Discovering an Electric Violin
by Dan Trueman
It was my first "concert" with electric violin. I had been playing electric violin for quite some time, but always in clubs, with weird rock bands, singing harmonies, using distortion and a big electric guitar tube amplifier. But this was a "concert," in a concert hall with nice acoustics at a conservatory and (mostly) attentive listeners. So I lugged in my Fender amp, my Alesis Quadraverb (a digital effects unit), my MIDI foot-pedal and my flying-V, strap-on 6-string fretted electric violin (made by Mark Wood), and played. Though I grew up playing chamber music, performing with orchestras, and doing "concerts" on a regular basis, it had been many years since I had played in a "concert" setting, and it felt simultaneously foreign and familiar.
I learned so much in that concert. Let me start with the details, because that's where the whole process began for me. I set up the amplifier facing the audience, and sat down next to it with my foot-pedal. Usually, at home or in a club, I'm standing facing my amplifier, getting it full blast in the face. But I had to sit for this (needing both feet for some complicated foot-pedal maneuvers) and didn't really think facing the amplifier would be appropriate I'd be blocking the sound and, regardless of Miles Davis's example, I didn't think I could play with my back to the audience. So I sat down next to the amp and played. It was shocking how different it sounded to me than when I was face to face with the amp. All the high frequencies were gone, so I began compensating by hitting the high strings hard and the lightening up on the low strings. I was also puzzled by the sound of my instrument in the hall. It was clear that people in the central seats were getting a decent sound from the amp, but as I wandered the hall with a long cord, I could hear that anyone outside of the central seating was also missing the high frequencies; what an awful sound!
I also noticed something subtler, but in some ways even more deeply troubling, and it wasn't until several years later that I was really able to put my finger on the difficulty. The acoustic instrumentalists I was playing with on that concert filled the hall beautifully; it was as if their instrument and the hall were one, working together symbiotically. My own instrument, on the other hand, sounded like it was from another world. Oil and water. I messed with the reverb on my amp, to try to get it to sound better. I turned the reverb off, but then it sounded just terrible; dry and ugly, and instead of hearing the warm sound of the hall, all I could hear was the sound of my instrument bouncing off the back wall, like an echo. In the end, it was an intractable problem; oil and water it was, at least for that concert. The only thing I could think to do was to put mics on the acoustic instruments and put us all through a PA, so we could be artificially mixed and essentially bypass the acoustics of the hall, but there wasn't time, and it just seemed like an unsatisfying non-solution.
Finally, my solid-body electric violin sounded painfully coarse in the same room as these fine old acoustic instruments, in spite of my best efforts to tweak the equalization (I have spent untold hours messing with EQ and electric violins). These instruments, with their piezo pickups directly under the strings in the bridges, produce a remarkably clean signal, one that can be quite effective when processed, but one that is decidedly bland and even (to my ears) ugly when compared to most acoustic instruments. And, between my stomping on the foot-pedals to try to control the effects box and negotiating my fretted, undeniably phallic strap-on instrument, I felt a bit like a zoo animal.
So I played the concert (it went ok, though was an unsatisfying musical/sonic experience for me) and spent the next several months wondering if I could ever make "concert" music with the electric violin. Now, I'm well aware of the current trend in concert music to amplify everything, make it loud, and bypass the natural acoustics of performance spaces, and I quite enjoy it at times, both as a player and listener, but I also feel like it shouldn't be the only option when we are interested in combining old instruments with new technologies in spaces that have their own beautiful acoustic properties. I also have a social problem with it; while it is certainly possible to make things sound good in a hall when you have a good system and good engineers, and it is even often possible to arrange for a decent monitoring system for the musicians, it is undeniably a different social experience for the musicians to make music in this kind of context. It is so different that not only is the performer's musical experience deeply impacted, our whole notion of how we make music and what kind of music we will make is changed. This is not just a practical matter, this is a musical, aesthetic matter, and we should have options so we aren't using a PA simply because that is our only choice.
From my point of view at the time, there were Three Problems: 1) the tone of my instrument and EQ; 2) the amplifiers/speakers; 3) the foot-pedal as controller for signal processing. I don't think I would have made much progress on any of these fronts if I hadn't had the opportunity to work with Perry Cook, a computer scientist, engineer, and musician at Princeton, where I became a graduate student the year after my formative concert experience. With Perry, I embarked on two separate projects to address the Three Problems.
