Introducing the Language Lab

At the Language Lab we are exploring the social and cultural impact of the Industrial Revolution as reported in newspapers and other types of textual sources.

We want to investigate how the relationship between humans and machines was discussed in the sources, and what prompted individuals to write about machinery. We want to explore the ways in which negative narratives and fears were articulated, as well as the stories of success and positive claims about progress. Did sentiment and emotions towards technology change over time, and, if they did, in what way? Can we find a way of modelling the voice of the press (in terms of verbosity, exaggeration, sensationalism, and language conservatism) with respect to the changes brought about by Britain’s rapid transformation into an industrial society? Radical changes in society are known to relate to changes in language too: can we therefore trace linguistic change in the context of the Industrial Revolution?

A common denominator among these questions is that answering them (or getting closer to answering them) will require an intertwined combination of distant reading and close reading. The vast scope of this project is something we find both thrilling and a bit daunting. We are working in close collaboration with the other labs towards the development of generalisable computational methods that will support processing large volumes of text and analysing and interpreting their contents at scale. It is our goal to plan and drive innovative experiments from conception to completion, that will push the state of the art of research in semantic processing of historical texts.

The Language Lab will be involved in a wide range of tasks, from ensuring the quality of the most basic linguistic preprocessing steps in our data, to performing sophisticated semantic analyses that will aim at answering specific historical research questions. Some of the tasks in which we are taking an active part (together with other labs) are: the creation of gold standard datasets and of a lexicon of machinery, the development of methods both for geo-referencing mentions of places and for lexicon expansion, and the assessment of the impact of bad OCR in historical texts. As a first minimum research outcome of the language lab, we are exploring to what extent machines were being seen as agents able to drive change. You can read more about it in this other blog post.