In The Big Humanities: Digital Humanities/Digital Laboratories, Richard J. Lane discusses a number of types of “humanistic labs, centres, projects, and products,” but he maintains a striking focus on the possibilities of the digital humanities (DH) lab. For Lane, the digital humanities lab provides an opportunity to bring a Big Science research model to the humanities, which includes a shift to “lab-based” knowledge production [Lane 2017]. Everything is larger: the research teams, the data sets, the duration of projects, the funding — evolution, so to speak, of what we might call a “lab ethos” [Lane 2017]. In addition, the construction of Big Humanities as a topic area correlates closely with a shift in philanthropy in higher education, which is “less concerned about creating in-perpetuity funds than solving a large, intractable problem” [Thorp 2010, 145]. DH funders such as the Mellon Foundation will fund projects for impact, which is viewed as highly interdisciplinary, with high levels of engagement, and market-consciousness [Thorp 2010]. The more scientific the model and technical the discourse, the more likely STEM funding may also be available [Thornham 2017]. For this reason, Big Humanities are in great demand, and labs compete for large-scale funding through the constructive and imagineering nature of their grants. Inevitably, the paradigm of funding shapes the kind of work that is done, and the tools and methods that are used.

Of course this is not the only value that institutions see in digital humanities labs. One need only inspect the technologies in a given DH lab to know what kinds of projects in which the lab is participating or able to support. A large, funded project will support the tools and methods needed to get the specific work of that project done. There may even be development of specific packages or software. By contrast, a lab that fits the computing support model will offer a broader range of technologies, tools, and personnel to fit a broad range of DH interests. An example of the former can be found in the humanities college-funded Digital Environmental Humanities Lab at California State University-Northridge, where students under the mentorship of one professor used XMIND software to visualize data related to the creation of the national park system in order to expose the exclusion of people of color in this narrative [Ruiz 2017]. An example of the computing support model may be found in the press release issued by Bowling Green State University to announce their Digital Humanities Lab:

“The goal of the Digital Humanities Lab is to encourage and support innovative, interdisciplinary research projects that make creative use of technology,” said Colleen Boff, associate dean for University Libraries. “We focus on supporting projects that are enhanced by visualization tools, multimedia applications, and manipulating data acquired through application programming interfaces (API).”  [Bowling Green State University 2015]

The press release then lists the training offerings for faculty and students related to the software that supports these kinds of projects. By comparison, a Big Humanities lab, much like a Principal Investigator's (PI's) scientific laboratory, may not necessarily have tools, methods, or training offerings for a wider academic audience at their institution. The DH lab, then, is fairly well-understood as a mechanism for advancing individual research goals, supporting faculty enrichment, striving for institutional alignment with scientific paradigms for enterprise-level research, as well as a hub for postdoctoral and postgraduate training — even in institutions where it is lacking.

However, what do we see when we de-emphasize the institutional and faculty aspirations so often associated with DH labs? The relationship between institutional role and research activity operates amidst other dynamics of research variation, and the DH lab is not uniform across it. As Mary Jo Deegan notes, the literature in DH in the United States (she mentions specifically Gold’s edited collections Debates in the Digital Humanities) tends to be more focused on theory and more focused on the Modern Languages Association and North America, whereas European DH tends to be more project and practice-focused [Deegan 2014]. The examples of DH labs offered by Claire Warwick tend to be more focused on the practices of large-scale projects [Warwick 2012]. Postgraduate and postdoctoral students tend to be situated as participants on these projects, who learn from PIs on the job.

In this article, we consider what invisible or less visible work becomes illuminated when practitioners understand the DH lab as a space for prioritizing graduate student needs. Graduate student perspectives may be sufficiently overlooked in the debates of “what,” “why,” and “who” is the digital humanities, despite the importance of graduate students to most DH teams reliant upon their labor. Thus, this article examines the ways that the DH lab can foster unique experiences for graduate students that fulfill what Alexander Reid postulates as the “central task” of DH — graduate education [Reid 2012]. More specifically, we argue that the DH lab as an institutional economic model does not necessarily promote a focus on graduate student professionalization, but that it can and should have a profound capacity to carve out space for deep thought in mixed-methods thinking, pushing back against competing discourses of neoliberal vocationalism, funding and labor precarity on one hand, and technological utopianism and tool fetishization on the other. Thus, as a critical and technological space committed to practices and capacities rather than a more exclusive emphasis on scholarly DH, the lab can train students in agile, contextual, and rhetorical mindsets with which to enter technologically-mediated workplaces and lives.

It is difficult to discuss the DH lab in a vacuum from the forces that create, shape, and reshape it over time. Here, we will review how the lab as a specific form of DH institution comes into being, with specific focus on funding models and projected inputs and outcomes from that funding. We will consider how the practice of DH in the United States and outside of it may affect the composition of labs and their graduate training. The graduate students who come to labor in the lab are a visible and direct result of the synergistic forces of funding and institutional shaping rather than consideration for the goals and objectives of graduate education. Based on these considerations, we will take a look at how various labs have approached graduate training and education as a part of that institutional model. We then argue that much of that graduate training and education is invisible because it is not tied to these funding models and resultant lab practices.

