LLS Colloquium: "A data-driven exploration of stylistic variation in historical English"

On Wednesday 24th November 2021, Bettelou Los (University of Edinburgh) will be presenting on "A data-driven exploration of stylistic variation in historical English", looking at stylistic developments in English historical texts using POS-tagged texts, building on work by Los & Lubbers.

A data-driven exploration of stylistic variation in historical English

The paper investigates stylistic developments in English historical texts using POS-tagged texts from which the lexical material has been stripped, leaving only the tags. These sequences of tags are then n-grammed; the n-grams are input to various statistical analyses that single out the sequences that are most significant markers of the variation between the individual texts. This talk reports on two case-studies: Old English texts and Horse manuals from Early Modern English to the present day. In the former, the clustering of the texts was clearly driven by text type – mostly narrative versus other types; in the latter, the clustering conformed to the actual chronology of the texts (i.e. a chronological progression bubbled up naturally even though the dates of the text were not part of the data fed into the various procedures). The chronological progression that emerged in the data reflects a change in styles. Horse manuals contain at least two text types, to varying degrees: instructional/procedural “cookery book” writing and the logical-argumentative text type of scientific writing. The latter has been known to exhibit higher frequencies of nominalization and passivization, and this is also evident in these texts. I will discuss the nominalizations in the context of the function of restriction, which can be expressed as premodification (adjectives) and postmodification (relative clauses and prepositional phrases) but also by conditional clauses. Passives are part of the same package, as they are used with inanimate subjects containing nominalizations, and represent a change in the type of referents that need to be tracked in the discourse: from protagonists like horses and their human carers, to scientific processes describing horse digestion and crop production. The data further suggest that the flow of given to new information is not adhered to as strictly in the earlier texts than later; in the earlier texts, pronouns are found more frequently in end-focus position, and the strategy of using the by-phrase of the “long passive” to manoeuvre a new agent into end-focus position is only found in the later texts.

The talk will take place at 4pm on Zoom, and there will be an opportunity to ask questions at the end - you can join using this link. 

Event poster: A data-driven exploration of stylistic variation in historical English