Live notes, so an incomplete, partial record of what actually happened.
Tags: dhist
My asides in {}
Stream/Deck: http://ihrdighist.blogs.sas.ac.uk/2017/01/01/tuesday-17-january-2017-tim-hitchcock-tbc/
Overuse of text in digital history .. beyond architects and gamers and archaeologists we aren't very good to 3d, at sound, at space
Plan for talk: text mining proceedings, then 3d modelling, then speech mining, then experience.
1694-1913, 197k trials, 127 million words of text, around half of which is speech. Legal record of a bureaucratic process.
Old Bailey of the Proceedings https://www.oldbaileyonline.org/ no longer exists.
How do we get an overall sense of the trials? Basic distant reading. Learnt a great deal about the corpus, a corpus he already thought we knew a lot about.
"We [historians who text mine] assume that all text is equal. Quite frankly it is not"
But linguists don't. Magnus Huber tagged the speech. Giving us a new object of study.
What is said to an audience shows us real people performing to an audience. Speech text is fundamentally different as evidence, moored to particular knowable experience. There is a feedback loop between creation of 'text' and validation via reception.
To know that validation we need to get our heads around the audience. The space. The courtroom.
Old Bailey circa 1774-1784 (Dance's rebuild) becomes model for other courts in England and around the world (though not USA, which is modelled on church).
Changes in the courtroom geometry not represented in text mining of the old bailey court. So modelling the space seemed crucial.
Building the model forces the historian to critique the sources, to decide what is right and what isn't.
Conclusions based on the model:
- compact space (unlike the Pugin/Rowlandson/Ackermann image)
- community of the courtroom (given the small space)
- importance of levels .. hierarchised spaces .. created the conditions for theatricality (Garrow et al)
Back to text mining, but this time of speech.
- decline in % of speech in court .. bureaucracy moved to police context
- decline in complex conversation
- increase in wordiness, combined with an increased muteness of defendants, guilty voices increasingly excluded [even before verdict]
Analysis of speech text reinforces the analysis of non-speech text.
Need to take more seriously the space as a place of speech. But not actual spoken speech, rather the sounds that differ by age, height, weight.
What do we learn from the classicists?
If the Old Bailey had survived in its 1770s form, would you have modelled it anyway. Would it matter.
How do we make an argument about this?
Adam: My question would be how the findings can be unpicked from the changing editors/reporters
Say more about the planned sonification?
sound like a fake foreign language. Close your eyes and hear an exchange.
Chris Warren (Carnegie Mellon) -- 23 May -- 6 Degrees of Francis Bacon
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Exceptions: embeds to and from external sources, and direct quotations from speakers