Training offerings in DH labs relate closely to the mission of the lab itself. As with the Bowling Green example, a computing support lab model tends to support general training offerings for the academic community related to “particular types of software or specialist tools” provided by the lab itself [Warwick 2012]. A Big Humanities lab, on the other hand, might offer graduate students the opportunity to work as RAs on a particular large scale sponsored research project, and be trained to support that particular project. These are the two polar opposites, and there can be a blending of these offerings, and other kinds of offerings as well.

The DH lab offers a space for experiential learning that challenges a more traditional classroom environment [Thorp 2010] in that it can be the professionals employed by the lab — often designated as support staff or unaffiliated with a traditional humanities discipline — that stand to offer the most expertise for experiential digital research training for graduate students [Nowviskie 2015]. Yet this same condition raises several questions regarding graduate student education: In the post-tenure university, Who does and should train graduate students, and to what ends? Who are graduate students to become (i.e., paid research assistants on Big Humanities projects, researchers on their own projects, future faculty in disciplines, “alt-ac” research professionals in higher education or industry)? At an institution like Arizona State University, for example, that has seen radical restructuring of its academic programs to as to create space for fostering interdisciplinary work, how do future faculty become effectively trained and usefully integrated into explicitly interdisciplinary spaces? Finally, why might a Big Humanities-focused DH lab participate in graduate student professionalization at all, if not incentivized under its funding model? These questions, answered or not, do much to influence how the DH lab views its mission in the university vis-à-vis its own disciplinary structure, and furthermore raise additional questions about the value of promoting such work and about the possibilities for the DH lab model. It is here where we begin our discussion of our own graduate training experiment that sought to make the work of graduate training visible in our DH lab.

We consider the affordances and constraints from such a model by offering a teaching case from our own Lab’s fall 2015 “Stories from Data” workshop in order to render visible a set of principles to guide professionalization of graduate students in the DH lab. The context for this case study is the Nexus Lab at Arizona State University. During our time employed at the Nexus Lab (as Director and Postdoctoral Fellow), it worked largely within the Big Humanities model: teams sought out large-scale and long-term funding; faculty partnerships for research spanned humanities and STEM disciplines; and upper administration aligned the lab with other units at the University invested in broad-reaching and big-impact work. However, part of the mission of the lab was to operate in a Big Humanities space while at the same time integrating technologies and lessons learned from that activity into graduate student training.

The Stories from Data workshop ran in the Fall Semester of 2015. The curriculum spanned fifteen weeks, beginning in the second week of classes. Hosted in the Lab, sessions took place from 1:00-3:00 p.m. on Friday afternoons. Attendance was free and open to all university staff, students, and faculty. We did not require consistent attendance in order to participate. We believe this contributed to enormous initial interest, from which emerged a steady cohort of staff, faculty, and students. The workshop began with seventy five participants, and after steady attrition, concluded with a group of twelve who remained consistent attendees and participants. Their backgrounds included rhetoric, sustainability, industrial engineering, computer science, and literature.

Stories from Data was an ongoing exercise in connecting participant expertise with the process and tools for decision making with data. In many ways, we sought to provide as much value as possible to participants by using the workshop as a way of codifying into the recognizable skillset of “visualization” the critical thinking, collaboration skills, and mixed methods analysis that unfolds perpetually in the context of interdisciplinary research. While the concept of “interdisciplinarity” can seem esoteric and elusive, the ideas of “visualization” and “data visualization” have a more immediate appeal outside academia and possess crosscutting benefits even within university research.

Our framing principle for the workshop was “People make decisions based on stories from data, not the data alone.” Our goals for the workshop engaged the cognitive, cultural, design, and narrative dimensions of visualization: 1) understand the tendencies and capabilities of users; 2) produce visualizations that help us think, not merely present information; and 3) learn how to tell a story from data that considers visual and non-visual narrative parts.

Generally each session fell into two parts, a presentation/discussion and a hands-on exercise. Each session lasted two to three hours. Consistently, we referred to the audience of visualizations as “users” rather than “viewers” to foreground the idea that visuals are rarely passively consumed.

The Stories from Data workshop provides a teaching case to make visible what the work of professionalizing graduate students might look like. This work temporarily suspends the goals of sponsored research and academic promotion. This work moves beyond the poles of teaching tools for machines and software purchased for the DH lab as a support center and teaching tasks for work on PI-driven Big Humanities projects. In fact, this professionalization work rejects easy assimilation into any lab model. It is not a lab during this kind of work, in the sense that specific instruments or a number of experts are there, all working around a common problem or issue. Yet, a DH lab is precisely the place where this professionalization work can take place, as a site of experiential, cross-disciplinary, cross-rank, academic-industry collaboration.

