Our current research projects

Digitising the National Farm Survey

Funder: The Lund Trust
Timeframe: October 2023 – March 2027

The National Archives is digitising the National Farm Survey (MAF 32 and MAF 73) in full, thanks to a generous grant of £2.13 million from Lund Trust. The 1941 National Farm Survey is one of the most comprehensive records of land that we hold in our collection and is a window in time on the UK’s agriculture and land use in the middle of the Second World War. Containing extensive data on over 300,000 English and Welsh farms, the survey is among the most-requested record series at The National Archives.

Currently, the complex filing of the paper record makes it difficult for readers to order and use, with the records only available in physical copy. This project will digitise the series in full and create a new digital cataloguing arrangement to make each farm searchable online. Genealogists, family and local historians will be able to consult the series for their own research, and the project will lay the ground for new analyses by historical economists, geographers and ecologists.

In pictures

The National Farm Survey, 1941–1943

A map of an area which has been divided up into sections and coloured in by hand

The National Farm Survey was created during the Second World War to increase food production and to plan for post-war agriculture.

Back on Track: Tilbury Riverside Station

Funder: National Lottery Heritage Fund
Timeframe: 2025 – 2028

The project will see the restoration of the Grade II listed Terminal complex at Tilbury Docks and combine commercial and community uses, supporting people into work through skills development opportunities and community activity, celebrating the global connections at the Cruise Terminal as a culturally significant part of the national story. For The National Archives this project is an opportunity to continue to build relationships with community partners and continue to raise awareness of national collections through a local lens.

Dr Chloe Lee continues to develop her work as a practitioner to ensure communities are able to access to resources and historic records, that are significant for regional areas, such as Thurrock. She is overseeing the delivery of support from The National Archives and will be working with the Education team and colleagues in Collections Expertise and Engagement.

Project Odyssey – Opening the National Archives' Legal Data to AI for A2J

Funder: UKRI Innovate UK
Timeframe: June 2024 – March 2026
In collaboration with: Tabled Technologies Limited (lead), Keele University, Northumbria University and Swansea University

The National Archives Legal datasets are the primary source of legislation and case law in the UK jurisdiction, available free of charge for computational analysis. However, the datasets have not been annotated or enhanced to optimise them for machine learning. This is an increasing problem in the era of pre-trained public LLMs that often fail to provide accurate legal interpretation to uniformed users.

The project addresses Sir Geoffrey Vos, Master of the Rolls and President of the Civil Courts’ statements:

If GPT-4 (and its subsequent iterations) is going to realise its full potential for lawyers… it is going to have to be trained to understand the principles upon which lawyers, courts and judges operate… the present version of ChatGPT does not have a sufficiently reliable moral compass

The Project comprises:

Enriching The National Archives XML files with additional metadata specifically designed to improve interoperability and accuracy of LLMs. An automated GPT-based annotation system will be developed to ensure the ever-expanding NA datasets remain annotated for all users.
Fine-tuning & evaluation of GPT-based LLMs on the corpus. State of the art LLMs (GPT, Llama2, Claude2) will be assessed, to hone the accuracy, ethics, and legal suitability for use.
Plain-English, standardised LLM prompts for legal subject matter information summarisation will be created for consistent input into LLMs.
For ESG purposes, the LegalPath A2J app for litigants in person and SMEs will be enhanced using the outputs of (i), (ii) and (iii). The app will enable members of the public to construct standardised prompts to access the fine-tuned LLMs, trained on the NA datasets, to achieve desired results.
SME consortium members will develop adjacent legal access products, while the academic institutions will disseminate findings. Recommendations on the use of LLMs in dispute resolution will be circulated to the Ministry of Justice, Regulators, and the Online Procedure Rules Committee.

Virtual Records Treasury of Ireland

Funder: Government of Ireland
Timeframe: 2023 – 2025
In collaboration with: Trinity College Dublin (other core international partners include National Archives Ireland, Public Record Office of Northern Ireland, Irish Manuscripts Commission and Trinity College Dublin Library)

The Virtual Record Treasury of Ireland is an all-island and international collaboration aiming to re-imagine and reconstruct through digital technologies the Public Record Office of Ireland, which housed seven centuries of Ireland’s archival heritage and was destroyed on June 30th, 1922, in the opening engagement of Ireland’s Civil War.

