Hypernaturalism™ and Junk-o-phonics
by David Thomas
I don't willingly use eq in the mix stage, and as little as possible in the tracking phase. You can add or subtract 15db anywhere across the frequency range and it cannot / does not / will not affect the essential nature of the sound. I noticed this very early on in my career but it took decades to understand the significance. I don't believe in torturing sound waves for convenience sake. (Paul Hamann, my long-time engineer, I leave to manipulate the EQ but he has the good graces to do it while I am distracted.) Over the years Paul has invented and built a number of specialized junk-o-phones for my use, microphone substitutes. These junk-o-phones are most often speakers of various sorts. The technologies of a speaker and a microphone are close enough to be considered identical, requiring only a little electronic tweaking and load-balancing to turn one into the other.
Early in the 90s, I realized that I was, essentially, tone-deaf, and always had been. This was an epiphany. It dawned on me that I didn't hear sound like others hear it. How was I to know? We all assume that we hear or see or sense like everybody else does. Why wouldn't you think that? I'd certainly had a reasonably successful career spanning decades as a significant musician. I started to experiment on myself and soon reached the conclusion that if I listen to two radically different frequencies sounding in an acoustically sterile space I simply cannot identify which one is a higher or lower frequency than the other. Characteristics of one are identifiable in the other. The harder I study one the more it sounds like the other. But if these two frequencies are sounding in real space, a room, an environment, with idiosyncratic acoustic features then those sounds take on a complexity and personality and are clearly identifiable not just as frequencies but as shapes and geometries laden with meaning. I began to suspect that I was synaesthetic, which is a theoretical condition in which the senses of certain individuals are scrambled. Color is sensed as a taste. Sound is sensed as a visual. I soon realized that sound, for me, was a geometrical function. I sensed it through shape, space and perspective. I could 'see' it and that by 'seeing it' certain kinds of meaning, significant relationships, were revealed to me.
The 'masonic' secret of the audio craft, what I call the Edison Principle, can be stated simply: The Way In is the Way Out. These junk-o-phones have names like The Box, The Fly's Eye, The Horn, The West, The 15, The 18, The Phone, and The Fan. The Box and The Fly's Eye are the most effective and most often used. Other less successful models have come and gone. Paul also worked up a way to use flat surfaces, like doors and windows, to act as microphones. These junk-o-phones are designed to capture the sound of an instrument or voice in a limited or idosyncratic way. I record most instruments with many different junk-o-phones. I use traditional microphones, usually, only for room / ambience recording. Typically, the drums will be recorded with 20 or more inputs though rarely do I use more than 4 - 8 of them for any one song. If I feel that an instrument in the mix stage needs to sound differently I alter the mix balance of the junk-o-phone tracks. Transience makes me nervous so I like there to be gaps in the transience, the "space" between notes. Junk-o-phones are especially useful for this. I don't like using store-bought audio effects. On Why I Hate Women I think I used only one such unit, a Lexicon on a minor backing vocal part - mainly because we were in a hurry and it was a small part. Otherwise all effects are from spatial acoustics, spring reverbs from an old Hammond B-3, Altec passive filters, the junk-o-phones and sometimes Suma's echo plate - again, most often when we're in a hurry, otherwise I simply broadcast the track into a room or space that will give me the reverb I want and re-record it. I like sounds to have plenty of spikes. If I use compression at all I only use it full on, hard-crunching, and unsubtle. If you dare to mess with sound then beat it into a bloody pulp, don't toy with it. But do it at peril to your mortal soul. Be convinced. Otherwise, don't torture the poor creature.
The three Drive He Said albums were remixed from scratch using up-to-date technology and techniques. Also significant is an increasing fluency with the 'black arts' of what's happening up beyond the range of human hearing.
Over the span of years, 1994 to 2002, I set out to acquire and master tools of production. At the same time, engineer Paul Hamann and I experimented with differing methods of production.
