Indeterminate Tudor

For the composers in both the New York and Darmstadt schools in the 1950s and 1960s, David Tudor was the interpreter for all new music written for the piano, and we undoubtedly owe Tudor credit for the large amount of piano music from the hands of Cage, Feldman, Boulez, Stockhausen, and their contemporaries. In later years, Tudor turned his attention more to electronics, but he remained committed to Cage's music throughout his life. His last piano recording, made in 1992, was a recording of Cage's Concert for Piano, a work he had recorded several times previously, starting almost immediately after the work was written in 1958.

Because of Tudor's stature in Cage interpretation, I looked forward with some anticipation to a recent two-CD set on Editions RZ of Tudor piano recordings from the 1950s and 1960s culled from the archives of WDR and other radio archives. The Swiss label, Hat Hut, has released a number of David Tudor CDs from the same radio archive, but many of them are now out of print. The new Editions RZ set includes five recordings from Cage's Music for Piano series; three recordings of Christian Wolff's Duo for Pianists I; one recording each of Cage's Winter Music, a realization of a graphic score of Sylvano Bussotti, and Morton Feldman's Piece for Four Pianos; and a reissue of Tudor's recording of Variations II on amplified piano, originally released on the landmark Columbia Masterworks recording of New Electronic Music. All of the individual works recorded here require some degree of realization (in some cases, a high degree) because of the open nature of the scores.

Cage's Music for Piano series consists of 84 unnumbered sheets of music, published in eight different sets (1, 2, 3, 4-19, 20, 21-52, 53-68, and 69-84). Cage generated all the notes from imperfections in the manuscript paper in conjunction with I Ching operations, and the works were to some extent exercises in composing more quickly after the long composition time for Music of Changes, his first major piano work composed with chance operations. Cage eventually found that one of the drawbacks of these pieces was that the music paper imperfections produced only single notes. Here is an example, one system (of four) from one of the sheets. Even though he specified that some notes were to be muted or played pizzicato, single notes still makes for a very sparse texture. To circumvent this problem, Cage allowed for different pieces to be performed simultaneously, by one or multiple pianists. Presumably Tudor's realizations on this set are illustrations of the combinatorial possibilities, but they're all still very pointillistic. Winter Music applies similar techniques, but combines the points into chords, and includes some additional rules about how to disambiguate mapping notes in the chords to different clefs. Compared to the Music for Piano pieces, Winter Music (played here by two pianists, Tudor and Cage) sounds positively lush.

As a bizarre footnote, one of the Music for Piano pieces was previously released on one of the Hat releases with a different title (Music for Piano 21, 22, 26, 29, 34, 36 here is the same recording as Music for Piano 52-56 on the Hat album). It makes me wonder: if the compilers can't decide which pieces are actually being played, what's the difference between these pieces and random notes?

Cage and Tudor must have had a relatively fixed program when they made the rounds of German radio stations in 1960. One of the Hat releases contained two versions of Christian Wolff's Duo for Pianists I, recorded in 1960 for WDR. This album contains two different recordings of the same two realizations, also recorded in 1960 but for Radio Bremen. (Jazz buffs get upset when archival releases don't contain complete discographical information. One might have the same complaint here.) Wolff's piece uses time brackets and fixed pitch classes from which to choose, but leaves many details up to the performer. In this example from For Pianist that uses similar notation, Wolff specifies one sound from pitch class C for 1 and 1/5 seconds, followed by no sounds for 3/5 seconds, followed by three sounds from pitch class C, one of which is played pizzicato, and one of which is lowered by a semi-tone. The whole system should take 2 and 1/10 seconds. Wolff wrote Duo for Pianists I for his own concerts with Frederic Rzewski as an experiment in controlled improvisation. Tudor created fairly strict realizations of the various indeterminate pieces that he played (despite occasional heated objections from the composer), but it's interesting to hear the proof with the comparison of these two recordings. This release also includes a third recording of the same piece, done three years later. By 1963, Tudor was feeling constrained by notated pieces. As he states in an interview fragment included at the end of the second disc, he only felt alive "in every part of my consciousness" when he was called upon to make actions that were "undetermined as to what [sounds] they are going to produce," so it is likely that the later recording was performed more in the spirit of the original composition.

One of the reasons I was excited about this release was that I recently started making a realization of Winter Music. Without going into too many details of the mechanics, I endeavored to understand Cage's egoless musical activity through study of one of his most famous indeterminate pieces. But I find the nearly random nature of the piece almost too daunting. I have a few minutes of music prepared, which I currently play very softly, as if it were a Feldman piece (Tudor peppers his realizations with occasional very loud percussive attacks). But so far I remain unconvinced. I am more comfortable performing music that has patterns that develop or repeat. Cage's indeterminate music deliberately eschews any kind of pattern, so the performer must supply a musical thread that simply doesn't exist in the printed score. It presents a different set of challenges than more fully notated music, a half-way house between composition and interpretation. Cage's systems may have made it quicker for him to compose, but the resulting works put an unusual burden on the performer.

But the creation of a realization of Winter Music isn't really an egoless activity. One of Cage's primary goals once he started composing with chance operations was to make compositional decisions without his ego getting in the way. But compositional decisions are different from performance decisions. When it comes to a performance of these pieces, once the performer has made the requisite decisions to transform the score into an audible piece of music, the performance characteristics aren't that different than for any other fully notated work. Wolff's pieces at least provide a framework for a performer to create a different work at each performance (even if Cage and Tudor opted to fix the performance characteristics for their 1960 radio tour). In this respect, Duo for Pianists I is more like the piano compositions of Henri Pousseur. As Tudor realized in the 1963 interview, the indeterminate nature of the piece should translate to its performance as well, and not simply to the ability to prepare any number of static realizations.

David Tudor's Music for Piano on Editions RZ is available at Erstwhile Records and other fine record stores.