Palaeographers are typically asked three questions of medieval or ancient handwriting: when was this written, where was this written, and by whom was this written. In general the people so asked are highly specialised experts with many years of experience, but their answers tend to be given with little or no justification (Derolez), sometimes explicitly arguing that ‘human’ processes such as handwriting cannot ever be subject to ‘scientific’ argument (Costamagna et al.), or that palaeography cannot be taught but only acquired through practice (Pfaff).
In contrast is an attempt to obtain greater rigour through quantitative methods, an approach which has undergone renewed interest since the widespread availability of computers. However, this approach typically reduces script entirely to statistical processes which are themselves difficult or impossible to evaluate, replacing the impenetrable human expert with an equally impenetrable digital one (Stokes). Such methods also depend heavily on subtle and often unrecognised assumptions about the underlying data (Scully and Pasanek), and focus on scribal identification at the expense of dating, localisation and indeed teaching.
The paper proposed here presents a different approach: using technology to make evidence available in new ways for discovery, citation, and teaching. It draws in part on Davis’s argument that visual evidence in palaeography is generally more compelling than verbal. Palaeographers have used visual evidence since the field’s inception, but technology has now allowed this on a new scale and in new ways: this has proven particularly successful with historical content that cannot fit neatly into fixed categories normally demanded by a computer, and such evidence is often more intelligible than either statistical processes or verbal description (Jessop). Possible examples include arranging images of letters on interactive timelines or maps, creating grids of letters arranged by scribe (or genre of text), and connecting these to images of the full page or descriptions of the manuscript. A working prototype will be demonstrated to illustrate these principles applied to Anglo-Saxon script, the goal of which is to allow exploration of content and the presentation of evidence in support of palaeographical arguments in a way which allows both for judgment and for data, perhaps thereby overcoming the weaknesses of the two competing schools while keeping the best of both.
Costamagna, Giorgio, et al. ‘Commentare Bischoff’. Scrittura e Civiltà 19 (1995): 325–48 and 20 (1996): 401–7.
Davis, Tom. ‘The Practice of Handwriting Identification’. The Library 8.3 (2007): 251–76. doi:10.1093/library/8.3.251
Derolez, Albert. The Palaeography of Gothic Manuscript Books. Cambridge, 2003.
Jessop, Martyn. ‘Digital Visualization’. Literary and Linguistic Computing 23 (2008): 281–93. doi:10.1093/llc/fqn016
Pfaff, R.W. ‘M. R. James on the Cataloguing of Manuscripts’. Scriptorium 31 (1977): 103–18.
Sculley, D. and B. Pasanek. ‘Meaning and Mining: The Impact of Implicit Assumptions in Data Mining for the Humanities’. Literary and Linguistic Computing 23 (2008): 409–24. doi:10.1093/llc/fqn019
Comments