Raygun Suitcase was the last album to be recorded in its entirety on Suma's Ampeg 24-track 2-inch tape machine. The recording relied heavily on the use of junk-o-phones - the system based on the stratagem 'The Way Out is the Way In.' A variety of speakers and broadcast enclosures were rewired by Paul to function as microphones. These included:
Paul added a vari-speed control to a box fan. On one side of the fan would be the sound source and on the other a microphone or a junkophone. The fan rotation would be adjusted to match a song's tempo. The Fan is prominent on 'Turquoise Fins' and 'Don't Worry.'
A metal high end speaker salvaged from a PA cabinet.
A wooden sound reinforcement enclosure from an old radio cabinet.
The Fly's Eye:
A transistor radio speaker.
The 15 and The 18:
Speakers pulled out of guitar amps.
There were other attempts at devices. We tried to turn a door into a microphone but it didn't give us anything desirable. I wanted discrete bands across the frequency spectrum. Suma has two Altec passive filters which we've used for decades. I wanted to expand the idea and nullify transients.
Junk-o-Phonics produces a degree of undesirable collateral noise. As well, I found the analog tape / mix desk interface bewilderingly restrictive of where I wanted to go with the song construction for the Raygun Suitcase sessions. The entire album was recorded to a click track, in the hope that Scott Krauss, who had quit the band, would reconsider. That added a level of uncertainty. On the last weekend of the session, Scott Benedict came in and recorded the drum parts in one of the most inspired and professional studio performances I have witnessed. He retired from music soon after to become a landscape gardener.
For the box set reissue, Paul transferred the RGS multitrack tapes to the digital realm at a 192khz / 24-bit resolution. I went through the individual tracks cleaning them. I don't like the sound of a gate, which is the automated way to eliminate undesirable noise, so I shaped every snare, every tom and foot drum hit, etc., by ear. Professionals were bemused - I was clearly nuts to spend weeks clearing out stuff no one would hear. My response is and was, 'It's there and if you pile up all the stuff no one is ever gonna hear, you're either gonna hear it or it's going to mask the listener's ability to see all the way to the bottom. Gating imposes a corrosive uniformity.'
The remix process for both 'Raygun Suitcase' and 'Pennsylvania,' which was digitally recorded at 44.1khz on daisy-chained DA88s, uncovered 'lost' vocal and instrumental parts. Some of these were reinstated, including the original vocals to 'Electricity,' outro vocals to 'Woolie Bullie' and other bits and pieces. Edits were made to a number of songs on both albums for one of two reasons: (1) to shorten the songs, when appropriate, in order to achieve an overall length suitable for vinyl, and (2) to tighten song structures in ways that were beyond my abilities at the time of the recording.
Pennsylvania was Steve Mehlman's debut album and the drums were recorded with the standard ten to fifteen inputs. Some junkophonics were used, as well as cardboard boxes substituted for toms in some instances. We nixed The Fan because we'd already done it to death. St. Arkansas, on the other hand, was recorded with just four drum inputs, one of them being The 15. Somewhere along the way I started nagging Steve about his "damn cymbals" - why he had to use them at all and what was the point of all that noise splashed everywhere. This was an issue we eventually worked out by using muted dummies, but that took ten more years. Along the way we agreed to disagree, though, as always, I was right and he was wrong.
By the time we got to 'St Arkansas' I was up to speed with the sequencer I was using, Digital Performer, and a home studio, the two of which gave me time to think about what I was hearing and the facility to record vocals when I felt like it. As a consequence, the remix of STARK yielded no alternative versions or out-takes.
I work fast and intuitively. Being able to study at leisure what just happened makes all the difference. In 1982, I spent fifteen hours recording the vocal to 'Semaphore' from the album Variations On A Theme (1983). That's fifteen hours straight, one take after another without a break. I remember a vast stretch of time trying to punch in for an 's.' I started at 11 in the morning and ended at 2 the next morning. Paul was a wreck. I'd probably gotten it within fifteen minutes but I couldn't know that. These days I do vocals at home in no more than five takes, the first to get levels and voicing, the second and third produce the bulk of the final composite vocal track and with the fourth and fifth takes I try to pickup weak words or phrases, usually without much success. If necessary, I will manufacture words and phrases out of individual syllables, if I have to. I'm good at that sort of thing - hours and hours of brain numbing detail work.