DHBeNeLux: Incubator for Digital Humanities in Belgium, the Netherlands and Luxembourg

Commencing in 2014, the Digital Humanities BeNeLux initiative originated from a group of researchers and practitioners from Belgium, the Netherlands, and Luxembourg. This set of researchers felt that there was a need for a platform for people whose work related to or involved digital humanities methodology but whose work was not as academically visible as it could be. These people seemed to especially lack specific academic outlets to communicate their work. As a result, the DH Benelux conference[1] builds a platform for collaboration, connecting international researchers in the field of Digital Humanities from the Benelux region together with researchers from the European Union and beyond. The conference has a broad disciplinary range, from a diverse array of research projects. It is targeted at building a community that bridges the gap between early career researchers and eminent academic scholars.

From their own experience the researchers and practitioners who started the initiative knew that over the past two decades the use of digital and computational techniques and methods in the humanities in Belgium, the Netherlands, and Luxembourg had been on the rise [van Zundert and Dalen-Oskam 2014]. However, most instances of such techniques and methods advancing humanities research emerged in disciplinary related isolated pockets of methodological innovation. This mode of rather compartmented innovation arguably has been exacerbated by the region's rich quilt of smaller language communities. Within the BeNeLux the major official languages are Dutch and French. However, it would be more fair to summarize the language situation as at least a patchwork of West and East Flemish, Zeelandic, Dutch, Frisian, West Low German, Limburgish, French, Luxembourgish, and German. And even that summarization is just an approximation of the actual number of distinctly separate languages spoken in the area. Often methodological innovation has been tied to the interdisciplinary curiosity and stamina of individuals involved with research in language, literature, culture or history. Sometimes it resulted from modest interdisciplinary collaboration between individuals from different disciplines such as humanities and computer science or software engineering. Though their work was interdisciplinary in nature there was little knowledge exchange between the innovators of methodology themselves: they would collaborate with computer scientists or would import knowledge from software engineering into their own work, but they would do so from the relative isolation of their own humanities sub-discipline or institutional context (e.g. libraries or archives). Much like grassroots digital humanities emerging elsewhere we suppose, there was little overarching coordinated effort and knowledge exchange. Due to the strong local, independent, and autonomous progress of method, there was little incentive and not many opportunities for innovators to network, to share their knowledge, and to leverage the experience from comparable innovations from like-minded practitioners in other humanities disciplines. As such, these researchers and practitioners had little opportunity to obtain peer-level feedback and reflection on the pioneering digital and computational work they were undertaking.

Arguably the disjunct and isolated nature of computational and digital methodological innovation in the humanities in the Benelux has in many cases resulted in such innovations being relatively short lived. Often innovative studies go unrecognized as relevant methodological contribution or as research contributions in themselves [Schreibman et al. 2011] [Van Dalen-Oskam 2013] [Besser and Vaeseens 2013, 194]. New methods and techniques might serve a particular research project, only to evanesce quickly after researchers moved on. Digital humanities innovation has often been judged a “parade of prototypes” [Wouters and Beaulieu 2007], questioning the viability and validity of computational methods in the humanities (and elsewhere). If no steady user groups can be identified and if no considerable impact can be measured, then where is the clear and decisive benefit of this work? The argument however can be mirrored: if interdisciplinary innovative work is not recognized and if there is no cross-domain support for it, just how much long lasting and generic impact can we expect these methods to have?

Digital humanities has had its advocates and antagonists in the Benelux region just as it has had in other places. Though stark accusations of a neoliberal plot [Allington et al. 2016] have not been voiced, certainly researchers have pointed out a “silent ideology” underpinning a perceived “push” for digital and computational methods. Piersma and Ribbens for instance point to the fact that digital technology is still often presented as a neutral technology that is thus pervasive enough to make non-digital methods obsolete in the very near future [Piersma and Ribbens 2013]. They also argue that only rarely the underlying assumptions of the technologies and methods are questioned. In their view, these assumptions almost always point to an application of the empirical, quantitative, “scientistic” style of research. However, increasingly we see attention for the hermeneutic and abductive reasoning[2] potential of computational methods too. Certainly from an international perspective [Capurro 2010] [Flanders 2009] [Gibbs and Owens 2012], but progressively also in the Benelux [Beyen 2013] [van Zundert 2016]. Thus methodological reflection, particularly on the assumptions underlying computational methods may not abound, but it certainly seems to find some traction.

