Overview
Background
The Scottish Court of Session Papers are documents used in the presentation of cases in the Court of Session, Scotland's supreme civil court. They are the written pleadings of contested cases, plus associated documents. From the early 18th century until the mid-19th century, these papers were printed, with copies being provided for all those involved with the case. The papers give a valuable insight into the social, economic, political and legal history of Scotland, during the period of the Scottish Enlightenment, many of the notable figures of which were advocates. As a court of appeal and of first instance, the Court of Session in this period held jurisdiction over contract and commercial cases, matters of succession and land ownership, divorce proceedings, intellectual property and copyright disputes, and contested political elections. The Session Papers have been referred to as “the most valuable unstudied source for Scottish history....in existence.”
Scope
The Session Papers have great potential value for interdisciplinary scholarship, for genealogists and for local historians studying the historical period from the immediate aftermath of the Union of 1707 through the Jacobite wars, the Enlightenment, the agricultural and industrial revolutions and the building of Walter Scott’s Edinburgh – many of the notable figures of which were advocates (members of the Faculty of Advocates whose status and function correspond to that of a barrister in England).
Recognising the insight presented into the social, economic, political and legal history of Scotland, the University has agreed to invest £55,000 in the project.
Objectives
The principal objective of this phase will be to develop tools that will automate the creation of records and enable each case to be machine readable and thus searchable.
Building on the work already undertaken by Mike Bennett, Digital Scholarship Developer in the Library Digital Development team, and using the material scanned in the initial pilot, the second phase of the project will use computer vision techniques, OCR, and intelligent text analysis to automatically extract and parse case-level data in order to create an indexed, searchable digital resource for the Session Papers. If successful, the tool can then be adapted for use with other material, allowing us to work with other early printed texts, thus opening up a wide range of possibilities for digital scholarship. Work will focus on “training” the open source Tesseract OCR engine. Details of the initial process are outlined here - http://libraryblogs.is.ed.ac.uk/librarylabs/
The project will also address questions on how and where to best store content and data for both long term preservation and access.
In this phase, a basic project website will be created and images will be made available through the Luna Image Management System. Longer term, it is envisaged that the image files will be moved to a IIIF server so that images from the three institutions can be compared alongside each other and perhaps with other material such as NLS maps. Integration with Geonames and Map APIs (e.g. the NLS Map API), infrastructure planning and IIIF development will therefore form a key part of this phase of the project.
Additionally, more digitisation will take place, allowing us to increase the sample size and further test the image tools under development. Following on from the recommendations set out in the Phase 1 pilot, a number of scanners were viewed and the one best suited to this material was identified as the i2s Copibook V Shape RGBA 2-600dpi Optical Scanner. One of these has now been purchased and installed within the Digital Imaging Unit in the Library. Moreover, by timing the digitisation work to coincide with the end of the Theses Digitisation project, some of the equipment and digitisation staff currently working on that project could be used for the work on the Session Papers. An initial assessment conducted by Nicole Devereux from the Conservation team has concluded that some of the scanners will be suitable for particular volumes of the Session Papers.
A sample of 300 volumes will be scanned in this phase of the project, selected proportionately across the three institutions and further split equally into the three condition categories outlined in the Phase 1 Conservation report – good, fair and poor.
This will therefore be broken down as follows:
- University of Edinburgh - 24 volumes (8 in each condition)
- Signet Library - 72 volumes (24 in each condition)
- Faculty of Advocates - 204 volumes (68 in each condition)
Deliverables
- Conservation of collections' material for digitisation and development of methodology for larger scale conservation
- OCR tools
- Digitised texts, fully searchable
- Project website
Benefits
The project will result in a digital Scottish Session Papers collection freely available worldwide via the website http://collections.ed.ac.uk/, which would have sufficient critical mass to attract further research interest that could support a major research grant proposal to conserve and digitise the entire collection.
In the longer-term, we aim to create digital images of every item and then apply sophisticated OCR to create a searchable text to underlie the photographic facsimile. Geo-referencing, linking to other datasets and linguistic analysis will open up this collection to a wide range of researchers and communities who care about the history of Scotland.
A fully-searchable digital resource will allow scholars to answer a number of key research questions such as:
- What role did the law play in the government of Scotland during a period without local royal or Parliamentary presence?
- How does the data on social history which can be extracted from the Session Papers (e.g. information on payments, property, family relationships) confirm or question existing historical accounts of the period?
In addition to opening up a crucial period of Scottish history, the project’s “digitise first” approach also has huge potential for other large-scale digitisation projects using the IIIF model, where access and discovery will be built around the digitised content rather than a traditional catalogue.
With new advances in text processing and machine learning, we will be able to release online large and complex collections which are discoverable and navigable, of which the Session Papers is just one example.
This will hugely enhance research and community access to collections and will create new sets of text and data which will open up new avenues for scholarship; the Session Papers will not only become a searchable resource in their own right but the data can be linked to other digitised content, such as those listed in 4.12, to build up a picture of life in Scotland at that time. Moreover, by creating a database of the language used in the cases, we can create an important resource for socio-linguistic data.
Success Criteria
Project Milestones
(Please copy and paste from Milestones log)