Tracing patterns of contact and change: Philological vs. computational approaches to the handwritings of a 18th century migrant community in Berlin

Historical manuscripts are an essential source for all humanities. Since many communities have relied on a purely handwritten culture for a long time, fostering access to historical handwritings is an important and potentially fruitful endeavour. Traditional philology and other text-centered humanities have developed a received methodology of accessing old manuscripts, which involves research on the text-external context, close reading, transcription, critical edition and much time-consuming textological "detective work". While direct support by computer science and engineering through reliable Optical Character Recognition (OCR) tools for handwriting is technically still out of the question, modern image and pattern recognition techniques promise being able to distinguish personal handwritings and isolate pre-defined graphic templates in them. The present project plans on confronting these two methodologies and reflect systematically upon the question which of the two approaches is more adequate and successful in the long run. As a major outcome, we will devise and implement an assistance system for accessing historical handwritten texts which puts to use the best of both methodologies. Our object of study are the previously unedited 18th century texts of the small, but very interesting Czech-speaking community of religious "exulants" to Berlin, the (ancestors of the) Herrnhuter Brüdergemeine in present-day Berlin-Neukölln. Their texts – autobiographies and sermons – are valuable for Czech language history, but hard to assess without knowing about the authors, scribes, translators and their background. Based on these documents, we will try our best in both approaches to uncover layers of handwriting, textologically important writing features (spontaneous vs. copied, self vs. external correction), and repetitive patterns of words. By answering these demands, we not only tackle a methodological desideratum, but also contribute to our knowledge of the Brüdergemeine, their historical situation and their everyday life. Most interesting from a linguistic point of view is the intensive contact situation with German, and its consequences for language change. In the best case, the methodology of historical philology and of computer science would inspire each other and lead towards a useful tool to support the analysis of handwriting.

Principal Investigators
Meyer, Roland Prof. Dr. (Details) (West Slavic Languages)

Volkswagen-Stiftung (VW)

Duration of Project
Start date: 04/2017
End date: 09/2020

Research Areas
General and Comparative Linguistics, Typology, Non-European Languages, Humanities and Social Sciences

Last updated on 2020-06-06 at 00:05