This–i.e. a potential modest rise in methodological reflection and the formation of theory–is arguably a good thing, as it points to the emergence of a theoretical stance that has been called for more than once [Liu 2012] [Robinson 2013] [McCarty 2016]. As one colleague aptly put it: method is the combination of theory with technique. If, as many seem to claim, digital humanities is a field, and a field of methodological interest foremost, then it cannot do without theory. And for that matter: can any academic field do strictly without theory? However, is it not far too early to speak of anything like theory with a capital “T” in digital humanities. We rather seem to be in the stage where each prototype is (or should at least be) a contribution to the formation of theory, which somehow emerges out of practice. Every database is a theory somehow [Bauer 2011], every digital edition is an argument [Shillingsburg 2013]. What we have not figured out so far is how to connect these individual dots of theory-turned-practical-prototype to mesh into a greater fabric that can be called theory of digital humanities.

Which takes us in a way full circle to the current status of digital humanities in the Benelux region. To practitioners in the field, the grand gestures and sweeping statements on theory and underlying assumptions of method are–apologies for the hyperbolic imagery–like the grand plans made up by generals being put to foot soldiers on the ground who are merely figuring out what works. They are digging trenches to shelter themselves from everything that is thrown at them from all sides, including organizational politics, institutional and funding policies, the resistance and accusations of traditional humanists, the overstretched promises of computational utopists, and difficult but justified questions of critical theorists. Many practitioners feel that this is too grand a challenge to be posed on digital humanities so soon: too much calling for unifying theory, too often questioning “where's the beef?” [Scheinfeldt 2010]. Like big infrastructures are a form of premature optimization for experimentation in the digital humanities [van Zundert 2012], so is calling for unified theories of it all while practitioners and researchers are merely figuring out what it is that we see with our new “microscopes” rather than why. We would argue therefore that the current state of praxis of digital humanities in the Benelux is excellent and appropriate. It is in an incubator era: the creation of prototypes thrives, many experimental projects are initiated. DH in the Benelux region is currently very rich, varied and diverse. A creative quilt that includes all: from advanced long term research by highly visible key players to many one off experimental projects, from institutionalized large infrastructure development to maker community projects tagging along on a shoestring. Prototypes indeed, and many of them. We would contend: the more the better. Computationally we still are just scratching at the surface of what we can do in the humanities with digital and computational methods. From this “prototype soup” some methodological commons and digital humanities theory may arise, but only if there are enough venues for practitioners to communicate and discuss and examine critically their methods and results.

Exactly for this–allowing practitioners to communicate and reflect on a methodological level–the DH Benelux platform was initiated. The digital humanities researchers and practitioners that initiated the platform wanted to facilitate the possibility to pioneers of computational and digital technologies in the humanities to share and learn from each other's work. They were much less concerned with a particular computational ideology or the establishment of a field. Even less, we would argue, were they occupied with institutional or organizational politics. They were simply looking for ways to mesh the individual nodes of methodological innovation into a fabric that would more effectively support computational and digital methods in the humanities in the Benelux. DH Benelux has therefore primarily been a community fostering collaboration: an informally governed group that furthers the networking and integration of distributed digital and computational methodological knowledge through community building. This is still at the core of DH Benelux: fostering knowledge exchange and collaboration on a methodological level. Most DH'ers seem not too concerned at all with questioning whether digital humanities is a field or “just” a temporary methodological pidgin, whether theory should have additional focus or not. That is not to say these questions are not important–we think they are. But the answers should emerge from the rich quilt of praxis that drives institutionalization of this methodological knowledge into digital humanities curricula that are slowly appearing on the map of the Benelux.[3]

Within its three years of existence the DH Benelux Conference has now become the foremost means for the Benelux digital humanities community to communicate, share and integrate knowledge. The first DH Benelux Conference took place in 2014. In The Hague at the National Library of the Netherlands some eighty presenters contributed 77 accepted submissions. Participants obviously derived from the three related countries. However, also a relatively large contingent joined from beyond these countries, which testifies to the embedding of the community in the larger region. The conference has since seen rising numbers of participants, a rising number of submissions (thus sadly also more rejections) and each year more contributions (from 104 presentations in 2015 in Antwerp, to currently 125 in 2016 in Luxembourg). Meanwhile the DHBenelux initiative also became a partner organization to the EADH[4], stretching its outreach further as may be gauged from submissions now including originating countries as far as the US and Taiwan. Further analysis of the audience and submissions show that the conference “is definitely not a clique of people who know one another, but actually a varied and growing community”  [Kemman 2016].

