The “Diktat für Schlick”: Authorship Research and Computational Stylometry Revisited

Abstract

Both the authorship and the dating of the so-called “Diktat für Schlick” (DFS), once attributed to Ludwig Wittgenstein and assigned by Georg Henrik von Wright to the Wittgenstein Nachlass as item 302, are debated topics in Wittgenstein and Vienna Circle research. Schulte (2011) and Manninen (2011) hold that DFS was authored by Friedrich Waismann rather than Wittgenstein. Applying techniques from computational stylometry to the authorship question, the paper concludes that DFS is located stylometrically in the middle between Waismann’s and Wittgenstein’s writings, but slightly closer to Wittgenstein, and so Wittgenstein authorship is hence stylometrically still not unlikely. The paper concludes by presenting a number of factors that speak in favour of the view that DFS might originally indeed have been dictated by Wittgenstein. For the computational stylometry component, the paper uses the Eder, Rybicki and Kestemont’s (2016) “Stylometry with R” package; the degree of similarity and dissimilarity between documents is calculated by Burrows' Delta measure; and the results are displayed using Hierarchical Cluster Analysis and Principal Components Analysis. For the text corpus part, the paper uses texts authored by Schlick, Waismann and Wittgenstein. For the archival research part, the paper refers to materials form the Schlick Nachlass in the North Holland Archives, the Waismann Nachlass in the Bodleian Libraries, the Rose Rand Nachlass in the Pittsburgh Archives of Scientific Philosophy, the Ludwig Wittgenstein Nachlass in the Trinity College Cambridge Wren Library, and the Cornell copy of the Ludwig Wittgenstein Nachlass. The paper is a follow-up on Oakes and Pichler (2013); for the current paper we have extended the Waismann text corpus with more texts written under the influence of Wittgenstein, a.o. Logik, Sprache, Philosophie (1976).