Introduction This dataset collection is created within the context of the digitisation project ‘First-Hand Accounts of War: War Letters (1935-1950) from NIOD Digitised’, that ran over a period of three years (2020-2023) and was funded by the Mondriaan Fund, the Dutch Ministry of Health, Welfare, and Sport, and the NIOD Institute for War, Holocaust, and Genocide Studies in Amsterdam. The aim of the project was to preserve, digitize, and transcribe the NIOD’s war letters collection and to enhance access to these historical records in various ways.

Creator The NIOD Institute for War, Holocaust, and Genocide Studies was founded in 1945. The NIOD is a national and international archival institution and research institute. The NIOD-researchers conduct interdisciplinary academic research into the history of wars, mass violence and genocides. The institute holds over 400 archives and collections (2500 meters) about various topics related to World War II and mass violence in the 20th century.

War letters The project ‘First-Hand Accounts of War: War Letters (1935-1950) from NIOD Digitised’ digitised NIOD’s paper archival letter collection, also known as ‘247 Collectie Correspondentie’. The collection contains personal correspondence written and received in the context of the German Occupation of the Netherlands (1940-1945) and the War of Independence in Indonesia (in the late 1940s). Many people have been donating personal correspondence to NIOD since 1945 and new documents are acquired on a regular basis. The vast majority of the letters are written in Dutch and originate from the period 1935-1950. The archival collection ‘247 Collectie Correspondentie’ currently measures 14,1 meters and is divided into different inventory numbers. The collection entails a wide variety of different kinds of personal correspondence from various letter-writers.

Contents The dataset ‘First-Hand Accounts of War: War Letters (1935-1950) from NIOD Digitised’ contains machine-readable data in plain text and structured file formats. The data is aimed particularly at researchers interested in the (computational or quantitative) analysis of personal correspondence (‘egodocuments’) in bulk. The dataset consists of four different folders:

1. Handwritten Text Recognition (HTR) model ‘NIOD_WarLet_1935-1950’

The Handwritten Text Recognition (HTR) model ‘NIOD_WarLet_1935-1950’ is trained using READ COOP’s Transkribus software (PyLaia HTR). The computer model is based on ‘Ground Truth’ transcriptions of high-resolution (600 dpi) scans of handwritten correspondence.

Contents:

README (.txt) with URL to web page with trained public Handwritten Text Recognition (HTR)-model and online demo version (building on the READ COOP’s Transkribus server).

1. Ground Truth War Letters Transcriptions

The folder ‘Ground Truth War Letters Transcriptions’ includes 966 manually generated and checked transcriptions of high-resolution (600 dpi) scans of handwritten wartime correspondence.

Contents:

README (.txt) Ground Truth Transcriptions in ALTO-XML, 966 files (.xml)

1. War Letters (1935-1950) Transcriptions

The folder ‘War Letters (1935-1950) Transcriptions’ includes a large number of automatically generated transcriptions of historical handwritten correspondence. The dataset also includes the 966 manually transcribed and checked transcriptions in the Ground Truth War Letters Transcriptions Dataset.

Contents:

README (.txt)

3.1 War Letters (1935-1950) Transcriptions Dataset (per inventory number) Transcriptions in plain text format, 1480 files, (.txt)

1. War Letters (1935-1950) Metadata

The folder ‘War Letters (1935-1950) Metadata’ contains a matrix with the metadata (inventory-number level). This metadataset contains additional information about the data, linked to identifiers (ISIL-codes) related to inventory numbers. This metadata scheme includes the following features:

isil: International Standard Identifier for Libraries and Related Organizations (ISIL) code as used by the NIOD Institute for War-, Holocaust-, and Genocide Studies and additional information about the collection (247) and the inventory number of the sub-collection. Example: ‘NL-AsdNIOD_247_0310’. inv_no: The inventory number of the particular sub-collection. Example: ‘310’. no_of_scans: The number of scans (retrieved from the Transkribus-folder containing all scans of each inventory number). Example: ‘1’. pid: persistent identifier of the particular sub-collection. Example: ‘https://proxy.archieven.nl/0/21FE75B1551287C9E0538A77ABC22FE9’. description: A short description of the contents of the sub-collection, derived from the archival index of the NIOD. Example: ‘Brief van Dolly aan haar vader en moeder’. date_range: Date of writing/sending a letter, or date range in case of a correspondence. Example: ‘15 augustus 1943’. keyword_inventory_1: A keyword referring to the contents of the inventory number, derived from the archival inventory index of the NIOD. Some keyword terms include synonyms (in Dutch). Keywords and index categories sometimes overlap, sometimes not. In the case both an index category and a keyword were available, the index category is defined as keyword_inventory_1, and the keyword as keyword_inventory_2 (see below). Example: ‘Duitse burgers en militairen’. keyword_inventory_2: For some sub-collections, a second keyword referring to the contents of the sub-collection could be derived from the archival inventory index of the NIOD. Example: ‘Militairen - Zie ook: Officieren, Soldaten’.

Contents:

README (.txt) Metadata dataset (.csv)

Access and Terms and Conditions

1. Access

NIOD Institute for War, Holocaust, and Genocide Studies (hereinafter ‘NIOD’) strives to enhance access to and use of data that is collects and publishes for academic research purposes. NIOD makes a big effort to protect the privacy of living people and to track down copyright rights owners.

If you wish to use these datasets you must ask permission via the button on the right (‘Access File’) or via: onderzoeksdata@niod.knaw.nl as Copyright, GDPR or other legal restrictions may apply.

1. Terms and Conditions

NIOD Institute for War, Holocaust, and Genocide Studies (hereinafter ‘NIOD’) strives to enhance access to and use of data that it collects and publishes for academic research purposes. NIOD provides any individual or (legal) entity (hereinafter ‘you’) with access to this Dataset free of charge. The use of this dataset is subject to the terms of this agreement (hereinafter “Terms and Conditions”). Use of any data derived from the dataset, which may appear in any other format than text, such as tables and charts, is also subject to these Terms and Conditions.

1. Use of this dataset is permitted on condition that the dataset is used for non-commercial academic research purposes;
2. This dataset may contain personal information relating to living persons or data that is protected by the Copyright Act or other applicable legislation and/or regulations. Data that is protected by Copyright and/or is relating to living persons or persons who still may be alive, are intended for personal study and are not published or otherwise made public or distributed, unless explicit consent has been obtained;
3. NIOD is not liable for (the consequences of) the unlawful use of the data (referred to under 2.1. and 2.2) by you.

NIOD aims to protect the privacy of living people and to track down copyrights owners. See our privacy policy for more details. If you have any privacy concerns, including requests to remove your name or any other personal information, please contact us via onderzoeksdata@niod.knaw.nl. If you think you have rights to materials that are used in this dataset, you can also contact us via the email address listed above.