In addition, I've had to move ahead in defining the area of interest for (1) my book list exam and (2) the eventual dissertation.
Let me sketch out the idea I have right now, and I hope readers will be kind enough to offer thoughts, encouragement, suggestions, cautions, and so forth. Here it is:
I am interested in approaching textual issues through the application of probabilistic analysis.
Specifically, I want to investigate the use of probability models to illuminate syntactic and semantic patterns in difficult passages of Anglo-Saxon poetry. For example, I think these models could help predict recurring dependencies in syntax and semantics. These dependencies, in turn, would enhance our analysis of textual cruxes and gaps in the Anglo-Saxon poetic corpus. We would then have an additional basis of evidence for supporting or challenging editorial readings. These models allow us to ask questions such as “given wherever we are at place p in the text - that is, wherever we are syntactically and semantically – what are the most likely options for what will follow next?”
Interestingly these models heaviy depend on the input, that is, on the text(s) imported into the database to establish the basis for probabilities. With regard to Anglo-Saxon poetry, these input texts would require us to deal with word breaks, caps, points, runic symbols, abbreviations, erasures and other intricacies of textuality in that period.
So a project like this places a premium on textual issues. At the same time, it encourages a kind of forensics of Anglo-Saxon scholarship where every crux and emendation has a textual and historical background that can be partially reconstructed.