The Shape of the Voice 1: Milton Babbitt

- 100 Warren Burt: The Shape of the Voice 1: Milton Babbitt I only had one extended encounter with Milton. It was in 1974 at UC San Diego, at the Center for Music Experiment, when we held a Schoenberg Symposium. Milton was one of the featured guest speakers, and as one of the Graduate Assistants there, I was heavily involved with the organizing and running of the Symposium. Milton gave an amazing talk about Schoenberg’s work one afternoon during the symposium. It ran well overtime – no one wanted it to end – and I recorded it all. I remember afterward that Kenneth Gaburo wanted a copy of the seminar, so I dubbed it off to cassette and gave him a copy. Years later, that copy emerged in his archives, found by David Dunn, who copied it to mp3 form, and emailed it to me. Listening to the talk, old memories came flooding back, as well as familiar voices during the question and answer period. After the talk, I was walking somewhere with Milton, and I asked him if any of the segmentation ideas he had been discussing would have any application to microtonality. Without missing a beat, he began telling me about Joseph Yasser’s Theory of Evolving Tonality, and had I heard of it, and I should get a copy and read it, but basically, Yasser says….. What was supposed to be a 10 minute walk across campus, bringing a guest to one place or another, turned into an extended personal seminar. This, mind you, after he had given a 3 hour seminar on Schoenberg. Wow, was I privileged to get that. Eventually, I did find a copy of Yasser’s book, read it, and Milton was, of course, right. Yasser’s ideas were indeed, at least in rudimentary form, a source of how to apply segmentation to various microtonal scales. (Years later, studying Erv Wilson’s papers, I found that he, too, had come to the same conclusions about Yasser’s significance for applying segmentation to microtonal scales.) Somewhere along the line, it might be from that seminar, or it might be from somewhere else, I acquired a recording of Milton saying: That marvellous relation called an interval which is unique almost entirely to the perception of sound and nothing else.” The sound quality of the sample is pretty miserable, but no matter, it was the content that counted, and I’ve always liked working with lo-fi anyway. Recently, I’ve been working a lot with graphics-to-sound conversion programs, especially those at the lower end of the cost spectrum. Two of those programs are Photosounder by Michel Rouzic (www.photosounder.org) and AudioPaint by Nicolas Fournel (http://www.nicolasfournel.com/ audiopaint.htm). Both enable you to do different things. Photosounder does a very nice FastFourier Transform analysis of sound, and you can save the analysis as a graphic to import to other programs. Photosounder enables you to do a resynthesis of a sound using any sample as your waveform. Together, they make a pretty powerful pair. It is also possible in Photosounder to horizontally flip a spectrum and to rotate it by 90, 180, and 270 degrees. The combination of these controls gives us 8 transformations of our original material. Not just O, R, I, and RI, but also the “flipped” versions of those. This work came about when I was investigating what “inverted” spectra would sound like (Rotate 180 degrees, followed by Horizontal Fiip). I placed the sample of Milton’s quote into the program, and inverted it, and transposed it a little so that the majority of the partials were in a mid-range audio area. The sound was quite engaging. I then stretched the sound to last 6 minutes, and the results were even more striking. A lot of the stretched and inverted voice sounds were quite noisy. I began thinking about the idea of interval, and decided to try transposing the sound, and combining it with the original stretch, to see - 101 if I could hear a quality of interval with moderately noisy sounds. I took the 6 minute stretched-inverted sound file. I made two copies. One remained unaltered except for...