Instead, we argue for centering graduate students and their development as individuals, to prepare them for a wide variety of contingencies in their future career paths. These principles do not invoke a particular set of skills, as the tools and methods used in industry and in academia are just tools, not development of the individual who may use them. As Goldman Sachs banking director Matthew William Barrett said in an interview, “I used to joke that if you can find me someone who has a degree in figuring out patterns of imagery in Chaucer’s Canterbury Tales, I can teach him to break down a balance sheet in 30 minutes. What you want is a mind” [Pitts 2010]. The set of principles behind the Stories from Data workshop for graduate students do not derive from a particular digital skillset. “Tools” are not the same as initiation into practice, or experience with the principles of mixed methods work. This set of principles may be applied to a variety of substantive areas.

In the last several years, several essays have been written about the state of the DH initiative in the United States and its successes and challenges. In 2011, Alan Liu issued a report and a critique of DH. He asked, “Are the digital humanities ready to live up to their responsibility to represent the humanities and higher education as the latter negotiates a new relation to postindustrial society?” [Liu 2012, 11]. Liu’s critique stemmed from what he saw as a divide between those who came out of humanities computing, who “lacks almost all cultural-critical awareness” and those working from new media studies, who are “indiscriminately critical of society and global informational ‘empire’ without sufficient focus on the specifically institutional — in this case, higher education — issues at stake” [Liu 2012, 11]. Many other critiques of the Mellon Foundation-funded American DH initiative have been raised in recent academic trade publications, such as Brennan’s recent polemic “The Digital Humanities Bust” in The Chronicle of Higher Education [Brennan 2017]. These critiques range from upbraiding a system that would fund particular types of scholarship perceived as lacking the rigor of more conventional forms of critique, to the questioning of the value of methods in DH for textual analysis [Brennan 2017].

The workshop model and its set of principles for professionalization offer an opportunity to visualize a DH that does a different kind of work than what is often targeted in these criticisms. Much of the unrest and debate mentioned above focuses almost exclusively on past or current PI-driven, sponsored research projects in DH. What a workshop like Stories from Data makes visible is how DH might inform a digital future populated with professionals who do different kinds of work, in the academy, in industry, and in public life. The graduate students we trained experienced a curriculum that was skill-oriented, but anchored in a commitment to critical thinking and articulating their own expertise to the skills at hand. As graduate students assessing their professional trajectories and employment prospects, they experienced and interrogated the political, economic, and cultural realities of working with digital technologies, inside and outside of the humanities. They recognize the changing face of the neoliberalized university as well as industry hiring practices that largely reward experience with up-to-date technology tools devoid of the context of use. And, by working as a place for student development, the DH lab had given them opportunities to prepare for both.

Indeed, a key value of this kind of training is that it connects ways of thinking in addition to instilling a specific skill set. The Stories from Data curriculum has since been taken up in multiple venues on our campus across disciplines and participant groups. It formed the foundations of a subsequent workshop, “Decision Design for Sustainability Approaches,” produced in concert with the School of Sustainability at Arizona State University. Additionally, the materials and lessons from the workshop continue to inform and be refined by interdisciplinary, data-driven research and leadership training among humanities doctoral students, STEM students, active-duty military, and student-veterans at the Library’s Unit for Data Science (co-author Simeone serves as current Director).

Finally, and perhaps most importantly, we hope to mentor graduate students who can render visible to the public how the academy contributes to knowledge production vis-à-vis technology. This is not a new argument; in fact, many digital humanists on large-scale projects echo this call to improve the public’s understanding of and engagement in DH projects. [Oh 2013] [Flanders 2013]. Further, on the week of the submission of this article, data scientist and author Cathy O’Neil wrote an op-ed for The New York Times titled “The Ivory Tower Can’t Keep Ignoring Tech.” In this piece, she argues “We need academia to step up to fill in the gaps in our collective understanding about the new role of technology in shaping our lives...It’s absolutely within the abilities of academic research to study such examples and to push against the most obvious statistical, ethical or constitutional failures and dedicate serious intellectual energy to finding solutions” [O'Neil 2017]. While controversial in several respects (including her description of how data science is taught in higher education), O’Neil points to exactly the issue Stories from Data is designed to combat — our graduate students can identify and translate their role in exactly this work to multiple stakeholders after they leave the lab, and in a way that communicates their value as problem solvers in the issues of concern that O’Neil outlines here. In the case of the lab, then, the impact on the issues of “math destruction” and data and society is not measured in what is formally studied and published by faculty with the help of graduate students, but the subsequent conduct of participants in the world outside of academia or a home department. This helps DH labs to envision the training of graduate students toward careers in their own right, driven by their convictions and talents rather than labs focusing on what faculty teach or research [Anderson 2016]. As a way of engendering reflective technological practice, the workshop space is designed to avoid replication of any model, either industry-based or academic. No graduate student in Stories from Data is pursuing a PhD in data science. Their work with data — and extracting knowledge from data — is meant to enrich their understanding of storytelling and reasoning with data. Crucially, that understanding includes critique, production, and translation. While we hope that scholars based in universities continue to shed light into the need for more deliberation and oversight when it comes to the power that big data, artificial intelligence, and algorithms hold over the lives of individuals. But it has been our mission that the graduate students we train in the lab — who eventually become professionals and citizens in any number of spaces — meaningfully intervene in those same systems as they continue to challenge our ethics, values, and possible futures.