The Treasury is an open-access resource, freely and permanently available online to all those interested in Ireland’s deep history at home and abroad. It brings together images, transcripts, translations and other data from surviving originals, official and antiquarian transcripts and other replacement records for those lost in the blaze. The National Archives supports the Medieval Exchequer ‘Gold Seam’, a digital edition of over 150 years of financial records deposited for audit at Westminster, and INSPIRE, which provides digital access and enhanced metadata to the papers of the secretaries of state in the late seventeenth and early eighteenth centuries. Together with the partners across Ireland and around the world, the project is democratizing access to invaluable records and illuminating seven centuries of Irish history.

virtualtreasury.ie

The Virtual Record Treasury of Ireland

Find out more on the project website

Focus on

Revealing the (corrupt) record in 14th-century Ireland

In 1325, officials found suspicious mistakes in the account of the former treasurer of Ireland and hauled him through the courts. What did they uncover?

AI as Infrastructure (AIINFRA)

Funder: The Australian National University (ANU)
Timeframe: 2024 – 2026
In collaboration with: The Australian National University and Kings College London

AIINFRA will explore whether Large Language Models (LLMs) could enable a new era of transnational historical research. The project will run from 2024 – 2026 and be led by the HASS Digital Research Hub at The Australian National University. The project team includes representatives from the Australian Parliamentary Library, the National Library of Australia, the Aotearoa / New Zealand Department of Internal Affairs, the UK National Archives, the UK History of Parliament project, and the ANU Library, with academic input from King’s College London. Indigenous guidance is provided by the Scaffolding Cultural Co-Creativity project, and Taiuru & Associates Ltd.

AIINFRA will design and build a prototype open-source LLM tool tailored for historical research, but this will be in service of the primary goal of understanding the technical potential of LLMs and developing test categories appropriate to the academic and GLAM communities. To limit the scope of the project, focus will be placed on historical research. Source material will be limited to Hansard records from Australia, New Zealand, and the United Kingdom from the year 1901. Biases in the Hansard documents will require the LLMs to manage cultural and ethical issues, prompting consideration of Indigenous AI and data sovereignty. A subsidiary goal of the project is to explore the potential of enriching LLMs with secondary sources and multimedia to provide broader cultural and academic context.

aiinfra.anu.edu.au

AI as Infrastructure

Find out more on the project website

Responsible use of AI in the Creation, Reactivation and Conservation of Artworks and their Archives

Funder: UKRI
Timeframe: 2024 – 2027
In collaboration with: Nottingham University and the University of Exeter

ukri.org

AHRC: BRAID Responsible AI Demonstrators

Find out more on the AHRC website

European Holocaust Research Infrastructure (EHRI) UK

Funder: AHRC
Timeframe: 2025 – 2026
In collaboration with: Holocaust Research Institute at Royal Holloway University, The Parkes Institute at University of Southampton and The Wiener Holocaust Library

EHRI-UK is the UK national node of the European Holocaust Research Infrastructure (EHRI), a European Research Infrastructure Consortium (ERIC) focused on securing and providing access to sources on the Holocaust. EHRI-UK aims to secure the long-terms sustainability of Holocaust research by:

representing institutions within the UK (and Crown Dependencies) which contain materials related to Holocaust Studies within a strong research consortium
connecting resources through a state-of-the-art digital infrastructure
developing innovative digital research tools
offering fellowships and training opportunities for researchers, archivists, and heritage professionals

ehri-uk.org

The UK National Holocaust Research Infrastructure

Find out more on the EHRI-UK website

The Prize Papers Project

Funder: Akademie der Wissenschaften zu Göttingen (Germany)
Timeframe: 2018 – 2037
In collaboration with: The Carl von Ossietzky University of Oldenburg (Germany), German Historical Institute, London (Germany), Verbundzentrale des Gemeinsamer Bibliotheksverbund (Germany).

This is an archival, research and digitisation project covering the Prize Papers of the High Court of Admiralty, 1652-1815. The collection comprises the court process (in London) of c.35,000 enemy and neutral ships captured in wartime) and the related exhibits (all papers found on the ships). The exhibits form an extraordinary global archive, including ships’ trading and navigation papers, personal archives of seafarers (mariners and passengers), and c.160,000 undelivered letters, as mail-in-transit. These provide matchless insights into the global connections of early-modern lives, from a multitude of backgrounds, in 19 different languages. The project includes (at The National Archives) the sorting, identification and initial cataloguing of the papers, ship-by-ship, and the preservation and photographing of the papers; and (at Oldenburg) the creation of research-oriented metadata and the presentation of the digital copies and the metadata in an open access research portal. The portal development reflects our continuous engagement with current discussions in the field of Digital Humanities. We cooperate with numerous international researchers and research institutions working on the Prize Papers and related areas, and run an annual seminar series. We are very grateful to the William Nelson Cromwell Foundation for their generous support of the initial sorting/cataloguing of Prize Papers.

prizepapers.de