Features & Columns

'The Sound of Innovation: Stanford and the Computer Music Revolution'

KEYBOARD COWBOYS: Stanford's CCRMA Laptop Orchestra reinvents classical music.

In 1934, as the United States was deep in the Great Depression, rural Salem, New Jersey welcomed its newest resident into the world: John Chowning. Chowning held an early interest in music, playing violin from the age of seven and percussion instruments from the age of 12. After high school, he served a three-year tour as a musician in the Navy.

Back in the United States, Chowning attended Wittenberg University in Ohio, graduating with a Bachelor of Music. He then moved to Paris to study with Nadia Boulanger—the French composer, conductor and teacher who counted the late American composer Aaron Copland among her pupils.

Post World War II Paris was the epicenter for the musique concrete movement. Pierre Schaeffer, an electronic engineer, employed rudimentary recording equipment, disc cutters, to isolate and capture naturally produced sound events. Schaeffer then experimented with how to manipulate and isolate portions of the sounds. Meanwhile, Cologne, Germany served as the epicenter of elektronische Musik, which arose around the same time. Elektronische Musik emphasized entirely synthetic means of producing sounds, drawing upon noise generators, filters and other devices to provide and manipulate sounds.

Despite their different roots, these two movements rubbed up against one another, and they increasingly moved away from their dogmatism in the late 1950s and 1960s. As Chowning arrived in Paris in 1959, the electronic music scene was one in which composers actively sought new direction and new inspiration for musical composition, drawing upon emerging electronic tools to further their musical visions.

Chowning's time in Paris solidified his interest in pursuing a DMA in music composition. He decided to apply to schools on the West Coast of the United States.

Experimentation

At Stanford, Chowning immediately became involved in the Society for the Performance of Contemporary Music—a small group of students who put on concerts. As Chowning describes Stanford, the music department embraced a wide range of musical styles, performing "traditional" classical music and also experimenting with a variety of new music.

Stanford's relative lack of barriers between departments, schools and programs was crucial in enabling Chowning's experimentation. As Chowning noted in a 2008 interview, not every university would have allowed a graduate student in music to access personnel and equipment from an artificial intelligence project: "That's, for me, the wonder ... that I was ever allowed to use a computer somewhere, even at night." Stanford in fact may have been somewhat unique in that respect. "Not every university [would have allowed it]. Loren Rush [who, with Chowning and others, would establish CCRMA] tried to do that [access a computer for musical purposes] at Cal not long after I started. They wouldn't let him. He couldn't break through. Sometimes the departmental/school boundaries can't be bridged."

Stanford, however, supported an environment that emphasized novel research, curiosity and exploration over departmental and disciplinary dogmatism.

Stanford's music program had grown organically, and it had only been a department since 1947. The department saw an opportunity to experiment with novel programs, since it had not yet established a reputation or devoted significant resources to any particular emphasis. In its early years, the department featured jazz studies, early dance and performance practice—none of which were standard offerings at comparable universities. Late founding and faculty autonomy combined to support experimentation and novelty.

Mods vs Freqs

Chowning's colleagues at the Stanford Symphony Orchestra were well aware of his growing interest in electronic music. In January 1964, one of the members, Joan Mansour, passed him a copy of an article from Science that described how a computer could be used as a musical instrument. The article's author, Max Mathews, based on work he had conducted at Bell Laboratories in New Jersey, explained the process by which numbers, the language of the computer, can be converted to sound, the basis of music. Fundamentally, computer music uses the numbers as "samples" of a sound pressure wave, with each sample corresponding to the state of the wave at a particular instant.

Chowning reasoned that Mathews' insights provided a way to pursue electronic music at Stanford. As he recalled in 1987, "Computers are general-purpose devices, and the idea of using them for music was very attractive because the computers were already there." An electronic music studio required a great deal of single-purpose equipment, which Stanford would not support; a computer, however, was already in place and justified on the basis of nonmusical applications. In many ways, in fact, the roots of the Stanford computer music program would grow from repurposing nonmusical entities—equipment, programs, people and funding agencies—in the services of musical aims.

In the early 1960s, computers were not common on the Stanford campus. Chowning's search for a machine quickly led him to the facilities of the Stanford Artificial Intelligence Project (Lab) or SAIL.

"The SAIL computer system, as Chowning recalled in his characteristic dry humor, "comprised an IBM 7090 that had an enormous memory of 36k 36-bit words and a hard disk whose capacity was well over 500k words and about the size of a large refrigerator."

In 1967, Chowning had his own technical breakthrough, which reflected the fruits of the AI lab environment: some late-night "fooling around," as Chowning described it, resulted in the discovery of frequency modulation (FM) synthesis.

"I was experimenting with just a sinusoid and kept increasing the vibrato rate, so all of a sudden it didn't sound like listening to a change in pitch in time, but rather I began to hear timbral differences. So the vibrato became very, very fast, hundreds of times per second, and very, very deep, as if the violinist had a different finger board, and the finger was whipping up and down at very high rates and very great distances."

In technical terms, FM synthesis involves varying the frequency of a carrier wave. These basics of FM were well understood as applied to radio transmission. Engineers, however, had not yet applied the technique in meaningful ways to the range that humans can hear.

Chowning then, in his own words, "did a number of experiments and got percussive tones, brass-like tones and woodwind-like tones. The FM discovery, as Chowning would characterize it, was "an ear discovery." In a 2006 radio interview, he argued, "I think without my musical interest and musical training, I would not have stumbled upon this."

Computer music afforded, at least theoretically, an unlimited sound palette with unlimited control and detail; whereas traditional instruments are objects with fixed sonic properties, the computer can create any sound you can imagine. Computer music compositions, therefore, opened a never-before-heard realm of musical expression.

