Use this URL to cite or link to this record in EThOS:
Title: Algorithmic categorisation in formal music analysis
Author: Anagnostopoulou, Christina
Awarding Body: University of Edinburgh
Current Institution: University of Edinburgh
Date of Award: 2001
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
A new method of formal music analysis is presented, Similarity and Categorisation Analysis of Music, which is based on Paradigmatic Analysis of music, as devised by Nattiez. The new method, which is independent of musical style, is applied to three pieces of diverse character: a Scottish folk tune, a solo flute piece by Debussy and a piece from a piano sonata by Boulez. As in Paradigmatic Analysis, the analyst produces a classification of the segments of a piece of music. However, the new method is set in a more formal framework in which the individual steps of the analysis are clearly delineated and the criteria for the classification are explicitly defined. Furthermore, in extension to Paradigmatic Analysis, the resulting classification is hierarchical, and new findings from categorisation theory are brought in to enhance the methodology and to act as a bridge to cognitive modelling of categorisation in music analysis. The new method is computationally modelled with an unsupervised neural network algorithm, thereby further formalising the classification process. The rationale behind this new approach is to allow for a formal analysis without restricting the analyst’s freedom of choice: the method acts as a framework for the analysis, making explicit previously intuitive decisions of the analyst, while the analyst remains free to choose his/her own analytical criteria. The thesis is divided into two parts: the first part provides necessary background and describes the new method in detail in musical terms. An example analysis is demonstrated for a Scottish folk tune. The second part describes the computational model of the method. The purpose of the computational analyses is two-fold: first, to produce entirely new analyses, and second, to reproduce human analyses in order to find the criteria on which the analyst based his/her decisions. Two more pieces of different character are analysed using the computational model: Syrinx by Debussy, which has previously been analysed by Nattiez in the framework of Paradigmatic Analyses, and Parenthèse by Boulez. The thesis concludes with an evaluation of the method and a discussion of the extent to which it is possible and useful to be formal in music analysis.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available