By Christina Li
By Dave Pehling
By Ian S. Port
By SF Weekly
By Ian S. Port
By Ian S. Port
By Ian S. Port
By Ian S. Port
The graduate course "Music Controller Design and Interface" meets twice a week in a crumbling old ballroom at Stanford. The course's professor, Max Matthews, is in his late 70s and slightly stooped, with the patient smile of someone who's spent a life clarifying esoteric subjects. During a recent lecture, he explains to his students why the objective of his course -- nothing less than creating one's own musical instrument from scratch -- is so difficult.
"Instruments of any kind normally take 10 or more years to master," he says to his eight students, all of whom are roughly a third of his age. "So if you want to make yours capable of creating good music in the near future, you have two options. One: You design it so it can be played like an instrument that people already know. Or two: You develop really smart software that makes the hardware a lot easier to operate. When I created mine, I did both."
Matthews mentions his own invention, the Radio Baton, in an offhand manner, as if it were of little historical significance. But in computer-music circles the Baton was a landmark creation -- an instrument that could conduct an entire orchestra of computer-generated sounds -- and it was far from Matthews' only one. Over the course of his 50-year career, Matthews has been at the forefront of the digital sound revolution, enabling everything from crystal-clear cell phone calls to laptop techno. Max Matthews is the grandfather of computer music, without whom the experience of sound would be unfathomably different.
In 1957, while working at Bell Telephone Laboratories, Matthews invented digital audio, the process of turning sound waves into ones and zeros and then back again into sound. This ability to translate analog sound into a common medium increased music's portability and malleability a thousandfold. The prevailing playback and recording formats of today -- CD, MP3, and DAT -- wouldn't exist without digital audio, and neither would modern dance music, with its computer-driven drum machines, samplers, sequencers, and synthesizers. While vinyl-collecting audiophiles still revere analog methodology, the digital domain is by far the dominant means for producing and reproducing sound. "I would say maybe 90 percent of the stuff we listen to involves computer synthesis in one way or another," Matthews estimates.
With the amount of human energy that has been poured into computers in the last half-century, digital audio would have most likely happened eventually, with or without Matthews. But his ability to generate sound on a computer as early as he did, with extremely slow equipment and no precedent whatsoever, is remarkable. Such scientific milestones as the discovery of DNA and the splitting of the atom may have been flashier, but Matthews' feat has had an equal impact on our daily lives.
"What really impressed me when I heard him talk about what he was doing in those early years is that he had to make all these innovations, just one brilliant idea after another," says David Zicarelli, president of the influential San Francisco music software company Cycling '74 (the flagship product of which is called MAX, in tribute to Matthews). "All these things that are really the basis of all the techniques we use now. When someone asked where he got the ideas, he just said, "Well, it seemed obvious at the time.' But it wasn't obvious to anyone else."
At the time, Bell was the only U.S. telephone company, and its vast resources allowed it to pursue all sorts of arcane technological questions. The company's huge budget and computer access, which were rarities at the time, attracted some of the most brilliant technical minds of the era. And since mathematical know-how is often accompanied by an aptitude for musical expression, it's no surprise that many of the engineers tinkered with music as well. For his part, Matthews was a great lover of classical music and played the violin ("Although neither wisely nor well," he quips). Matthews' boss, also a music man, allowed Max to use the company's computers after hours for experiments, although the whole thing was unofficial because Bell was legally constrained from developing commercial products outside of its telephone business.
Matthews' earliest synthesized sounds were created as listening tests to evaluate the quality of phone lines. Grasping what this new capability meant in an artistic context, he developed a program called Music 1 that allowed him to play music on a computer. The initial piece produced with Music 1 was by all accounts a horrible-sounding 17-second composition titled "In the Silver Scale," crafted by another Bell employee. But Matthews and various colleagues improved upon the program over five subsequent versions, which were sent out to interested parties for free. This was in the days before the advent of keyboards, so the Music program was mailed as stacks of 3,500 punch cards with a note reading, "Good luck!" Only a select few professors with access to mainframes and a surplus of patience managed to get Music 1 up and running on their machines.
