The Adoption of STEM Praxis in the Digital Humanities

Submitted by Sarah Stanley on Fri, 07/15/2016 - 20:27

The following is a transcription of Matt Hunter's and my presentation from Keystone Digital Humanities, held at the University of Pittsburgh from June 22-24th.

Matt: Since the digital humanities are often said to emulate STEM-like practices such as PI-driven funding and lab-based collaborative models, we originally proposed this talk to discuss ethical and practical considerations of funding in DH, concerning mainly how we should counteract any influences funding directives might have on project goals. Then, that LA Review of Books article (“LARB” for convenience)  came out, and now it’s impossible to have this funding discussion without recognizing the fact that we’re all living in a post-LARB world and beholden to answering for our neoliberal sins. As a quick refresher, that article bemoaned the fact that DH and its practitioners are complicit in what the authors think of as the “neoliberal” turn in Academia.

Sarah: Neoliberalism is actually a very precise term (theoretically), yet the authors expand it a bit beyond its original definition. And indeed, accusations of neoliberalism have been levied against the university as a whole, not just our discipline. Chris Lorenz’s “If You’re So Smart, Why Are You Under Surveillance” (apologies for paywall) provides pointed criticism of a university in which there is “a constant decrease in the level and quality of employment [...], which comes down to a steady process of deprofessionalization and a reduction in the number and the quality of jobs; and (3) constantly rising prices for the consumers of services such as education.” So, as we can see, this criticism cannot be made only against the digital humanist; it is endemic to the university as a whole.

Maybe at this point, we should define neoliberalism. For our purposes, we’re adopting  the definition given by William Davies in his “Bibliographic review of neoliberalism:”

neoliberal policy targets institutions and activities which lie outside of the market, such as universities [...] so as to bring them inside the market, through acts of privatization; or to reinvent them in a ‘market-like’ way

So, compared to this definition, there are some valid (yet overstated) arguments in the LA Review of Books article, but we believe they should not be applied to digital humanities practitioners so much as to the management systems in which they are forced to operate.

In this spirit, we would like to complicate the LARB article by introducing what we’re referring to as the administration/practitioner divide. The administration being the system that polices lab-based management practices and focuses, among other things, on securing grants (and all the logistics/gatekeeping therein). The practitioners are the ones whose focus is on the production of scholarship, rather than the management of it.  (And we do realize that we too are being reductive in creating this arbitrary divide; you can obviously be both, but for the sake of argument, we’ve had to oversimplify a bit.)

Matt: The LARB article criticized DH for: “pushing the discipline toward post-interpretative, non-suspicious, technocratic, conservative, managerial, lab-based practice.”  From what we’ve seen, 4 of these criticisms have already been addressed in detail by scholars: accusations of post-interpretative, non-suspicious, technocratic, and conservative practice. The remaining two have been somewhat less addressed: criticisms of “managerial” and “lab-based practice” in DH, which were luckily the two things we were going to talk about when we proposed this talk anyway!

What we’d like to talk about is the complicated relationship between the radical labor practices  found in certain DH circles and how the administration has coopted them for the tactical gains of the true “neoliberal,” corporatized university. Matthew Kirschenbaum speaks to this in his article “DH as/is a tactical term” by saying that the very name of the discipline itself is utilized by managers for financial purposes since it “is a term possessed of enough currency and escape velocity to penetrate layers of administrative strata to get funds allocated, initiatives under way, and plans set in motion.” The managerial structure views the work that we strive to do as merely a cash cow, highlighting the distinction between those who are passionate about the work itself and the structures in place that are solely interested in the capital opportunities.

Sarah: John D. Martin and Carolyn Runyon argue in their article named “Digital Humanities, Digital Hegemony” that project-based work allows for cultural hegemony to be perpetuated through funding agendas. In their words, money is “controlled by the small groups that have control over the allotment thereof. Each project is seemingly disconnected from every other and no direct connection can be drawn between their features, except that they all focus on some broadly-construed, necessarily vague theme.”

