Processus Universalis

Certain procedures in historical recipe groups became the focus of extensive reworking and circulation. One such cluster of texts is the Processus Universalis, which exists in over 100 known manuscript versions.

Computational approaches are used to categorise alchemical recipes that belong to the same recipe group but survive in different stages of development. The goal is to better understand how these texts relate to one another, especially since many are undated and their relationships remain unclear. Unlike a traditional philological stemma or collation process, reconstruction is complicated by the fact that practitioners actively experimented with these recipes in the laboratory. As a result, textual transmission is not straightforward. Many of the texts are interrelated. Digital methods help identify patterns that may bring greater order to this complex corpus. Besides text reuse detection, I have employed stylometric and phylogenetic analyses on this coprus.

I started the Processus Universalis project in 2018 together with the Gotha Netzwerk Alchemie.

The edited volume resulting from our collaboration at that time is scheduled for publication in 2026 and is currently in press (missing reference). This work has also inspired me to think about the relationship between experimental and computational methods in history of science and knowledge (Lang, 2022).

Because the methods used in the original project have since evolved, I revisited the material in 2026. A number of forthcoming publications re-examine the original data and ground truth by investigating the relationship between linguistic surface features and content. To do so, I apply methods that focus either on language features or on knowledge extraction and representation.