However, digitisation poses challenges too. These new digital sources are no longer “originals” in the classical sense, but information that has been constantly manipulated, recoded, and reframed. Dealing with a retro digitised source on a screen is very different to sitting in an archive and holding an original letter in your hand.
The methodology of source criticism, developed in the 19th century, is a key scientific competency for historians: being able to understand whether a source is original and authentic, or if it has been manipulated, copied or changed in any way. This methodology needs to be updated for the 21st century as “historical data criticism” — but that poses some foundational questions. What constitutes evidence? How can we authenticate both sources that have been retro-digitised and those that were born digital?
Future historians will need new skills and competencies in order to understand how to analyse and interpret digital sources and documents. In the era of deep fakes and fake news, this isn’t just an issue for historians; given their expertise in source criticism and critical contextualisation of evidence from the past, historians have a lot to contribute in the emerging field of “digital forensics”.
Digital history emerged as a field 30 years ago, closely related to what was then known as computational humanities. Its first phase focused on how new mainframe computers could be used to collect and manage historical data – for example the collection of economic data (sales statistics), environmental data (weather records), and demographic data (marriage and birth rates) over long periods of time. The second phase, starting in the 1990s, was defined by the mass digitisation of analogue sources by cultural heritage institutions such as archives, libraries and museums.
Currently, we are in the field’s third phase, which is focussed on the critical reflection of how new digital research infrastructures and tools change, how we think about history and what narratives we produce. In short, how does the digital affect our imagination of the past and thus our conception of history?
This involves an all-encompassing reflection on how the digital interacts with the way we do historical research, including how we search for, manage and curate data; how we analyse data with new text mining tools; and how we use methods of data visualisation for the production of new forms of transmedia storytelling. In addition to source criticism, we need new competencies and skills to critically evaluate how we search, analyse, visualise, and interpret historical data. Understanding how digital infrastructures, data sets and tools are altering our object of study is called digital hermeneutics.
Next to the methodological challenges, the digital revolution has fostered new epistemic values and virtues such as openness, transparency, traceability and sharing, which now sit alongside the traditional epistemic values of objectivity and accuracy. However, divisions and inequities in the infrastructures that determine access to knowledge are being reproduced in the digital realm, with even greater impact. For example, issues such as the overrepresentation of English and European languages in mass digitalisation of cultural heritage collections mean that the rich cultural heritage of countries such as China or Japan or of the many oral knowledge treasures of indigenous cultures around the world is not feeding into the new digital knowledge base that we are building.
This cultural disenfranchisement has been referred to as "epistemic injustice" by the philosopher Miranda Fricker, and is becoming increasingly relevant when it comes to questions of democratisation and access to historical data, as well as the knowledge and historical narratives that are produced from the data that is available. To combat this risk, we must actively invest in the retro-digitisation of non-European, non-Western cultural heritage, over the long term.
Looking forward, moving into the fourth phase of digital history will require historians to develop a new form of reading of historical sources. Machine-driven reading of vast digital corpora can identify statistically relevant patterns and semantic topics, but this “explorative reading” of a dataset needs to be combined with the “close reading” of individual sources. Combining artificial with human intelligence will be key but requires an understanding of both the strengths and weaknesses of these two forms of rationality and logics of sense-making. Learning to move easily between these two forms of reading will require training a new generation of historians in a new cultural technique of information retrieval and interpretation which I frame as “scalable reading”. This is not only true for historians, but also for practitioners of other disciplines in the humanities, social sciences and journalism.
Change, however, requires co-evolution and time. With tools and technologies being invented and deployed at a much faster pace than their critical appropriation, our education systems are struggling to balance these two factors. In addition, technology has opened up the universe of information to everyone, rather than a well-schooled elite. Teaching must therefore focus on developing critical competencies for dealing with a wide range of sources rather than instilling students with deep but specific corpuses of knowledge. However, this will in turn require drastic changes in the curricula and management of academic institutions such as schools and universities — and these are generally very conservative organisations.
A second big challenge for the future of history as a field is to engage in interdisciplinary approaches and to practice history as collaborative and situated form of knowledge production. Digital history can be thought of as a “trading zone” between many disciplines (history, computer & data science, web-design, statistics, library and information science, cultural heritage & archival science, etc.), which asks for systematic co-design of research projects and outputs. Learning each other’s “languages” and understanding the different epistemic traditions require long-term investments that are usually not funded by short-term research grants.
Such collaborations also affect questions of authorship and recognition culture. The co-production of historical knowledge in the digital age necessarily involves co-authorship in terms of publications and outputs. Paired with the aim of multi-perspectivity and multi-vocality when it comes to the production of historical narratives, the future of digital history as collaborative practice will hopefully contribute to the democratisation of historical storytelling and the promotion of shared authority.