PÄRT’S EVOLVING TINTINNABULI STYLE1 MICHAEL CHIKINDA CURIOUS PARALLEL is found in the compositional development of Arvo Pärt and Arnold Schoenberg; both composers experienced a creative hiatus followed by a period in which they fervently explored a new compositional technique. For Schoenberg, the pan-tonal works of the pre-war years were followed by the pre-determined order of the twelve-tone technique. For Pärt, the twelve-tone technique served as a point of departure before his hiatus, which was followed by the beginning of the tintinnabuli compositions. While Pärt is certainly not the first composer to have abandoned the twelve-tone technique in favor of a new form of expression, his reason for doing so is unique. For instance, Steve Reich had reservations about a compositional process that was not immediately perceptible to the listener.2 Could the row be heard by even the most attentive listener? Thus, his response in his early compositions was the use of the phasing technique, which he felt A Pärt’s Evolving Tintinnabuli Style 183 would be immediately perceptible to his audience.3 By contrast, Pärt’s concern was not about transparency; rather, his departure from the twelve-tone technique in favor of the tintinnabuli style was to find a method that more fully expressed spiritual considerations.4 A brief review of the tintinnabuli style will prove helpful before discussing these considerations and their evolution in later compositions. There are two elements that work in tandem in the tintinnabuli style: the M-voice, which represents a melodic line (typically using conjunct motion) and the T-voice, which represents the note of a triad (typically a major or minor triad).5 Traditionally, the T-voice has been understood to stand in relation to the M-voice; that is to say, as the scalar pattern of the M-voice unfolds in the music, the presence of the T-voice is gauged by whether it is articulated above or below the Mvoice and by the intervallic distance between the two. For example, if the M-voice articulates the following scalar pattern: C4 – D4 – E4 – F4 – G4 and the T-voice articulates, in a first-species counterpoint style, E4 – E4 – G4 – G4 – C5, it would be considered a 1st-postion, superior tintinnabuli voice: superior, because it is always placed above the M-voice, and 1st-position, because each member of the C major triad that it articulates is in ‘closed’ position. By contrast, had the Tvoice articulated instead G4 – G4 – C5 – C5 – E5, then it would be considered a 2nd -position, superior tintinnabuli voice because there is another member of the C major triad that could be inserted between the M-voice and the T-voice.6 Using this schema, it is also possible to construct a 1st- and 2nd -position inferior T-voice and an alternating pattern that switches between superior and inferior positions.7 Hillier privileges, in both his book and in his Grove Music Online article, the M-voice by suggesting that the T-voice is to be understood as something that is placed in reference to the M-voice, which assumes a compositional process whereby one starts with a cantus firmus and then writes a T-voice counterpoint; however, considering the relationship of these two elements in a different way, one could privilege the T-voice by suggesting that the conjunct motion of the M-voice is the result of ‘filling in the spaces’ between the members of the T-voice. I will return to this idea later. It will be helpful now to turn to an example from an early work from Pärt’s oeuvre that demonstrates the schema just outlined. Example 1 shows a brief passage from his Cantus in Memory of Benamin Britten (1977), which isolates the divisi violin II part and the campana for the purpose of clarity.8 The campana, throughout the entirety of this work, plays A4, which acts as a center of gravity. The upper part of the divisi violin II is an M-voice that articulates a scalar pattern over a 184 Perspectives of New Music two-octave span (in this context the A aeolian scale), while the...