From sources to data: historical people in the digital archive
Event Information
Description
The creation of digital archives and editions of historical texts opens up new possibilities for the creation of structured data on human populations of the past. This workshop aims to open up the journey from archival sources to data to critical scrutiny from a range of different disciplines, with a specific focus on the potential and challenges involved in the reconstruction of statistics about historical populations through case studies presented by the The Cambridge Group for the History of Population and Social Structure, the Casebooks Project and The National Archives.
A sandwich lunch will be served, so please help us cater accurately by cancelling your ticket if you book and find you can no longer attend.
Provisional agenda
11 – 12.15 - Session 1
The Aviva Archive. Its potential uses and plans for digitisation.
Oliver Dunn, Cambridge Group for the History of Population and Social Structure / Anna Stone (Aviva Archive)
When the combined Commercial Union, General Accident, and Norwich Union archive was established in 2000 it was considered to be the most important insurance archive in the United Kingdom. The collections derive from the merging of hundreds of historic insurance companies, some of which are the first of their kind. Life and fire policy records are particularly prominent from an early time, giving minute detail about health and living conditions from 1696 to the present day. These records seem to be suited to data-led historical research and genealogy, although suitability for these ends has never been fully explored. Access is a major obstacle to carrying out further investigation. Very few researchers have used the collections at the Aviva offices in Norwich. Aviva plc aim to increase access for scholars and the general public in consultation with Cambridge. To this end, this session will introduce the archive and show a selection of its records. We seek feedback about potential uses of the collections and how to increase their prominence.
Automated Text Recognition in the digital archive
Mark Bell – The National Archives
The Prerogative Court of Canterbury collection (PROB 11), held at The National Archives, contains Probate Wills spanning over 450 years (1384-1858). Historical research in this collection tends to involve sampling 10s or 100s of wills, out of around a million, for a particular study. With recent developments in Handwriting Recognition Technology (HTR) the potential is there to automate the transcription of the entire collection, opening the door to researchers using big data techniques to perform large scale longitudinal analyses of the wills. So The National Archives has started experimenting with the Transkribus platform to understand how well the technology works with this material. This talk will describe our experiences so far and explain how we hope to turn this fascinating collection into data.
12.15 – 12.45 Lunch
12.45 – 2 Session 2
Casebooks Project
Lauren Kassell, Casebooks project
This talk will introduce the casebooks of Simon Forman and Richard Napier, focusing on their data about 62,000 people living in England c. 1550 to 1650. It will set out what we have done with this data, and seek views about how other scholars might wish to use it and what we can do to facilitate their work. (Note: We are in the process of moving to a new site. If you would like access to it in advance of the workshop, please let me know.)
Data mining English Family History Society records
Gill Newton, Cambridge Group for the History of Population and Social Structure
This paper exploits several million burials of individuals in England from the 1500s to the early twentieth century to consider patterns of age-specific seasonality of death by settlement type. Salient information, drawn from transcriptions of parish burial registers created and maintained by local Family History Societies in many different formats, has been standardised to a common data structure. From this we obtain counts of burials per month and year, subdivisible by age into adult and child categories, for more than two thousand English locations/parishes. Each location is georeferenced to permit regional comparisons and facilitate the use of existing data on historical population size and other characteristics to create a simple urban/rural/other settlement typology.