Die Digitalisierung des goldenen Zeitalters – stilometrische Autorschaftsattribution am Beispiel des Quijote

For quantitative literary analysis, the Spanish Golden Age seems to be a perfect field of research. Many texts are provided by Virtual Libraries like the most known BVMC (Biblioteca Virtual Miguel de Cervantes) in various editions. Though for textual analysis like stylometry not only the text itself, but also the metadata should be complete and correct. As the contemporary editions of the Golden Age are themselves difficult to judge regarding the textual originality, additional information (usually called metadata) about the origin of text, author, date, place is needed. The following paper discusses the problem of existing textual basis for stylometric analysis using the example of authorship attribution for the apocryphal second volume of the Quijote by Avellaneda (pseudonym). The aim is not to discover the real identity of the author Avellaneda, but to test the method with various parameters (cosine Delta, selected wordlist, culling, rolling Delta) and to discuss the results in the context of other existing authorship attributional methods and in the context of traditional working literary history and studies.