HPSCHD (review)

The much anticipated—now heralded—studio-realized compact disc recording of John Cage/Lejaren Hiller's HPSCHD (1967–1969), for harpsichord(s) and computer-generated tapes, was released in December 2003 by EMF Media. It is the first such studio version for a recording medium since the 21-minute LP recording by Cage and Hiller for Nonesuch Records in 1969. On 16 May 1969, the five-hour premiere performance was presented for seven amplified harpsichords, 51 monaural tapes, and thousands of NASA (and other) slide projections, all experienced by the roaming audience of 8,000 in the Assembly Hall at the University of Illinois, Urbana: a maximal, musico-theatrical, "sixties" event. This new recording is, according to liner notes by co-producer Joel Chadabe, not an attempt to "capture the dynamics of a performance [of HPSCHD]. [Co-producer] Bill Blakeney and I believed that the sound of HPSCHD, which cannot be preserved through any conceivable musical notation, must be documented." And the new release is a compelling document, musically, graphically, and historically.

Like the first studio-realized recording of HPSCHD for LP, this new production presents a composite recording of harpsichord and computer-generated sounds. In the first recording, three harpsichordists (Antoinette Vischer, Neely Bruce, and David Tudor) each performed/recorded one of the seven different Solos in the piece. Each Solo was around 20 minutes long, the nominal duration of an LP side. The performers used different instruments, leading one to conclude that they made the recording in-studio, all three playing at the same time, rather than individually recording their different Solos on the same instrument on separate tracks of a multi-track recorder. In the new recording, all seven Solos are performed/recorded by a single harpsichordist, Robert Conant, on a single harpsichord on seven separate tracks, these combined, in post-production, in diverse ways with one another and the 51 computer-generated tapes for a total duration of 65:27, over three times the length of the LP version and closer in spirit to the extended temporal scope of the original and the subsequent live performances through the years. Both versions work well and have the same astonishing density and intensity of this amazing, massive collage of unrelenting harpsichord musics and computer sounds. In the new version, the quality of the recorded computer sounds is cleaner and sharper, due to the restoration process of noise reduction and the elimination of clicks and glitches from the tapes, all accomplished by William Blakeney. Because Mr. Conant recorded all seven Solos with the same instrument, there is homogeneity of timbre . . . although that is certainly not the case in live performance or on the LP version, where a diversity of the harpsichords' sound quality is heard.

It is interesting here to note that in a 19–20 June 1968 interview I conducted about HPSCHD with Cage and Hiller in Urbana, Illinois, Cage explained:

It's called Harpsichord—HPSCHD—because of a commission [in 1962] from Swiss harpsichordist, Antoinette Vischer, to write a piece for harpsichord. I must admit I've never particularly liked the instrument. It sounded to me like a sewing machine.

(Larry Austin, Source, Issue 4, Vol. 2, No. 2, 1968, pp. 10–19)

But the model for the computer sounds developed by Hiller was, indeed, the harpsichord sound, with its complex sidebands and rapid, plucked-string attack and two-ramp decay. The resulting sounds were, according to Hiller, mostly sawtooth waves. To my ears, these synthetic, rather awkward approximations of the harpsichord's sound and amplitude envelope are sometimes harsh and distorted. But they are part of the sonic, hybrid mix of the piece. They belong there, together with the live harpsichords—an anomaly—a paradoxical mix of the old—softly delicate sounds of the harpsichords, amplified—combined with...