The Selection of Clausula Sources for Thirteenth-Century Motets: Some Practical Considerations and Aesthetic Implications

Abstract

In addressing questions of compositional process, scholars of medieval polyphony have relatively little on which to build. They cannot gain insight from reading explicit written testimonies by medieval composers; nor can they look to evidence such as sketches, drafts, or revisions for guidance. Instead, they can only study theoretical accounts of how to compose good discant and examine the surviving pieces themselves to increase their understanding of medieval compositional process. l Of all the surviving types of medieval music, the early Latin offers perhaps the best opportunity to explore aspects of compositional process, for the vast majority of early Latin motets were created in several observable stages. More than one hundred years have passed since Wilhelm Meyer's pathbreaking report that many early motets originated with the the addition of text to preexisting discant clausulae (Meyer 1898). While Meyer'S discovery prompted an intense effort to identifY related motets and clausulae (Ludwig 1910; Gennrich 1957; van der Werf 1989), other elements of compositional process were largely overlooked. Norman Smith (1989) recognized this lacuna and drew attention to the process of converting clausulae into motets; in this paper, I shall focus on an earlier stage of compositional process-the process of selecting clausula models to convert into motets. In brief, I shall identifY factors that may have attracted medieval to select certain types of clausula models for early Latin motets.2 I shall then consider the extent to which their criteria for selection apply to other segments of the thirteenth-century repertory. As I will show, the initial selection criteria not only reflected practical considerations, but also had long-range aesthetic implications for the stylistic development of the medieval motet. The surviving sources of Notre Dame polyphony indicate that the composers of the earliest motets did not indiscriminately add texts to all of the approximately 950 passages of discant; instead, they focused their efforts on less than twelve percent of the available repertory.3 To investigate potential criteria for selection, I compared a motet group of fifty-five clausulae that were converted into motets with a control group of 103 clausulae that did not become motets.4 The purpose of this comparison was to look for possible stylistic differences between the two groups; if