Breakthrough

The CCRMA-Yamaha relationship serves, for many observes, as a model of university technology transfer: a university-based research group and a commercial firm collaborating over many years to develop a technical breakthrough into a widespread product. Undoubtedly, the relationship has yielded enormous benefits for each organization and for computer music as a whole.

No musical instrument manufacturer in the United States had been interested in Chowning's FM technique. Chowning (and a partner at Stanford) reached out to Yamaha in 1974. Though its U.S. market share was limited, Yamaha was a major manufacturer on a worldwide scale.

In December 1974, Yamaha's Yasunori Mochida responded to a letter from Chowning, sharing technical details of a digital synthesis system that Yamaha was developing.

Yamaha's interest was not indicative of a "sure thing." The FM sales pitch, however, was helped immensely by the three features: Yamaha was already working in this area; they had come up against a technical barrier that Chowning's system solved; and they happened to have an engineer—not a business development or marketing person—evaluate the opportunity.

Chowning's FM technique reduced the computational requirements and, in turn, the bulk and cost of associated hardware. On March 19, 1975, Yamaha finalized its license for FM, with the clause that Chowning would be highly involved in its development. This license would later become one of the most profitable in Stanford history.

(Yamaha's) DX7 synthesizer, which would become the best-selling synthesizer in history and which, 30 years later, retains the No. 2 spot in that category. At the DX7's heart: Chowning's FM synthesis. Stanford—and CCRMA—would receive money from each unit sold.

College Keyboard Scores

In 1983, Yamaha released the DX7 synthesizer. It was an instant blockbuster. Chowning recalled that he first heard the instrument at a bar in Palo Alto.

"My wife and I had been out to see a movie and we stopped off at a local bar for a nightcap. I knew the keyboard player in the bar, and when he saw me he waved me over excitedly to come and see this "incredible new instrument" he had sitting on top of his piano. I was astonished. It was an awesome moment. I had no idea that people had been waiting in line to buy DX7s. I had no idea at all."

The DX7 would be heard around the world. As Max Mathews recalled in a 2008 interview: "The DX7 brought the entry level [for digital synthesis] down to $2,000, and that really expanded the field and got a lot more people and a lot more schools involved. Now universities all over could get into computer music or electronic music programs."

The DX7, however, would influence not only academic programs, but also popular music. Indeed, the DX7 quickly became the best-selling musical instrument in history, completely redefining both the synthesizer market and popular music. A generation of musicians—including Madonna, Phil Collins and Toto—relied on its shimmering digital sounds to define a new sound in music. Thus, the DX7, more so than any other instrument, marked the transposition of digital sound synthesis from the realm of academic computer music centers to popular use by individual artists.

The large number of adopters, however, also drove economies of scale around digital synthesis. Suddenly, it was affordable. This affordability enabled the broad expansion of electronic music programs and of electronic music generally. As hundreds of thousands of musicians bought DX7s and other Yamaha products, these revenues enabled CCRMA itself—once on the brink of bankruptcy—to be financially viable. In short, commercialization success tied to a user-driven invention laid the foundation for an academic center, for the expansion of a new academic discipline and for the emergence of a new brand of popular culture.

Technology commercialization continues to play a central role at CCRMA. The shape of commercialization has changed in recent years, however, as intellectual property-focused direct commercialization efforts in the 1990s gave way to free and open-source-oriented efforts in the 2000s. In turn, CCRMA has found that the most fruitful commercialization approaches may lie not in "formal" technology transfer, but rather in informal and free sharing that builds long-term personal relationships.

This relationship between university-based research, education and firm- based technology development has continued to play out in other recent CCRMA activities. For example, Ge Wang, director of the Stanford Mobile Phone Orchestra (MoPhO) served for several years both as assistant professor at CCRMA and as chief technology officer and cofounder at Smule, a maker of musical instrument applications for smartphones.

Wang's motivation is notable: free and open sharing enables both diffusion and subsequent improvements by others. The fact that he took a cue about free and open sharing from his PhD advisor at Princeton, Perry Cook, whose own perspective was shaped by his experience at CCRMA, reinforces how both people and practices in computer music transcend organizational boundaries.

As with Cook, too, Wang's engagement in free and open sharing is not to suggest that he was blind to the commercial potential of his creations. Quite the contrary, in fact: Wang's audio programming language, ChucK, is the same one used by Smule, Wang's mobile-phone music app start-up. Yet the path from ChucK to Smule reveals a far more complex picture than that of an academic seeking financial profit from his research work: at Princeton, Cook and Dan Trueman (an associate professor of music) pitched Wang on the idea of a "laptop orchestra" that would use ChucK. Wang loved the idea. In 2005, he helped start the Princeton Laptop Orchestra, or PLOrk.

When Stanford hired Wang as a professor in 2007, he transposed the laptop orchestra idea to CCRMA, founding the Stanford Laptop Orchestra, or SLOrk. The group cut holes in IKEA salad bowls and inserted car stereo amplifiers and speakers. The result is a custom half-spherical speaker that resembles the head of R2D2. Each SLOrk performer has one hooked up to his or her laptop. Today, digital music is widespread. Our phones function as musical instruments and media players. Electronic keyboards fill shelves at Target and Walmart.

As we reflect on the surge of digital music into everyday life, CCRMA deserves much of the credit. The center's faculty, staff and students invented and commercialized key technologies. They populate academic programs and commercial firms around the world. Perhaps most important, their success with crossing disciplinary boundaries, engaging in open innovation and commercializing their research added legitimacy to these activities, facilitating the emergence of other computer music groups and encouraging other explorations at the boundaries of technical and artistic, commercial and academic.