For Problem #1, we decided to study the natural EQ that instrument bodies create when they filter the sounds of strings being driven through them. This is, after all, the missing element in a solid-body instrument. We went about taking what are called "impulse responses" for a variety of violins, guitars, and mandolins. Here's the idea: you whack the bridge of the instrument with a very small hammer, and record the way the body colors that sound (the strings are left on, but damped). This response then serves as a model for building an equalization filter that you can then use with an electric violin (or any other sound, for that matter). Engineers have done this kind of thing for years, and it is analogous to the way reverbs are built these days; generate an impulse (via firing a pistol or something similar) in a concert hall, record the way that impulse is dissipated, and that becomes your model for building a reverb (or in the case of some recent reverbs like Altiverb, the response is used directly to create reverb via a process known as convolution). But instead of just recording the response with a single mic, we set up an array of twelve microphones pointing inwards so we could see how the sound of the instrument varied in space.
An array of 12 microphones, with an instrument inside
So we went about whacking the bridges of several instruments and collecting sets of 12 "directional impulse responses" which could be used to create EQs for my electric violin.
Simultaneous with this process, I was thinking about Problem #2. Taking it step by step, I thought things might be improved if I used two speakers: one for the audience and one for me. This way I could hear the high frequencies and feel more comfortable with the sound of the instrument. Of course, it would still sound lousy to the musicians on stage with me. Ok, then let's try three speakers, adding one for the other musicians. Well, you need more than three if you have more than one other musician, so let's keep adding. Finally, it occurred to me that maybe we could set up an array of speakers pointing outwards, arranged like a sphere. Without having any idea how this might be done, I mentioned the idea to Perry and his eyes lit up. A couple days later he invited me over to his house to see something he had built:
Looking like something that NASA might build if they could only shop at Home Depot, the Boulder was the inverse of the microphone array we were using to collect impulse responses: a set of 12 speakers pointing outwards. Now we could take the impulse responses we were collecting, apply them to my electric violin through these 12 speakers and create a virtual replica of the acoustic qualities of the instruments we had abused. Thus our first spherical speaker was born (I should point out that spherical speakers have been used since the 1950s, though primarily for acoustic studies, not as musical instruments).
I noticed something remarkable when I started playing with the Boulder (and the Bomb, a spherical speaker made out of two metal salad bowls, and later, R2, a speaker made out of two wooden salad bowls and tom-tom legs): I didn't need reverb anymore. With the sound of my electric violin coming out all directions, it was filling the hall more like acoustic instruments doI was actually engaging the acoustic qualities of the performance space, and getting the natural reverb that acoustic musicians are accustomed to (and boy was it fun to bring it into the bathroom!). This was particularly true when using the 12 EQs from the impulse responses, but even with a mono signal going through the speaker I was able to get a full, reverberant sound. This was a revelation, and I have been using spherical speakers (or, more recently, hemispherical speakers, which are excellent as well) ever since.
It is difficult to overstate the impact these speakers have had on my subsequent work. After building a few speakers by hand and coming up with a more refined design (with my father, physicist Larry Trueman), I began (with my long time collaborate Curtis Bahn) to pursue manufacturers in the hope of having many of these speakers built. At RPI, Curtis eventually initiated a project where they built over 50 hemispherical speakers, which in turn spawned a mini-business (see http://www.oddnoise.com/spheres.html) intended to meet the requests of several other artists interested in using them. Over the years, my electronic improvisation ensemble interface (with Curtis and the dancer/shakuhachi player Tomie Hahn) has performed regularly with them, sometimes using dozens of them at a time. interface had always performed with a PA system, and it was remarkable how much our music changed when we started using the new speakers. The speakers have a sense of presence in a space; they are definable, localized sources of sounds, like old fashioned musical instruments. They also help create a sense of intimacy rare in electronic music; rather than hurling sound at an audience from behind the plane of separation that PA systems create between an audience and performers, we felt like we could create delicate sonic spaces that invited listeners to lean forward, and even enter our space and walk around. With these speakers, the whole notion of the on-stage monitor becomes irrelevant; the speakers are instruments, and we simply set the volume to whatever sounds appropriate. This has noticeable social ramifications; when we play, it feels more like an old fashioned chamber music experience than the familiar PA/monitor situation. Our improvisations began to change, becoming more detailed, with a wider dynamic range (particularly on the quiet side).