This vagueness, specifically vagueness with regards to the rubric against which projects are being judged, allows for the administration to award money to projects without justifying the award’s merit within the context of humanistic and cultural criticism. So, for clarity, what I am arguing is that the standards against which the project is judged should consciously incorporate humanistic theories (e.g. postcolonial theory, feminist theory, critical race studies, etc). The obscurity we see surrounding grant money relates to Chris Lorenz’ argument that, “Information, especially financial information, is increasingly treated as the private property of management. Money is power, and so is information about money.” Vagueness in the terms of grant funding allows funders to be unclear about the ways in which funding is impacting real world labor practices and social issues. So while many practitioners may be pushing back against hegemonic structures through their projects, the managerial systems currently in place are not beholden to the criticisms of labor practices that DH as a discipline has long been discussing.

Matt: Right, and this isn’t all a bunch of theoretical nitpicking either; managerial practices are in reality preventing projects from getting off the ground and preventing people from doing research how they need, across all disciplines. For an example, just a few weeks ago I had to sit down with a STEM-discipline PI and help rebudget a project after he was forced to cut two postdoc appointments from years to months merely because of stifling administrative rules imposed on already insufficient budgets. Current administrative systems are directly affecting how scholarship gets done, and by whom.

Sarah: So perhaps, projects should be increasingly funded by localized bodies (such as institutions) rather than external bodies (such as Google), since increasingly baroque systems of management (in the external chain of command) prevent meaningful research from being conducted effectively. At the very least (if externally funded) grant-funded research should be conducted in a way that attends to local practice and acknowledges the impact of local relationships on the research product. So, for example, grant funding should acknowledge that writing “a programmer” (as a role) into a grant is often not just writing an arbitrary person whose job it is to know PHP; it is usually a person with his or her own research agenda. And often (when people are writing grants) people with complex skills are hired simply as something like “web developer,” and go on to contribute intellectually beyond that role. Those intellectual contributions become essential to the success of the project (even if it’s sometimes unacknowledged). And if that person is unable to participate during the term of the grant, it’s often more complicated to replace them than simply “hiring another web developer.”

So an example of this may be a grad student studying early modern print who is also competent in web development. She was hired to the role of “web developer” to represent prints on the internet, but her knowledge of 16th century print technologies helps her in her role. Say she even continues writing scholarship on early modern text technologies as they relate to web representations. If she gets a unicorn TT job and has to leave, it’s not going to be as simple to replace her as just writing in a new developer. Indeed, the PI may want to do something different with our “developer” role, based on the actual contributions the person in the developer role made. (and grants often don’t allow you to treat roles as fungible in that regard.)

Matt: The other problem with bureaucratic systems surrounding grants is that those systems take into account efficiency of funding and “relevancy,” (as those things are determined by a nebulous and undefined “public” consensus) rather than by the scholarly community. Now this all is not to say we don’t like publicly-funded programs! They have been able to produce great projects which furthered thoughtful and insightful humanities research and scholarship. We are just very suspicious of the fact that the scholarly community is increasingly judged by a conservative (and neoliberal) legislative branch which questions the validity and value of our work based on its “efficiency.” (And, if you wanna know how screwed the humanities are, take a look at Senator Jeff Flake’s recent oversight report, Twenty Questions, a “waste report” in which STEM researchers were criticized for their wasteful and frivolous research.) We should not be reliant upon bureaucratic systems that force us to constrain our research based on profitability or the whims of Congress.

So, to bring this back around a bit, when we’re looking at how we’re following in the footsteps of STEM (which is regularly grant funded), ought we to play into the hands of these managerial structures? Like I just said, these structures are the ones that have been judging value of STEM research based on the marketability of their findings (read: deliverables). Rita Raley cynically claims in her “Digital Humanities for the Next Five Minutes” that “in an institutional context in which a corporate administrative class is already mystified by humanities research that it cannot assess in terms of the amorphous metrics of ‘excellence’ and ‘innovation,’ one might say that the digital humanities is also particularly well positioned to exploit the expectation that we should be affectively awed by instrumentation.” And by instrumentation, we’re meant to read “shiny new tools,” which we are able to create with the help of these grants. I would challenge the sentiment that we should be attempting to exploit the system, since this labor of playing into neoliberal metrics detracts from the metrics that we in the humanistic community have already internally decided upon. Why are we still allowing ourselves to be judged based on quantitative benchmarks or cool factor rather than self-ascribed metrics like critical and intellectual contribution?

