that are chosen for their sonic and semantic impact.
I’ve been using Omnigraffle to graph the connections, focusing on categories of sounds (animals, tools, transport etc), and drawing direct links between words (fly-paper) and more general similarities (gun, cannon). It’s a very useful piece of software as it redraws the network every time you make a new link.
I think I’m nearly there with this, but it’s surprisingly more complicated than I anticipated.