Using the same to make different. Analysis of a traditional musical repertoire based on centonisation

This paper deals with music based on Centonisation, a compositional technique broadly used during the Middle Age, still at work in many musical traditions around the world. Based on a limited stock of melodic patterns, it consists of using them in as much as many different ways as possible. The multiple possible combinations of ABCD musical patterns (ABCD, ABDC, etc.) can create as much as possible different musical pieces. As a result, understanding a musical corpus built on centonisation is quite a challenge for the listener: most of pieces sound the same although they are always different. In other words, the centonisation technique does not simply deals with the same and the other in music: it only works on these principles. As a very simple theoretical music technique, it possesses a very high potential for complex development in practice. In that sense, such a technique could be of great value as a basis for computational analysis that deals with automated pattern identification and extraction. Similarly, centonisation as an implicit/explicit system seems well adapted for experimental studies that analyze the cognitive processes involved in listening, such as identification, memorization and recognition of mobile melodic patterns in a closed system that is purely experimental or effective, like in the presented Ethiopian case.