Subscribe to our mailing list to receive updates about forthcoming issues. JCMS will not disclose your details to any third party.
Volume 2, Issue 1, September 2017
Computation, Style, Innovation, Clustering, Information Theory
Using corpora of 19 common-practice tonal composers, this article uses a variety of similarity metrics including cluster analyses and cross entropy to identify points of particular stylistic uniqueness. Throughout four computational experiments, we find that geography and chronology determine much inter-corpus similarity; however, points of dissimilarity – in particular, dissimilarities between larger groupings of similar composers – are described. Specific properties of stylistic imitations are also investigated and found to be more normative than average. These data are interpreted to arise due to emergent phenomena rather than being driven by individual “creative” composers.