# Data Spirits: Data practices along and across the boundaries of humanities research > [!tip] Reference > Paper for the panel *Demarcating boundaries of and with data: Boundary work in the age of datafication* at [STS Hub 2025](https://sts-hub.de/25/), Berlin, March 11–14. ## Abstract 💁 Even within the humanities, disciplines speak and work differently. There is a considerable variety of methodologies and felicity conditions. As a result, boundaries are blurred, both between disciplines and to other forms of research. This makes collaboration tricky, but an interesting area of research. Data is an integral part of research and collaboration in this field, well beyond what is called digital humanities. Even if researchers do not use the term, working with data is as essential as it is mundane. This infrastructural layer is not just another problem to be solved, but a starting point for tracing ways of doing research in the humanities. The paper draws on a praxiography (Mol 2002) embedded in a long-term research institution consisting of a dozen of projects in the humanities. Here, the notion of _data spirits_ refers to the multiple ways, in which these researchers create and mobilize data through intelligent procedures to generate what they are looking for: spirited ideas. In these processes, data is used to validate but also to cross boundaries within the field. Diffractive attention to shared data can be one methodology to foster lateral collaboration across these boundaries by facilitating “respectful engagements with different disciplinary practices” (Barad 2007: 93). ## The site 🗺️ Imagine an interdisciplinary, humanities-based research institute at a German university. Round about 50 people from media studies, art history, medieval and literature studies. Many of them work on contemporary and digital issues, but most don’t employ digital-humanities-like methods. Their research data is neither big nor quantitative. Instead, they work with Word files on cloud drives, PDFs in folders, and scans of historical manuscripts, or images of architecture and carpets. The researchers at this site treat all of these digital and non-digital objects as research data; that is, they treat it as traces of the things that they are interested in. As research data, these objects are created and processed in *data practices* like sorting, annotating, or analysing. By these practices, objects and materials get mixed, marked up with colors or comments, dissected or fused by categories and concepts. At this site, I conduct an ethnographic study with a focus on research data practices of the not-so-digital humanities. Not everyone agrees that this kind of research uses data at all. At least, it’s unclear what role it plays or should play. I’m not an external researcher, but I hold a position in one of the subprojects of this institution. Doing an ethnography means I’m researching my colleagues: I work with them. I visit events and meetings. I organize workshops myself. I participate, I observe, I write ethnographic protocols, and I do interviews. ## The framework 🖼️ The title of my paper, the notion of *Data Spirits* is shorthand for the multiple ways, in which the humanities at my site creates and mobilizes research data to generate and care for what they are looking for. The term *Spirits* is derived from the German word *Geist*, which is the essential part of the German term for the humanities: *Geisteswissenschaften*. What is Geisteswissenschaften about? It must be *Geist*, right? Very much in the spirit of STS (pun intended), this is not to belittle the ways humanities scholars do their work. Quite the opposite: I like to celebrate the many ways in which research is being done here. It’s not the same data practices as in the natural sciences. It’s not even the same practice as in the digital humanities. That’s why I believe it’s important to understand these ways of mobilizing research data. I would like my research to enrich and multiply what research data can be, how it is used, and what can be achieved with it. So, there cannot be one, but many ways of *doing research data* as a humanities' researcher. To address this multiplicity, I use a framework revolving around the two main concepts of the paper: *data* und *spirits*. Here is a diagram to show the two components and their relations. On the one hand, there is *data*, that is, the processed and worked-with materials of research. On the other hand, there are *Geister/Spirits*, that is, the objects of interest but also products of research. As you can see, both components are mediated by data practices. ![[geistkreis-eng-250306.png]] ## Ghosting data 👻 Let's take a closer look and start with the weird one: *Geister/Spirits*. This is first and foremost a heuristic strategy. The term marks that, which the field tries to achieve without necessarily being clear about it. It states the assumption, that there are objectives within humanities research. And because these are probably heterogeneous and multiple, it might be smart to keep them vague. That’s why, I choose this open and deliberately naive term. As such, *Geist* designates, what humanities research is interested in. Both as an object of research but also as a product. At the same time, *Geist* is not without any meaning. It refers to the generic medium of meaning (*Sinn*) (Luhmann 1992: 44, fn. 47). While at the same time leaving undefined, to what extent meaning happens in heads, in the air, on paper or on screens. *Geist* (in