Bridging the Gaps in and by Teaching

Letting Collaboration Be the Teacher

At the University of Stuttgart, the master’s program for digital humanities is centered around the department of digital humanities, which is part of the Institute for Literary Studies. Because Stuttgart also has a longstanding tradition in computational linguistics, textual analytics is a strong focus. Nonetheless, our master’s program is open to all kinds of humanities BA graduates. Students have backgrounds in linguistics, history, literature, art history, sociology, and communication; to cover a wider disciplinary spectrum that reflects these backgrounds, we add courses offered by other departments, mostly humanities and computer science, to our DH curriculum. However, for a variety of reasons, it is not always possible to integrate a spectrum as differentiated as would be desirable for our students. For example, sometimes courses require elaborate prior knowledge, and sometimes departments simply do not focus enough on digital methods to provide adequate courses. In chapter 24 in this volume, Heather Richards-Rissetto and Adrian Wisnicki describe similar problems and propose agile DH curricula suited to the needs of students, fostering independence and necessitating experimentation.

The DH program at the University of Stuttgart addresses the need for developing skills necessary for project-oriented interdisciplinary collaboration through project-oriented seminars, or project seminars, as they are called in the curriculum. Placed in the curriculum right after a first semester that is intended to give a broad, yet input-intensive overview of concepts and methods in the field, the project seminar releases that richness of information into research- or application-driven student projects to let students experience how to use this knowledge from start to finish. For the seminars, we split the cohort into small groups of three to four students, each of whom deals with a different topic. Topics, broad research questions, and pertaining context are provided by professors and research staff who act as teachers and facilitators for each group. Students rank the top projects on which they want to work and are then grouped according to their preferences but without overcrowding any specific group.

Digital humanities programs in Germany are focused on graduate education and on providing digital skills for undergraduates who bring backgrounds in the humanities. These programs often have different focal points due to the highly varied nature of local digital humanities scholarship and the individual traditions of humanities computing: some might emphasize the humanities and computer science skills necessary for building digital editions, some might focus more on natural language processing (as is the case in Stuttgart), and some place attention on reflections on digital practice. Most often, these programs are the result of institutional collaboration between the computer science and humanities departments, though sometimes courses are primarily offered through a dedicated department.

Within this framework, some of the programs in Germany offer courses that cover the organization of team-based DH projects. However, seminars explicitly aiming at the practical, team-based DH project are not the norm. If they are offered, this often happens in the form of seasonal electives, not as an integral part of the curriculum.

In our experience, depending on contacts with cultural institutions like archives and museums and on the interdepartmental collaboration at the university level, it is sometimes difficult to find a sufficient number of projects that also offer enough variety for students’ very specific backgrounds. For instance, as our questionnaire has shown (see Evaluation section to follow), students with backgrounds in disciplines like art history, which are not easily mapped to the most common DH tasks, sometimes do not feel as represented in the project offerings.

To cover a wider disciplinary spectrum than can be offered by the department itself, we invite colleagues from other faculties and external partners from local institutions in the cultural heritage sector to provide real-world research problems from their domain that are suited to digital approaches. In the past, this has led to seminars that dealt with a variety of topics like propaganda detection in newspaper articles, virtual collections in the Stuttgart Kunstmuseum, a network visualization of historical sources from the Regesta Imperii, or reflections on gender bias in data mining, to name but a few.

Conceptualization and project management are an explicit part of the learning experience. Although most courses do not teach any specific strategy for project management, workflows, or methods, groups are tasked with elements of self-organization and project management, which are communicated as part of the grading scheme at the start of the seminar. Self-organization and management also include the definition of goals and outcomes under the guidance of a supervisor, identification of the necessary means to achieve these goals, and regular assessments of (temporary) outcomes through group presentations and discussions in the small seminar group and in the larger group of all project seminars. This aspect of project management is part of the reflective project documentation—the written part of the graded semester work. The final event of this process is a poster slam and poster session, held at the welcome event for new first-year students.

Students must find answers to many questions: should we use project management software like Trello? Versioning or no versioning? How frequently should we meet? How should we coordinate between meetings? How should we document the process? Last but not least, the group defines and redefines its goals based on the time frame, necessary steps and skills, and the projected outcome. For example, in a project on network analysis and a dataset on the connections of narrative tropes on tvtropes.com, students quickly noticed they lacked some of the necessary expertise in network analysis. Thus, the group tried to find a more appropriate research question for their skill level and differentiated tasks into small work packages (e.g., data cleaning, report writing, and presentation). The resulting, independently defined analysis could only cover a small section of the dataset and was therefore limited in its informative value, but it yielded more confidence in the claims they could make. This does not mean that projects are pure exercises in project planning. Research is the main goal, with a realistic impression of research processes.

