Is it Beethoven, or is it software

February 28, 2000|By Bob Tedeschi | Bob Tedeschi,NEW YORK TIMES NEWS SERVICE

Anyone who plays Beethoven, Mozart or Bach faces issues that have plagued performers throughout the centuries: namely, the composer's score provides scant guidance on how to approach many facets of the music. Performers make educated guesses as to how loudly to play individual notes, for instance, or how to accent four consecutive quarter notes, but they remain guesses.

Now Dr. Manfred Clynes, a neuroscientist and concert pianist of Sonoma, Calif., has discovered what he says are clues as to how exactly the masters would have wanted their pieces performed.

Clynes, a former neuropsychology professor at the University of Melbourne, who coined the term "cyborg" and has performed to the acclaim of people like Yehudi Menuhin and Einstein, has identified what he says are patterns in the music of master composers.

Those patterns, or "pulses," show how a composer favors louder and longer notes at consistent points in a phrase, much as poets resort to patterns of phrasing or painters favor certain brush strokes. Clynes argues that similar patterns exist on higher levels in musical compositions, like the way composers stress certain bars in a movement.

Performances of Beethoven, for instance, carry his characteristic strength and intensity when, in four-note sequences, the second note is played shorter and softer than the others, whereas the notes in Mozart's sequences are more consistent in length and amplitude.

These discoveries were made in a series of experiments that built on the work of Gustav Becking, a German musicologist who first put forth the idea of composers' pulses in 1928. Becking himself had extended the theories of Eduard Sievers, who found he could mark the shift in the Bible's authors by following the rhythm of the writing with the movement of a finger. The loops that Sievers traced while reading each writer were distinct, he found.

Just as each writer evoked consistent finger patterns, Becking found, composers produced similar patterns. Clynes first measured those shapes in 1967 by recording how musicians like Pablo Casals reacted to certain pieces of music while using a device Clynes invented, called a sentograph.

A person using the sentograph listens to music and touches a pad, which finely records finger pressure and translates that pressure into shapes. When several such musicians reacted to the same composers with similar shapes, Clynes said, "we knew we had something good."

In the early 1980s, using digitized performances of pieces by well-known Western composers, Clynes experimented with variations in each composer's pulseand tested those pieces to see how true they were to the patterns he had established for each. In this way, he was able to chart the "micro-structure" of a composition -- the most subtle aspects of a score, like the attack on an individual note, which composers leave to performers and conductors to discern.

Clynes codified these findings in a computer program that allows users to experiment with the pulses of various composers -- say, what a Mozart symphony might sound like with Schumann's pulses -- and build interpretations of their own, using digitized music rendered by computer. That program, Superconductor, is now used at the Juilliard School and has won praise from musicians and from scholars at the Massachusetts Institute of Technology's Media Lab.

In so doing, Clynes has provided new tools for composers and musicians.

"Manfred has shown that a computer simulation can be made good enough for many people to immediately take to it, and that's not a trivial feat," said Dr. Jerry Lettvin, professor of neuroscience at MIT. "In fact, the difficulty of that is quite staggering."

Previous incarnations of digitized music "were a joke," said Alice Shields, a composer who is the former associate director of the Columbia-Princeton Electronic Music Center. "They were wooden and boring and had none of the emotional qualities that noncomputerized music has," she said.

Part of the reason, Shields and others said, is that computerized music often plays every note virtually the same, with few subtle variations in aspects like tone or tempo found in performances. In his computer program and in 240 computer-generated recordings of works like Bach's Brandenburg Concertos, Clynes uses the pulse to avoid such an effect, but he also relies heavily on another of his theories: predictive amplitude shaping, which calculates the volume of notes in a phrase.

For many instruments, such as a violin, individual notes can be shaped differently, depending on how the amplitude rises and falls during a note. A note can have a gradual build-up or can diminish quickly.

Clynes observed that the shape of a note's rise and fall is affected by the next note. If the next note is a higher pitch, then the amplitude will fall more sharply at the end. If the note that follows is lower, the amplitude will fall more gradually.

Baltimore Sun Articles
Please note the green-lined linked article text has been applied commercially without any involvement from our newsroom editors, reporters or any other editorial staff.