During Matthews' tenure at Bell, which lasted until the company's dissolution in 1987, computer-music research was largely a fringe pursuit of university music departments and the classical music world. The vast majority of composers either ignored or scoffed at the strides he and his colleagues made with computers. "Most of them and a good part of the rest of the world were absolutely terrified of computers," Matthews recalls. "If you remember the film 2001, where the computer HAL went mad and had to be turned off, thatwas the general feeling at the time." (Fittingly, during HAL's death throes, it hummed Matthews' computerized version of "A Bicycle Built for Two," which marked the first tune a computer sang.) "By contrast," he continues, "there was a small number of very farseeing, very famous musicians who saw the possibilities here, which was that computers can synthesize any sound that the human ear can hear."
Composers such as Milton Babbitt from Princeton, Vladimir Ussachevsky from Columbia, and Frenchman Jean-Claude Risset lent their input to the development of the Music program. By the end of the '60s, the foundational program had inspired various software architects outside of Bell to create offshoots of their own, which evolved over the years into the robust digital applications used today. Computer music had become more than the extracurricular activity of a few Bell engineers, and academic departments devoted to it sprang up at universities like Stanford and MIT.
"Max was absolutely formative in the early years of digital audio," observes Joel Chadabe, computer-music historian and author of Electric Sound: The Past and Promise of Electronic Music. "But as everyone else jumped on the bandwagon, he ... began to be interested in other kinds of performance devices, because at the heart of it, he is a musician. Looking retrospectively, it seems very logical that the first thing he would do is develop sound and then develop something to play it with."
Along the lines of creating new ways for musicians to interact with computers, Matthews developed Generated Real-time Operations On Voltage-controlled Equipment (GROOVE) in 1970. Until then, there was no way to perform computer music in real time because early digital synthesizers took hours to generate sounds. Analog synths, on the other hand, could produce music instantly, but they were not programmable. GROOVE was a computer that sat between a performer and an analog synth and allowed the latter to be programmed in something approaching real time. GROOVE was the ancestor of the modern sequencer, which allows electronic musicians to arrange sounds into a song structure.
"I could play [GROOVE] with a piano keyboard, joystick, and a telegraph key with which one conductor could beat time," Matthews describes. "That was the first virtual orchestra."
But GROOVE suffered from the same problem Matthews spoke of in his class: It wasn't played like a familiar instrument. In 1987, after years of refining the idea, Matthews and Bell Labs colleague Robert Boie developed a tool that was more user-friendly: the Radio Baton. To use the instrument, a conductor or composer entered a score into the accompanying Conductor Program software and then conducted it by waving a pair of batons that had low-frequency radio transmitters at the ends. The transmitters relayed the motions of the wands back to the computer, which altered various characteristics of the score. Today, soloists and singers who can't afford to hire an orchestra often use the Radio Baton in performance, as do modern composers who want to re-create complex music live. Some music-technology experts consider innovations like the Radio Baton to represent a trend toward easier interaction with sound for nonmusicians, which may one day have far-reaching commercial applications.
As Matthews puts it, "I have envisioned a new way of appreciating music called "active listening.' Instead of buying a recording and sitting in a chair listening to somebody's interpretation of Beethoven's Fifth Symphony, you'll buy the score for the Conductor Program and you will conduct your own interpretation -- at the tempo you want, as loud as you want, with the dynamics you want."
Nowadays Matthews splits his time between selling Radio Batons out of his San Francisco garage and teaching at Stanford. He says the Baton business is far from profitable, but he expects that some of his students will be more successful in the future. Their gadgets -- which have used lasers and accordionlike mechanisms and sported such tantalizing names as the Metasaxophone, Holey and Phony Controllers, and Crystal Mountain Environmental Sensor -- might one day play some unimaginable music or allow new ways of interfacing with sound.
As music technology continues to increase in speed and power, Matthews hopes the field he engendered will stay true to its original course -- toward human empowerment, not Stanley Kubrick's vision of alienation. "I think we are and should be in control of computers in terms of their job descriptions," he says, "and we should only ask computers to do things that we do not want to do ourselves."