Evaluation

To relate past and present participants’ broader experiences to our own, we devised a questionnaire based on the standard learning outcome evaluation questionnaires at the University of Stuttgart. We used additional items with custom, more detailed questions. The questionnaire was then distributed to all previous and current project seminar participants. Thirty-five students and alumni filled out the questionnaire. Because regular evaluation is limited to seminars with a group size of five or more, our project seminars do not participate in the usual university evaluation cycles. Thus, our evaluation did not focus on the learning outcomes of single groups but on the overall experience of students with the seminar format. Some of the answers came as no surprise, but others certainly did.

Among the least surprising findings were the pairings of project categories and student backgrounds. The majority of respondents chose a project that had some aspects of their background field incorporated. Students with backgrounds in history chose projects that were more about building digital editions and historical databases. Students with a background in linguistics or literature chose projects that featured elements of computational text analysis, building classifiers, or reflecting text generators.

A minority of the students did not participate in a project close to their background. A rather unexpected finding was that although there were very mixed groups overall, in terms of disciplinary background and experience with digital methods, most respondents did not think this negatively impacted their experience; most students found that they were able to learn from the practices of their colleagues and vice versa. However, this was not true for all participants: for the groups with mixed backgrounds, roughly half (thirteen participants) indicated that they perceived their group to be moderately mixed with regard to background. The other half (fifteen participants) indicated that they had strongly mixed backgrounds. The participants with moderately mixed groups all seemed to have a favorable experience when it came to mapping their past education to that of their colleagues, but three participants from strongly mixed groups indicated a negative experience. Far from being a trend, this still highlights the possibility that some of the students might have had problems in environments where it proved difficult to establish common ground.

We suspect that many groups do have to initially “get an idea” of what a digital humanities project would actually entail. At this point in the curriculum, students only have a broad overview in mind and no idea of how different DH projects, or projects in computer science, statistics, or web development, for that matter, are actually organized. Often, project members do not even know what is possible with the algorithms and frameworks proposed in the initial project description.

Just like it is for many first-time DH researchers from the humanities or computer science, the frame of reference for these project seminar participants is often limited to the practices and expectations of their backgrounds. Within the constraints of DH-specific practices of publication rhythms, presentation styles, and usual work modes, how the research actually comes together and how it is produced, argued, and presented varies with the interdisciplinary multiplicity that often makes up a project. We try to emulate this with the project seminar that lets students work out the collaborative “how” for themselves in a safe environment by offering complex, research, or application-focused projects to be completed over a semester and finalized by reflective written documentation and poster presentations at the end.

Speaking from our own experience, many of the final project presentations are excellent, especially given the short time frame, and most are unique, out-of-the-box approaches to their respective subjects. To get to this point, the groups have to find a way to play to the strengths of their members, without working in different directions. Our questionnaire showed that, indeed, interdisciplinarity was not perceived to be an obstacle on the path to reaching this goal. But this experience was not the same for all participants. The focus on project organization was partly perceived as being too implicit. Self-organization and project management techniques need to be taken into account with discussions about collaboration and different work modes. A closer dialogue with students, based on our evaluation, might uncover more detailed areas of improvement. Though focused on various infrastructures in a program and the practices they can generate, Jacob Richter and Hannah Taylor’s contribution in chapter 28 in this volume also emphasizes the way that reflective “consideration of material, discursive, and rhetorical infrastructures,” and reflections of practices in a teaching and research environment in general, are vital for consolidating them. The contribution by Ermolaev, Munson, and Martin even explicitly relates a need for teaching critical project management skills, so as to enable future DH project managers to see possible reproductions of bias due to certain ways of project organization. Digital humanities curricula in general must remain agile to be able to react to the rapidly changing ecosystems in the DH landscape, as Richards-Rissetto and Wisnicki make clear.

As Ted Underwood notes in chapter 29 in this volume, having students experience other disciplinary cultures (in this case, information science) can be beneficial and prepare them for a multiplicity of research environments and requirements. Given the experience we had with our project seminar, we agree. The strengths perceived in DH, like interdisciplinary communication and, by consequence, reflection, do permeate to teaching. The trans-practical intersections between disciplines offer synergies that are important for the challenges of the often entirely new research questions that can be built in the digital humanities.

Bibliography

1. Pickering, Andrew. “From Science as Knowledge to Science as Practice.” In Science as Practice and Culture, edited by Andrew Pickering, 1–26. Chicago: University of Chicago Press, 1992.
2. Reed, Ashley. “Managing an Established Digital Humanities Project: Principles and Practices from the Twentieth Year of the William Blake Archive.” Digital Humanities Quarterly 8, no. 1 (2014). http://www.digitalhumanities.org/dhq/vol/8/1/000174/000174.html.