From a recent email with Dan in which we discuss this audio excerpt he sent [LISTEN]. My comments in bold.
There are two ways in which we can approach this thing I think. I was listening to blackbird fly again. Something just like that except with this text would probably work great. I like the speed of it and its musicality and the granulated sound. So I guess that would be the "music composition" approach.
Obviously I have a bias toward this approach, but in this case I'm not sure it would work (or at least I'm not sure how to make it work). What drew me to Blackbird Fly a a poem was the poetics of it---there are clear moments of emphasis that lend themselves to a musical treatment. I actually wrote out the pitches and rhythms of Blackbird Fly the same way I would set a song. The codex text has some characteristics that work against this. First, it is prose which deprives it of the concise rhythms of poetry (prose has rhythm but is tends to be more vague). Second, the codex prose is at once bland and overblown. It is stated in a matter-of-fact way that doesn't lend itself to poetic treatment and the words themselves are over-the-top (magic, the fluid metals, time of change) in a way that perfectly balances the matter-of-factness when read but sound ridiculous when treated more poetically ("Magic!"). I've heard prose such as biblical text set, but their success often depends on the characters involved (a sort of operatic leaning) and the hieratic nature of the text.
I really appreciate you going through this with me because I couldn't figure out at first why I was confounded by how to treat the text. Everytime I hit the important words as I normally would it sounded silly.
I think if we were to want a more compositional approach, it would essentially have to be in between the action or accompanying action that doesn't further the plot. This is, I think, the basis for the MTV moments in so many TV shows: moody pop song plays while character walks and looks moody and interacts in meaningless ways (plot-wise) with other moody people. The TV show Alias used to do this about 5 or 10 minutes before the end of each show.
The other strategy is the "sound design" approach. In terms of this approach, I have some specific comments.
A. For some reason, size implication is really important to me. The implication of something small, homely, home-made seems important and congruent with the film in general. I think it's also a personal preference (or obsession). When I was using Reason a lot, I was constantly running audio through any Scream setting that evoked small devices (cassette, tinny speaker, small transistor radio, etc.). I ran your audio through something like a telephone EQ and it seemed more 'aesthetically correct' to me I think because of this small device association.
Absolutely. I had planned to run it through a reverb that would give it the sense of being played in a confined space, but your telephone EQ approach takes that kind of idea back to the source itself.
B. If you remember a long time ago we had a discussion about pitch shifting and I guess this sort of follows up on that. I think it's important that there's a sense of a warbly speed, like a turntable with a bad motor. I ran your audio through my Reaktor ap that does exactly this--you draw a curve that speeds up/slows down the audio using a hand drawn curve and that basic sound seemed right. I think the reason this is important is because it reiterates the old school technological feel of the devices--the analog type distortions that might come from a spinning apparatus.
Again, a good idea. My tape speed glitching was extreme and occasional leading to interesting thumps and squeals, but not what would come from an actual physical source.
C. I also tried a version in which I inserted some blank spaces randomly throughout the piece. Breaking up the piece seemed to evoke the tape splicing idea. I know you weren't going for that yet, but in the long run, something that implies splicing would be important from a sound design type standpoint.
This works nicely. My favorite spots in my own version are things like "sssssecret" and the very slow "controlled" because they seem like they're straining to get out (they remind me of the effort that people who stutter put into their speech). Your random blank spaces give it that held-back feeling. There's a little space after "sssecret" that is there because I needed to adjust the rhythm, but it acts a bit like the spaces you mentioned.
D. I was thinking about the idea of withholding. I don't know if you ever read the blog entry I wrote about the green shrine...
You can find it at http://noisefilm.blogspot.com/2006_06_01_noisefilm_archive.html
(see story of the Green shrine)
Here's an excerpt of the important part:
Making the shrine was really educational. It helped me to realize that these props are performers, just like the actors. You even have to think like an actor to make one of these things. Ben was saying that in film acting you have to consciously try to withhold emotions. The same thing is true with the shrine. One of the reasons the first attempt looked so bad is that the shrine looked as if it were saying "look! I'm old!" The more I covered things up, the more I tried to withhold the appearance of age, the better it looked. Sounds so much like Heidegger's idea that entities withhold themselves. I should look that up. You can see the green shrine in the desert pre-vis.
I wouldn't even know how to do this in terms of audio)... but it would be great if we could do the audio equivalent of the green shrine. Right now the audio is sounding purposefully distorted. It should sound more like the sound is just distorted and noisy but somehow, legible words are forcing their way out. Right now the distortion sounds more like someone trying to make the glitches and distortions apparent. It should in instead, sound like it just IS glitchy and noisy and the legibility is pushing its way through despite the best attempts of the noise to keep it back. Does this even make sense??
That makes a great deal of sense. it should feel like the problems are from the source rather than laid on afterward. Part of the aesthetic that I'm (accidentally) bringing that's left over from the compositional approach is the conscious crafting of materials so that the artist's hand is apparent. Despite my appreciation of and what I thought was an understanding of the sorts of ideas that Cage propounded, I guess I'm still a traditionalist at heart----the artist makes the art!
This is, I suppose, what really makes a sound designer (as opposed to a composer): the craft it hidden, the goal is mimesis. I'll try to think this way some more. I realized that I created the glitching and tape speed squeals using statistical approach (random number generators) in Csound, but perhaps this feels like too much of a post-process. While creating the intercut and looking for the words, I was essentially recreating Ben's activities which helps to explain why the intercutting sounds authentic to me (although as you mentioned, there's not enough). I was thinking about inserting blank spaces using Csound, but maybe I'll just do it in Digital Performer some more---that might give it a more hands-on sense.
Perhaps I should kill off my own time stretching (I wrote out all of those as rhythms) and keep Tod's relative pacing (but a touch slower) along with the warbly tape speed. Interestingly, once I started glitching it I barely noticed my own time manipulations---I was too busy trying to hear what the hell was being said through all the glitches. I played some versions for Bekki and she had a lot of trouble understanding the text once the glitches started.
Here's what I'm going to try next:
1. start with the intercut version
2. tape-speed manipulate it so that it is slower overall (but with the same relative pacing) and has gentle changes in speed/pitch
3. use the blank spaces insert idea
4. telephone EQ it
5. see where we are---this should be relatively understandable, but not completely so.
I like the delay, but I need to rationalize it a bit to figure out where it comes from in the process.
No comments:
Post a Comment