Beyond Bars: Distribution of Differences in Music Prints

Abstract

Our goal is to analyse differences between prints of Beethoven’s Piano Sonatas. Since there is a huge number of prints, we will only compare sample encodings instead of full encodings. The aim of the presented work is to evaluate three different algorithms to draw these samples and find the one which draws the samples that best represent the prints. To that end we used six editions of Beethoven’s Bagatelles Op. 33 as a test group and drew 1000 samples from each Bagatelle for each algorithm. Then, encodings of these samples were compared, resulting in 45.000 comparisons for each Bagatelle. To visualize the results, they are being plotted, so that the number of differences can be seen on the x-axis and the frequency, with which this number of differences occurs, on the y-axis. As a result we obtain a normal distribution for each of the algorithms but with different scale parameters. We interpret this finding to demonstrate that thinking of music in terms of bars is not sufficient to find a sample that represents the basic population. Instead, it is necessary to take into account the density of the musical events in the scores.

Fabian C. Moss

Digital Music Philology and Music Theory

Fabian C. Moss is an assistant professor for Digital Music Philology and Music Theory at Julius-Maximilians University Würzburg (JMU), Germany.