I found all of this tremendously inspiring, and began to compose chamber pieces that combined my electric violin and spherical speakers with cellos, violin, guitars, percussion, etc.... My first electronic chamber piece, Machine Language, was for electric violin (no fancy signal processing or laptop, just a single spherical speaker for the sound of the e-violin), violin, cello and percussion. This was followed by a trio, Still, for violin, cello, and electric violin/laptop (which was premiered at the ACO's first OrchestraTech festival in 2001) that introduced noisy electronic textures into the ensemble (using three spherical speakers, one for the electric violin, two for the electronics, all placed in and around the ensemble). My next piece, Counterfeit Curio, for Pierrot/percussion/electric-violin/laptop, required more speakers to match the size of the ensemble, so I used four hemispherical speakers (by now I had decided that hemispheres were actually more appropriate than spheres, spheres being almost too reverberant). Emboldened, I entered the most challenging of chamber music contexts with the piece Traps, for string quartet and electric violin/laptop; with Traps, the four hemispherical speakers sit discretely on the floor, one near each player, and the electronics create a kind of gentle sonic halo from within the ensemble.
Over the course of composing these pieces, the speakers went through many changes, and I also continued to explore different kinds of electric violins. I eventually abandoned the strap-on flying-V instrument (though I still have it!) and went through several other designs, all solid-body. Finally, in the summer of 2002 I commissioned a 6-string acoustic/electric violin from Eric Aceto. This instrument has a full resonating body (but no F-holes), and rather than having pickups directly under the strings, the pickup is on the top of the violin and gets a spectrally rich sound, colored by the body of the instrument.
Also, having a full resonant body, the instrument can take the full weight of the bow arm; one of the ironic things I found with solid-body instruments is that, in spite of their ability to be very loud, you can't really sink the bow into themthere is no place for the energy to go, so the strings just give out. I also found that with the Aceto instrument, the need for applying equalization to the sound goes away; the resonating body takes care of that.
So, it is with this wonderful new Aceto electric violin and eight hemispherical speakers that I join the ACO on January 21. Traps Relaxed, for strings, percussion and electric violin/laptop, is a re-imagining of my earlier piece for string quartet and electric violin/laptop. With both Traps and Traps Relaxed, I am exploring a subtle algorithm where the sound of my violin is cut into tiny pieces (grains), transposed, and scattered across the speakers within the ensemble. The algorithm attempts to transpose my sound to a given "trap" pitch (or pitches). So, if the trap pitch is a high F, and I play an A below that, it will transpose my sound up a minor sixth:
(the little high F indicates the trap pitch)
Sometimes, however, it will also still remember notes I had played slightly earlier, so if I played a low D and then returned to the high A, we might also hear a bit of Bb (the D transposed up a minor sixth):
(note that in this sound example, I repeat the low D a couple times, so this notation is incomplete and shows only the first low D)
Or, I might have two trap notes, so we get a richer sonority:
And one of my favorite examples:
(note how when I play the low double stop we can hear a noisy spattering of grains; these are the algorithm's attempt to transpose that low double-stop up to the high trap A).
Both Traps and Traps Relaxed go through a series of "traps," exploring their various harmonic and textural possibilities. Traps Relaxed is a bit more expansive than the original Traps, lingering in places, discovering new possibilities that the larger ensemble allows.
The electric "violin" (which includes the Aceto, the speakers, and the laptop) that I will be using in Traps Relaxed is one of the main instruments that I am working with these days, and I am thrilled to have the opportunity to try with the ACO at Zankel; each new piece, ensemble and hall that I play these instruments in brings new revelations. I have never used them with such a large ensemble (17 instruments); if all goes well I'll try 16 hemispheres (or more) with a full orchestra!
Some of you are probably wondering about Problem #3 (the Foot-Pedal as Controller for Signal Processing Problem). It is beyond the scope of this article to describe what we've been doing to address this problem, but I will leave you with what I hope will be an intriguing image:
This is the Bowed-Sensor-Speaker-Array (BoSSA), an instrument I built that takes its inspiration from the violin. It consists of: a spherical speaker; a violin bow with sensors that detect bow position, speed, pressure; an ebony fingerboard with sensors that detect finger position; and a set of sensors arranged like a "bridge" that can be bowed. The data from these sensors are used to control synthesis and signal processing algorithms on a laptop, which in turn sends sound out the twelve elements of the spherical speaker. It sort of feels like a cello to play, with the speaker sitting in my lap. Sort of.... If you are intrigued, check out a video of BoSSA here and visit the BoSSA website.BoSSA is also the inspiration for PLOrk (the Princeton Laptop Orchestra), a new ensemble being formed at Princeton that combines a dozen hemispherical speakers with a variety of sensor/control devices; stay tuned!