Today’s interview features Christoph Sander. He studied philosophy in Freiburg and Berlin. In 2019, he received his Ph.D. from the Technische Universität Berlin with a thesis on the conceptions of magnetism in the early-modern period. Since then, he has been a postdoctoral researcher at the Bibliotheca Hertziana in Rome (Max Planck Institute for Art History) and at the Max Planck Institute for the History of Science in Berlin. His interests include the digital humanities, www.raramagnetica.de being his main platform to share his digital research.

Editor: What got you interested in marginalia?  

CS: I initially became interested in marginalia because they offer an intimate insight into the reception of a text at a specific time. Simply put, I wanted to see what a past reader spontaneously noted in their own copy. I was especially intrigued to see if this specific interest could be linked to the broader reception of a work. 

Editor: Do you remember your first marginalia?  

The first marginalia I examined as a historian were from the medieval commentaries on Aristotle’s writings. I noticed that translators oddly translated Aristotle’s rare use of the word “demon” as “scientia”. Without delving into why they chose this translation, it was evident that individual readers of these translations either restored the original word “demon” in the margins or, conversely, added the translation next to another passage containing the word “demon”. I have written on this: Christoph Sander, “Der Dämon im Text: lateinische Lesarten von De somno 453b22 und De divinatione per somnum 463b12 zwischen 1150 und 1650,” Recherches de Théologie et Philosophie médiévales 83, no. 2 (2016): 245–311.

Image 1: Aristotle, Textus Paruorumnaturalium Arestotelis per tractatus [et] capitula distinct[us] De nouo emendat[us], Leipzig 1496–99. Source: HAB Wolfenbüttel (A: 95.7 Quod. 2° (2)).

Editor: What do you find most interesting or powerful about marginalia?  

CS: As mentioned, I’m drawn to marginalia primarily because they provide a close look into readers’ understanding and interpretive horizon. This is particularly intriguing as it often reveals subversive readings that may not be captured in the reception history. Lately, another aspect has caught my attention: I aim to learn more about the contexts for marginalia, such as which themes in texts are most frequently annotated by readers. I view marginalia broadly as any intentional trace of reading and sense-making, including not just text in the margins but also underlining, strikethroughs, dog ears, and any marking interpretable as a trace of reading. Evaluating these traces quantitatively in relation to the subjects covered in the marked texts provides a complex network. Analyzing this network practically requires powerful computers and specific models. Also, a large amount of material is necessary to make connections visible. 

Editor: What marginalia are you working on now?  

CS: In my “Magnetic Margins” project, I’ve examined all the copies of early modern editions of certain works on magnetism and recorded all reading traces in a database. This database is publicly available at www.magnetic-margins.com. The benefit of this approach is that it’s essentially the same text across hundreds of copies. Thus, statistically, it’s relatively easy to determine which content particularly interested readers. The more markings related to specific content, the more interest in that content can be inferred. My hypothesis is that marking a chunk of text reflects an interest in its conceptual content. This assumption seems especially plausible for scientific texts. 

Image 2: Preliminary statistical graphical evaluation for William Gilbert’s De magnete (PNG download). All annotations are summed per chapter across all copies of all four editions and multiplied by the number of copies in which the corresponding chapter was annotated. This results in a relative weighting (so absolute y-values are irrelevant), which normalizes if very few copies have very many annotations for a chapter.

Editor: What do you see as the future of marginalia studies?  
 
CS: I believe the future of this research field lies primarily in collaboration – not only among researchers but also between researchers and libraries, and between researchers and IT specialists. There are already initial experiments using artificial intelligence to automatically identify marginalia in scans. This would be a huge help, even if it’s not always 100% accurate. 

Editor: Is there anything else you want to tell us?  
 
CS: I find it notable that there isn’t a unified vocabulary for describing marginalia. While the reasons are clear, library sciences have typically been more pragmatic. Historians and philologists, deeply engaged with this phenomenon of marginalia, sometimes overemphasize minute specific differences or discuss historical categories at a very abstract level, rather than recognizing the potential of a shared vocabulary. This shared vocabulary, even if imperfect or somewhat anachronistic at points, can create international synergies for collaboration. Furthermore, it’s disappointing that many libraries still don’t provide their scans with an open license. This can complicate the lawful use of this material and can sometimes be prohibitively expensive.