the singular) understood as a medium means there are many forms and entities possible. So there isn’t just one *Geist*, but many *Geister/Spirits* (Kittler 1980). *Geister/Spirits* is the heuristic term to circumscribe the entities which the researcher in my field endeavor to study and produce – and for which they generate and process data. ## The data layer 💾 Let’s talk about data next. Frist, a short version of my working definition of research data: As I employ a praxeographic approach, data always exists in data *practices* (Mol 2002). These practices share the goal to fix traces of the phenomena of interest in a suitable medium (Rheinberger 2021). The result of such practices are *stable but mobile objects* (Latour 1990), which serve to witnesses the phenomena in question (Borgman 2015). This generic description is necessary to recognize data and their practices, even if they are not labelled as such in the field. Combined with careful participations and conversions, I have compiled a list of data practices from my field. Each entry is composed of more than one situated practice, so these are basically categories. If incomplete, the list gives a clear overview of the spectrum of research data practices at my site. ``` - Recording social processes (e.g. interviews, observations) - Recording technical processes (e.g. sensor data, software code) - Creating and editing text (e.g. notes, protocols, publications) - Annotating texts, images and other files or objects (e.g. comments in PDF or Word files) - Searching and collecting literature and other materials (e.g. Google, repositories) - Generating lists of literature and other materials (e.g. link lists) - Managing literature in specialized software (e.g. Zotero, Citavi) - Managing material in specialized software (e.g. MaxQDA, Obsidian, work database) - Managing local or remote files (e.g. Finder, Sciebo) - Individual or collective analysis or synthesis of material (e.g. coding, group interpretation, hermeneutics) - Compiling and organizing material visually or spatially (e.g. mind mapping, timeline) - Transfer of texts, files and other materials (e.g. e-mail, link, shared folder) - Archiving of texts, files and other materials (e.g. RDM, backup, external drive) ``` Most of these practices are focused on handling traces of thought and communication, forms in the medium of meaning, that is, traces of *Geist*. This is what I want to refer to with the term *Data Spirits*: The hunt for traces of meaning, refining and recombining it by these small-ish datafication procedures, to gain new traces of meaning compiled as books or PDFs. The notion of *Data Spirits* also refers to all the research data practices necessary to conjure and catch these spirits. All of these data practices happen on what I call the *data layer*. This *data layer* is the plane of research where these data practices are being performed, practically, materially, and close to technical infrastructures. ## Along and across boundaries 🚧 At least in my field, this data layer is undervalued and somehow neglected. Which brings me the boundary work at my site. There are two basic types of boundaries here: internal and external. *Internal boundaries* are those within the research institution, that is, between individuals as well as between subprojects. Research at my site is pretty dispersed and individualized. In that sense, almost all attempts to work together involve some kind of boundary crossing. Now, a lot of collaboration happens on the *Geist layer*. That is, like we do here today, presenting and commenting on finished papers. I propose, there could be put more care and attention into cultivating ways to collaborate on the *data layer* instead. I think, if there were more deliberate research data practices, this could foster collaboration across internal boundaries. The *data layer* is an underused medium for collaboration across boundaries. *External boundaries* are towards other institutions or disciplines. Most prominently, of course, towards the natural sciences. Research data management efforts are on the rise, but the notion of research data is very much coined by the natural sciences. In this situation, the not-so-digital humanities could either insist on not caring for data or – and this would be my proposal – care for a more deliberate research data practice that multiplies what research data can be. So, I end with a plea for boundary work that aims at cultivating one, two, many ways of doing research data! ## References 📚 - Barad, K. (2007). Meeting the Universe Halfway: Quantum Physics and the Entanglement of Matter and Meaning. Duke University Press. - Borgman, C. L. (2015). Big data, little data, no data: Scholarship in the networked world. The MIT press. - Kittler, F. A. (Ed.). (1980). Austreibung des Geistes aus den Geisteswissenschaften: Programme des Poststrukturalismus. Schöningh. - Latour, B. (1990). Drawing things together. In M. Lynch & S. Woolgar (Eds.), Representation in scientific practice. MIT Press. - Luhmann, N. (1992). Die Wissenschaft der Gesellschaft. Suhrkamp. - Martus, S., & Spoerhase, C. (2022). Geistesarbeit. Eine Praxeologie der Geisteswissenschaften. Suhrkamp. - Mol, A. (2002). The body multiple: Ontology in medical practice. Duke Univ. Pr. - Rheinberger, H.-J. (2021). Spalt und Fuge: Eine Phänomenologie des Experiments. Suhrkamp. - Star, S. L., & Griesemer, J. R. (1989). Institutional Ecology, Translations and Boundary Objects. Social Studies of Science, 19(3), 387–420.