The DH Benelux Conference as a platform is still developing. The second conference in 2015 was the first to experiment with possibilities for researchers to publish their work in more expanded form. This special issue is a result of that initiative. For future installments of the conference we hope to be able to offer more general publication possibilities for paper contributors. This special issue draws upon some of the best work that was put forward at the DH Benelux 2015 Conference that was hosted and organized by Antwerp University.

Starting at the beginning, before digital humanities research can take place, the digitization of cultural heritage collections, often plays an important role. Yet, despite the increasing amount of digitized cultural heritage content being made available online, the accessibility of these collections remains limited. This is particularly due to a lack of user-friendly tools to explore such collections. Taking this as their starting point, and an online collection of Dutch folktales as their object of study, in “Supporting the Exploration of Online Cultural Heritage Collections: The Case of the Dutch Folktale Database” Iwe Muiser and his colleagues describe how they made use of user-centred design approaches to develop new interfaces to facilitate the browsing and exploring of their collection by both folk-tale experts and members of the general public.

Increasing the usability of digital cultural heritage collections continues in Max De Wilde and Simon Hengchen’s article “Semantic Enrichment of a Multilingual Archive with Linked Open Data,” this time with a digitized corpus of Belgian newspapers as the case study. In their article, they explore whether Natural Language Processing (NLP) techniques, such as Named Entity Recognition (NER) and entity-linking, can improve the search experience for end users of online historical collections. Addressing the particular problem of multilingual collections–their chosen corpus includes newspaper articles in Dutch, French and English–they demonstrate a language-independent method of mapping entities to the Linked Open Data cloud. As part of their research, they developed and tested MERCKX (Multilingual Entity/Resource Combiner & Knowledge eXtractor), with the aim of enabling semantic enrichment of digital collections by small and medium-sized cultural heritage institutions.

With a digitized and “cleaned” corpus there are a wide variety of methods that can then be applied for analysis. In “Coca-Cola: an Icon of the American Way of Life. An iterative text-mining workflow for analyzing advertisements in Dutch twentieth century newspapers,” Melvin Wevers and Jesper Verhoef show an example of this by analyzing to what extent Coca-Cola functioned as a symbol of an American way of life within the Netherlands using a corpus of advertisements from the National Library of the Netherlands (KB)’s digitized newspaper collection. In their article, they intentionally used a combination of traditional and computational methods to construct a sub-corpus for analysis and use the corpus linguistics tool AntConc, to answer their cultural historical research question.

Although with different research questions, J. Berenike Hermann also explores the multi-methodological setup of digital humanities. In her article, “In test bed with Kafka. Introducing a mixed-method approach to digital stylistics,” she aims at raising epistemological and methodological awareness within her research field of digital stylistics by using a practical, hands-on and ‘mixed-mode’ approach to analysing the prose of Franz Kafka. Using a combination of quantitative hypothesis testing, quantitative exploration and quantitative text analysis, she aims to pragmatically demonstrate the value of such an approach to her peers. While at the same time, doing justice to hermeneutic and empirical traditions in the field of literary stylistics.

It seems a small step from the Kafka’s disquieting world to the world of dreams. In “Unraveling reported dreams with text analytics,” Iris Hendrickx and her colleagues explore whether it is possible, using a combination of text analysis methods, including text classification, topic modelling and text coherence analysis, to computationally distinguish texts describing dreams from other personal narratives, such as diary entries. The intended goal of their research was to lay the foundations for innovating methods of dream analysis, including automatic detection of dream descriptions.

Moving from dreams to emotions, in “Mining Embodied Emotions: a Comparative Analysis of Sentiment and Emotion in Dutch Texts, 1600-1800,” Inger Leemans et al., explore how the mining of sentiments and emotions, can be applied to tracing the historical changes in emotional expression and the embodiment of emotions over time. In their article, they both present their Historical Embodied Emotional Model (HEEM) and reflect how it compares to other sentiment mining techniques using a corpus of historical Dutch theatrical texts as a case study.

We hope the reader will see how these articles demonstrate the rich palette of topic, method, and application that is currently the hallmark of digital humanities in the Benelux region. True to its mission to be most inclusive, some of the work presented at the conference and in more expanded form here, originated from outside the Benelux. We therefore trust that this special issue will give a good impression of some of the work currently going on within digital humanities in the Benelux and beyond. Exciting work that presents truly new research results. Work that also carves out new methods for the humanities. A small but high quality sample of the groundbreaking work that digital humanities researchers are contributing to the flourishing of the humanities in the Benelux.