Sarah: One way in which we are buying into neoliberal metrics that’s really making things difficult is the way in which we are choosing to fund grad students (as I will elucidate). We are starting to emulate STEM practices such as relying on grant funds to matriculate our master’s and PhD students. In this new model we are starting to adopt, funding cycles are determining academic participation. This means that cohorts could be entirely dependent on grant funds. Is that what we want to copy? And we do need to remember that it’s not the grad students getting the grant, either, it’s the PI. Which makes grad funding dependent upon PIs getting funding. And this, I think, is what DH claims it’s not doing. Our collaboration (we say) is equitable and collaborative. So, how true is that? How collaborative are we really? In their surveys on student labor and training, Anderson et al. find:

student survey respondents worked an average of 11.5 hours a week on their digital projects, but only spent an average of three hours working with others (SS 9; 10). More surprisingly, 43% of the respondents — a near majority — reported working with others for only one hour a week or less (SS 9; 10). Additionally, when asked to rate the collaborative nature of their project, 67% of the students rated their work as minimally to moderately collaborative (SS 18), whereas in the equivalent faculty survey question 63% rated their projects as highly collaborative (FS 5).

The fact that students are often working alone (and often on data entry) shows that DH projects are structured as assembly-lines, where students are often treated as replaceable parts within the research machine. And actually, this is very similar to STEM, wherein one person works on research design, another works on the experiment, another on making the figures, etc. tee-bee-aitch collaboration for us looks like collaboration in STEM.

Instead of working in models where students are relegated to data entry and menial labour, the Digital humanities should strive to champion less top-down structures in its research praxis, so that we can combat the issues that have already been identified by STEM practitioners (we are aware that this is already happening in STEM and in some humanities labs, but it should be more universally adopted). (As this PhD comic about authorship shows...)

Matt: On the question of assembly-line-ism we talked about a second ago, I’d just like to bring up that piecemeal collaboration in STEM does work in situations where individual contribution is discrete but still valuable. If we adopt this practice of disparate collaborative tasks, we must likewise ensure that value is attributed to the individual’s intellectual contribution and not that position’s task. The mindset of valuing a task over intellectual labor demotes that collaborative role to a replaceable part of the economic machine. This further exacerbates the problem of treating research as merely a commodity assigned value for its profitability rather than its critical and intellectual contributions to the field.

Sarah: Assembly-line models of collaboration promote systems in which technical specialists and other contributors are only consulted at the specific points in the process (“hey, generate my schema,” or “hey, do a visualization on my data”) which is toxic for digital humanities projects (and their collaborators!). PI-centric lab models not only reinforce hierarchical intellectual strata, but also limit the ways collaborators can can participate in the project. So, for example, graduate students who are hired for data entry cannot be compensated (or sometimes even acknowledged) for contributions that aren’t related to data entry. They often (as we saw in the Student Labor piece) cannot and do not participate in activities which could give them credit and capital within the field (e.g. they do not tend to write papers). This means that contingent laborers are not always participating in activities that could allow for their growth, or to the growth of the field at large to which they would otherwise be encouraged to contribute. (this relates to the problem that always gets brought up about digital dissertations)

Matt: So we’ve talked a lot about all the things we as digital humanists can borrow in toto from STEM practice, as well as how there are a number of sticky issues that come along with wholesale adoption. Now we have to ask ourselves, as digital humanists, will we accept the bad for the good? What are the parts that we can adopt in good faith, to follow the spirit of STEM collaboration, while not necessarily the letter. That is to say, what parts of current models should we be careful of adopting, so as to not emulate praxis harmful to our discipline?

To put that yet another way: is funding (which currently seems to be necessary to navigate within the corporatized academy) something that we are prepared to sacrifice our humanistic values for, or is there too much unwanted baggage?

Sarah: The funding aspect is a very difficult one to approach, since it is so dependent upon “culture” writ large. While we want to move away from grant funding models, we understand that it is necessary in the current climate. However, we need to start discussing alternatives and pushing for them. The best way to do that is within our own research and scholarly communities, specifically in the way we address collaboration. We cannot emulate lab models in which certain collaborators are devalued and replaceable. By pushing back against this type of lab model, we can actually push up against restrictive funding practices. We need to stop treating researchers like replaceable parts in an assembly line and start structuring our projects around the idea that each individual has his or her own valid intellectual contribution. Only in this way can we actually start resisting funding agendas which value the status quo over critical engagement and tacitly perpetuate hegemonic control over scholarship.