MariaGisina/istock/getty images plus
Everyone in academe, especially administrators, seems to love it when faculty members engage in interdisciplinary projects. We praise it at every turn. But that said, interdisciplinary collaborations are hard. And if you are embarking on one, we think it’s important to agree with your collaborators on some common issues up front in order to head off common problems.
Let’s begin with an example. Say you’re a linguist and you’ve found the oncologist of your dreams. You’re both excited to research the impact of throat cancer on speech patterns -- you’ve had coffee, have read each other’s papers and have even been awarded a small seed grant. Before you assume that you and your collaborator are on the same page, take a few minutes and discuss the following questions. Addressing these issues before things get real can save a lot of future heartache.
No. 1: What Does Success Look Like?
Different disciplines have remarkably different incentive structures and communicate research findings to their peers in different ways. Some privilege peer-reviewed manuscripts, while others may honor conferences, conference proceedings or even a device going to market.
Peer-reviewed manuscripts are usually longer than conference papers, and the same goes for their review time. They’re usually submitted on a rolling basis, while conferences occur on a schedule with fixed submission deadlines. Publications in the “wrong” kind of outlet may yield little to no recognition for principal investigators and students, and other people in their own disciplinary community may not recognize the results.
Why not just publish your findings in multiple outlets? Unfortunately, that’s usually considered unethical. Tenured faculty may shrug and go with a venue outside their home discipline, but graduate students and other early-career researchers often cannot afford to “waste” a publication in an outlet that others in their community don’t value.
Time scales for research outputs also vary wildly. A medical device may take many years to be approved for use, while some areas of computer science move so rapidly that even last year’s results may be outdated. Often, the best tool in an interdisciplinary application may be considered too well established to even be considered “research” in a rapidly developing field.
For example, computer vision offers enormous potential for analyzing images for wildlife research, but the analytical tools needed, while too complex for the average wildlife biologist, are often too simple to interest a researcher from computer vision. Projects falling into this “novelty gap” -- too new to entail off-the-shelf solutions for one discipline, but too old to be considered cutting-edge in the other -- can be nonstarters even if the underlying ideas are high impact.
So make sure to ask one another: What do you need for this to be a success? If this work is funded, what would realistic success look like to you, and what would it look like for the funder?
No. 2: How Will We Communicate and Delegate?
It’s practically a cliché to point out that different disciplines use different vocabularies, even when discussing similar types of problems. The resulting Tower of Academic Babel can drastically slow the pace of research. Even something as simple as defining a geographical location can become a point of contention. At the Deep Carbon Institute, experts were grappling with three very different methods of identifying global coordinates, and it took some time to realize that a tool was needed that could translate among them.
So make sure to ask one another: Whose vocabulary are we going to use? Do we need to design a translation system?
Each discipline also uses its own tools. Working with someone’s else’s software program of choice can slow you down. While graduate students may learn a suite of new skills quickly, their faculty elders aren’t always as fast. That can even make it difficult for them to participate in, or meaningfully oversee, a project. Even a simple divide between a group using licensed software and another group committed to using open-source tools could pose challenges. Disciplines also have different expectations for the release of raw data after a project is over, especially medical data with personally identifiable information.
So make sure to ask one another: What tools would allow us to work together most effectively? What are our plans for data dissemination?
Another key aspect of communication boils down to the basic organizational structure of a team: Who knows what and when? Large research groups may have a layer of middle management between faculty members and students that could include postdoctoral researchers, permanent research staff, technical or administrative staff, research programmers, or lab or field technicians. Faculty may have little involvement with the day-to-day operations of the research, as they focus instead on proposal development, reporting and administration. But faculty from fields with smaller research groups may expect to collaborate directly.
So make sure to ask one another: Who will be coming to the meetings? How often do meetings need to be held? Could most of the work be completed asynchronously with occasional sharing of results, or will this project require real-time, in-person collaboration among members of each team?
Even once the collaboration has hit its stride in terms of workflow and communication, the order of authors for the team’s final products can be a point of contention. In many STEM fields, the first or “lead” author will be in the first position, whereas the senior or supervising author will be in the last one. In other disciplines, the order of authors is in order of seniority from beginning to end, while others alphabetize authors and include a statement of effort. Numerous other unwritten expectations hold sway in different disciplines. Such varying norms can cause confusion and conflict.
So make sure to ask one another: What are the standards in your discipline for authorship? Who else might be involved with this project and how much engagement would entitle them to co-authorship on the resulting products?
No. 3: What if One of Us Has to Step Away From the Project?
By its very nature, interdisciplinary research usually means each person brings their own expertise to the party, so one of the sneakiest challenges -- one that becomes apparent only when a collaboration goes wrong in some way -- is a lack of redundancy in terms of the skills required. This makes interdisciplinary work much riskier than ordinary collaborations. Let’s return to our opening example of the oncologist and the linguist. If the oncologist fails to deliver on their component of the project, the linguist clearly lacks the skills to jump in to fill the gap -- and the reverse is obviously true, as well. Even worse, the collaborator left to pick up the pieces might lack the knowledge to articulate the specialized skills needed to recruit an appropriate replacement. If COVID-19 has taught us anything, it is to expect the unexpected.
So make sure to ask one another: If you were not able to complete this project for some reason, can you suggest someone else who could? What skills would be needed?
In summary, interdisciplinary research can provide a wealth of exciting opportunities, both personal and professional. But this excitement can often run headfirst into some very real difficulties that we rarely discuss openly in academe. Because traditional academic training is so focused on disciplinary expertise, it’s easy to overlook the on-the-ground details that result when disciplines combine.
These details aren’t as interesting as the thrilling combustion of ideas that gets these projects started, but an ad hoc approach to them can derail even the most exciting interdisciplinary effort. When we communicate with colleagues in other disciplines, we must make explicit what may be left implicit in our own disciplines. A frank discussion at the outset can ensure we’ve set the stage for later success.